tag:old.nabble.com,2006:forum-4361Nabble - MonetDB - Monet Database Management Syst2009-11-29T15:52:13ZMonetDB is a database management system developed from a main-memory perspective using a fully decomposed storage model, automatic index management, extensibility of data types and search accellerators, SQL- and XQuery- frontends. MonetDB - Monet Database Management Syst home is <a href="http://sourceforge.net/projects/monetdb/" target="_top" rel="nofollow">here</a>.tag:old.nabble.com,2006:post-26567726Announcement: New Nov2009 feature release of MonetDB suite2009-11-29T15:52:13Z2009-11-29T15:52:13ZStefan ManegoldDear all,
<br><br><br>The MonetDB team at CWI/MonetDB BV are pleased to announce the new
<br>"Nov2009" feature release of the MonetDB suite of programs.
<br>This feature release is backward compatible with the previous
<br>releases (Nov2008 through Aug2009-SP2).
<br><br>Caveat 1:
<br>In MonetDB/XQuery, we extended and improved the use of value indices on
<br>comparisions between a value and an XPath ending in text() or attribute
<br>access (e.g., doc(x)//x/y[z/text() = "foo"] and for $x in doc//x/y where
<br>$x/@z = "bar" return $x); such queries do not need scanning data, but start
<br>at the index and navigate backwards from there.
<br>This required a change in the physical database storage. While the Nov2009
<br>release of MonetDB/XQuery can still read older (pre-Nov2009) databases, it
<br>is recommended to (back-up and) re-shred all documents to benefit from the
<br>new indices and achieve optimal performance.
<br><br>Caveat 2:
<br>When upgrading from a release older than May2009 (i.e., Nov2008
<br>through Feb2009-SP2), you need to move your write-ahead log
<br>following the "recommended" instructions for
<br>SQL (<<a href="http://monetdb.cwi.nl/Development/Releases/May2009/#SQL" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/May2009/#SQL</a>>)
<br>and/or
<br>XQuery (<<a href="http://monetdb.cwi.nl/Development/Releases/May2009/#XQuery" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/May2009/#XQuery</a>>).
<br><br><br>More information on the new release is available at
<br><<a href="http://monetdb.cwi.nl/Development/Releases/Nov2009/" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/Nov2009/</a>>.
<br><br>The sources and binaries are available at
<br><<a href="http://monetdb.cwi.nl/Download/" target="_top" rel="nofollow">http://monetdb.cwi.nl/Download/</a>>.
<br><br><br>Enjoy!
<br><br>The MonetDB team at CWI/MonetDB BV.
<br><br>--
<br>| Dr. Stefan Manegold | mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26567726&i=0" target="_top" rel="nofollow">Stefan.Manegold@...</a> |
<br>| CWI, P.O.Box 94079 | <a href="http://www.cwi.nl/~manegold/" target="_top" rel="nofollow">http://www.cwi.nl/~manegold/</a> |
<br>| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
<br>| The Netherlands | Fax : +31 (20) 592-4312 |
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-developers mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26567726&i=1" target="_top" rel="nofollow">Monetdb-developers@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-developers" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-developers</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-developers-f4365.html" embed="fixTarget[4365]" target="_top" >monetdb-developers</a></p>tag:old.nabble.com,2006:post-26567730Announcement: New Nov2009 feature release of MonetDB suite2009-11-29T15:52:13Z2009-11-29T15:52:13ZStefan ManegoldDear all,
<br><br><br>The MonetDB team at CWI/MonetDB BV are pleased to announce the new
<br>"Nov2009" feature release of the MonetDB suite of programs.
<br>This feature release is backward compatible with the previous
<br>releases (Nov2008 through Aug2009-SP2).
<br><br>Caveat 1:
<br>In MonetDB/XQuery, we extended and improved the use of value indices on
<br>comparisions between a value and an XPath ending in text() or attribute
<br>access (e.g., doc(x)//x/y[z/text() = "foo"] and for $x in doc//x/y where
<br>$x/@z = "bar" return $x); such queries do not need scanning data, but start
<br>at the index and navigate backwards from there.
<br>This required a change in the physical database storage. While the Nov2009
<br>release of MonetDB/XQuery can still read older (pre-Nov2009) databases, it
<br>is recommended to (back-up and) re-shred all documents to benefit from the
<br>new indices and achieve optimal performance.
<br><br>Caveat 2:
<br>When upgrading from a release older than May2009 (i.e., Nov2008
<br>through Feb2009-SP2), you need to move your write-ahead log
<br>following the "recommended" instructions for
<br>SQL (<<a href="http://monetdb.cwi.nl/Development/Releases/May2009/#SQL" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/May2009/#SQL</a>>)
<br>and/or
<br>XQuery (<<a href="http://monetdb.cwi.nl/Development/Releases/May2009/#XQuery" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/May2009/#XQuery</a>>).
<br><br><br>More information on the new release is available at
<br><<a href="http://monetdb.cwi.nl/Development/Releases/Nov2009/" target="_top" rel="nofollow">http://monetdb.cwi.nl/Development/Releases/Nov2009/</a>>.
<br><br>The sources and binaries are available at
<br><<a href="http://monetdb.cwi.nl/Download/" target="_top" rel="nofollow">http://monetdb.cwi.nl/Download/</a>>.
<br><br><br>Enjoy!
<br><br>The MonetDB team at CWI/MonetDB BV.
<br><br>--
<br>| Dr. Stefan Manegold | mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26567730&i=0" target="_top" rel="nofollow">Stefan.Manegold@...</a> |
<br>| CWI, P.O.Box 94079 | <a href="http://www.cwi.nl/~manegold/" target="_top" rel="nofollow">http://www.cwi.nl/~manegold/</a> |
<br>| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
<br>| The Netherlands | Fax : +31 (20) 592-4312 |
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26567730&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26563233[ monetdb-Bugs-921173 ] SQL: extremely long and complex query causes SEGFAULT2009-11-29T08:01:05Z2009-11-29T08:01:05ZSourceForge.netBugs item #921173, was opened at 2004-03-22 17:04
<br>Message generated for change (Comment added) made by nielsnes
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=921173&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=921173&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: None
<br>Group: SQL CVS Head
<br>>Status: Closed
<br>Resolution: Fixed
<br>Priority: 6
<br>Private: No
<br>Submitted By: Sjoerd Mullender (sjoerd)
<br>Assigned to: Niels Nes (nielsnes)
<br>Summary: SQL: extremely long and complex query causes SEGFAULT
<br><br>Initial Comment:
<br>The attached file contains a query from the crash-me
<br>script that causes a SEGFAULT in the server.
<br><br>My guess is that there is a buffer overflow in
<br>sql_gencode. I'm looking into it.
<br><br>----------------------------------------------------------------------
<br><br>>Comment By: Niels Nes (nielsnes)
<br>Date: 2009-11-29 16:01
<br><br>Message:
<br>closed, we no longjmp out of the recursion.
<br><br>The overflow problem which we used to hit here (when we didn't crash)
<br>isn't fixed but is covered by the 'case overflow bug'.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-16 22:19
<br><br>Message:
<br>test has been disabled in Nov2009 branch, only.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-08-14 12:02
<br><br>Message:
<br>tagged subject
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-08-13 16:40
<br><br>Message:
<br>Should this test be disabled in the Aug2009 branch, again?
<br>cf.
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-02-15 20:08
<br><br>Message:
<br>test
<br>sql/src/test/BugDay_2005-10-06_2.9.3/Tests/huge_expression_and_column_name.SF-921173*
<br>has been disabled in the Feb2009 release branch as the bug won't be fixed,
<br>there.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Martin Kersten (mlkersten)
<br>Date: 2008-11-10 08:08
<br><br>Message:
<br>The overflow is caused by the SQL default to assume a tinyint type
<br>as argument. We cannot assume the compiler to be more clever.
<br>The result is an overflow which is correctly catched
<br>with a null. (provided tinyint is 0-255 ;))
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Niels Nes (nielsnes)
<br>Date: 2008-06-02 07:13
<br><br>Message:
<br>Logged In: YES
<br>user_id=43556
<br>Originator: NO
<br><br>The problem here is overflow detection. On gdk and m4/m5 level we need to
<br>introduce some overflow exceptions.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2008-06-01 19:44
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>Re-opened as the respective test fails again (still?).
<br><br>With M5 server (both default and "algebra" SQL compiler), only the output
<br>seems to differ (returns "NULL" instead of "2001"):
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests5103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests5103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTestsG103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTestsG103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.out.00.html</a><br><br>With M4 server, the test triggers a segfault (at least on some
<br>architectures):
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests4103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests4103/GNU.64.64.d-Fedora8/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2006-01-15 13:15
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br><br>Closed as the actual/original bug is fixed.
<br><br>The segfault on some platforms is file in this new report
<br>#1406591 "several tests cause segfault on the same subset
<br>of
<br>platforms"
<br><a href="https://sourceforge.net/tracker/index.php?func=detail&aid=1406591&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/index.php?func=detail&aid=1406591&group_id=56967&atid=482468</a><br><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2006-01-14 12:35
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br><br>re-opend as it segfauls on the same platforms as
<br>#1314982 "kill(0) causes SEGFAULT"
<br><a href="https://sourceforge.net/tracker/index.php?func=detail&aid=1314982&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/index.php?func=detail&aid=1314982&group_id=56967&atid=482468</a><br>and
<br>#1292727 "Mserver segfault because of 'col_name'"
<br><a href="https://sourceforge.net/tracker/index.php?func=detail&aid=1292727&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/index.php?func=detail&aid=1292727&group_id=56967&atid=482468</a><br><br>Fedora Core 4 (32-bit)
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-Fedora4/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-Fedora4/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html</a><br>Gentoo 1.6.13 (32-bit)
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-Gentoo1.6.13/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-Gentoo1.6.13/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html</a><br>SuSE 9.3 (32-bit)
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-SuSE9.3/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-SuSE9.3/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html</a><br>SuSE 9ES (32-bit)
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-SuSE9ES/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/sql/.mTests103/GNU.32.32.d-SuSE9ES/src_test_BugDay_2005-10-06_2.9.3/huge_expression_and_column_name.SF-921173.err.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Niels Nes (nielsnes)
<br>Date: 2005-10-23 11:57
<br><br>Message:
<br>Logged In: YES
<br>user_id=43556
<br><br>check for to big (recursion depth) queries is added.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Niels Nes (nielsnes)
<br>Date: 2005-10-22 22:11
<br><br>Message:
<br>Logged In: YES
<br>user_id=43556
<br><br>check for to big (recursion depth) queries is added.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2005-10-06 10:34
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br><br>BugDay_2005-10-06: CLAIMED BY stmane
<br><br>BugDay_2005-10-06: TEST ADDED / FAILURE
<br>test (excl. stable.out) added as
<br>sql/src/test/BugDay_2005-10-06_2.9.3/Tests/huge_expression_and_column_name.SF-921173.*
<br><br>bug re-opened, as the segfault re-occurs (at least with
<br>MonetDB 4.9.3 + SQL 2.9.3 compiled with gcc 4.0.1 on my
<br>64-bit Fedora Core sytem)
<br><br>It might very well be that we run out of stack space with a
<br>rather deep recursion...!
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Niels Nes (nielsnes)
<br>Date: 2004-03-22 19:09
<br><br>Message:
<br>Logged In: YES
<br>user_id=43556
<br><br>A to large column name was generated. Column names are now
<br>protected,
<br>but a full audit (buffer overflow) of sql_gencode is needed.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=921173&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=921173&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26563233&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26560029[ monetdb-Bugs-2902320 ] SQL: COPY multiple files2009-11-29T00:02:26Z2009-11-29T00:02:26ZSourceForge.netBugs item #2902320, was opened at 2009-11-23 08:01
<br>Message generated for change (Comment added) made by nielsnes
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2902320&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2902320&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: SQL/Core
<br>Group: SQL CVS Head
<br>>Status: Closed
<br>>Resolution: Fixed
<br>Priority: 5
<br>Private: No
<br>Submitted By: Martin Kersten (mlkersten)
<br>Assigned to: Niels Nes (nielsnes)
<br>Summary: SQL: COPY multiple files
<br><br>Initial Comment:
<br>create table tmp(i int);
<br>copy into tmp from 'file1','file2';
<br><br>only produces the MAL code for copying file1.
<br><br>----------------------------------------------------------------------
<br><br>>Comment By: Niels Nes (nielsnes)
<br>Date: 2009-11-29 08:02
<br><br>Message:
<br>added test copy_multiple_files.SF-2902320.sql
<br><br>Added the missing code for handling multiple files
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2902320&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2902320&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26560029&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26557995SABAOTHgetMyStatus Too many open files2009-11-28T15:27:08Z2009-11-28T15:27:08ZMatthew Jones-5Hey Folks, we are continuing to get deadlocks about once a day and I've noticed this in the logfile:<br><br>2009-11-28 14:31:05 ERR reporting_primary[31485]: !SABAOTHgetMyStatus: IOException:sabaoth.getStatus:Unable to open directory /opt/MonetDB5/dbfarm: Too many open files<br>
<br>Not sure how many open files it had when it went south, but immediately after a restart I saw this in files-nr:<br><br><font size="2"># cat /proc/sys/fs/file-nr<br>
1408 0 3164775<br><br>That third number being the 'files-max' on the system currently... I think that must be being dynamically set.<br><br>This is with the latest release August2009-SP2, the net effect is a deadlock on all connections to merovingian and a segfault of the mserver process.<br>
<br>Neils, this is what led up to that 'log' file being truncated in sql_logs a few days ago... this message occurred just before this happened. I've restarted my database and should see the problem reproduce itself again within 24 hours. I am recording the number of file descriptors in use every minute by mserver, merovingian, and overall for the system so I can know what the numbers were leading up to it.<br>
<br>Is this a file descriptor leak in Monet or is my files-max set too low (even though it appears to be dynamic)?<br></font><br clear="all">73,<br>Matthew W. Jones (KI4ZIB)<br><a href="http://matburt.net" target="_top" rel="nofollow">http://matburt.net</a><br>
<br />------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br />_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26557995&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26542888Re: Fatal error: aborted too deep recursion2009-11-27T05:49:15Z2009-11-27T05:49:15ZHarm van Beek(DT)Hi Peter,
<br><br>We indeed execute a large query (which is constructed by progra mcode,
<br>so hard to change).
<br>You can find the query at the bottom of my reply.
<br><br>I understand that you introduced the limits to not run out of stack
<br>space. However, since we are in full control of the system on which the
<br>query is executed, we might be able to adapt the limits to our
<br>environment. Can you give us directions on where to set this?
<br><br>Thanks again,
<br><br>Harm
<br><br>--
<br>query:
<br><br>element result {
<br> for $name in distinct-values( (
<br> let $var1 :=
<br> let $var2 :=
<br> let $var3 :=
<br> let $var4 :=
<br> let $var5 :=
<br> let $var6 :=
<br> let $var7 :=
<br> let $var8 :=
<br> let $var9 :=
<br> let $var10 :=
<br> let $var11 :=
<br> let $var12 := doc('document.xml')//properties/..
<br> return $var12[ (self::registry) ]
<br> return $var11/ancestor-or-self::file[1]/properties/..
<br> return $var10[ properties/(name)[ contains( lower-case(.),
<br>'system' ) ] ]
<br> return $var9/child::registry/properties/..
<br> return $var8/child::key/properties/..
<br> return $var7/child::key/properties/..
<br> return $var6/child::key/properties/..
<br> return $var5[ properties/(name)[ . = 'Enum' ] ]
<br> return $var4/descendant-or-self::key/properties/..
<br> return $var3[ properties/(name)[ contains( lower-case(.), 'usb' ) ]
<br>]
<br> return $var2/descendant-or-self::value/properties/..
<br> return $var1[ properties/(name)[ . = 'LocationInformation' ] ]
<br> )/properties/*/name()
<br> )
<br> return element name { $name }
<br>}
<br><br><br><div class='shrinkable-quote'><br>>-----Original Message-----
<br>>From: Peter Boncz [mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=0" target="_top" rel="nofollow">P.Boncz@...</a>]
<br>>Sent: Friday, November 27, 2009 10:26 AM
<br>>To: 'Communication channel for MonetDB users'
<br>>Subject: Re: [MonetDB-users] Fatal error: aborted too deep recursion
<br>>
<br>>Hi Harm,
<br>>
<br>>The 'too deep recursion' error is typically triggered by very
<br>>large queries that
<br>>send the pathfinder compiler into a very deep (and time
<br>>consuming) search. The
<br>>error is given to prevent a crash due to lack of stack space.
<br>>Recently, I do
<br>>recall that the limits were tightened as there are some
<br>>operating systems in our
<br>>nightly test farm that give their processes little stack
<br>>space, and could still
<br>>cause pathfinder to crash due to lack of stack space.
<br>>
<br>>So, it depends. The limit when to break off search is
<br>>arbitrary and based
<br>>currently on the lowest common denominator among the systems
<br>>we used. We could
<br>>potentially increase it (and possibly fix the settings on our
<br>>testing machines
<br>>to allow more stack space). We could even introduce a
<br>>user-controller limit.
<br>>
<br>>However, this phenomenon typically happens only on strange
<br>>queries (e.fg doing
<br>>tens or hundreds XPath steps after each other).
<br>>
<br>>Could you share what query produced such an error for you?
<br>>
<br>>Best,
<br>>
<br>>Peter Boncz
<br>>
<br>>
<br>>> -----Original Message-----
<br>>> From: Martin Kersten [mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=1" target="_top" rel="nofollow">Martin.Kersten@...</a>]
<br>>> Sent: Thursday, November 26, 2009 4:28 PM
<br>>> To: Communication channel for MonetDB users
<br>>> Subject: Re: [MonetDB-users] Fatal error: aborted too deep recursion
<br>>>
<br>>> Harm van Beek(DT) wrote:
<br>>> > During the development of our tool, using Xquery on
<br>>MonetDB, we ran onto
<br>>> > the following error:
<br>>> >
<br>>> > ERROR = !fatal error: aborted too deep recursion
<br>>> >
<br>>> >>From previous posts (I guess) I can conclude that this
<br>>relates to an
<br>>> > Xquery that simply is too big (too many operators) to
<br>>create a good plan
<br>>> > for it.
<br>>> All systems have limits that either cause an emergency stop.
<br>>In this case,
<br>>> the operating system has such a limit.
<br>>> >
<br>>> > I tried several MonetDB releases (May2009-SP2,
<br>>Aug2009-SP2, Nov2009-DB0)
<br>>> > on Red Hat Enterprise Linux Server release 5.4 (Tikanga),
<br>>all showing
<br>>> > the same result.
<br>>> >
<br>>> > Does anyone have any directions on how to solve this?
<br>>> Yes. Programming cleverness comes into play. Especially if
<br>>you use a tool
<br>>> to generate queries, then that tool should be a little more
<br>>conservative
<br>>> in the kind of queries generated.
<br>>>
<br>>> > Can we reconfigure the compiler to fix this (or recompile it with
<br>>> > options)?
<br>>> No. At some point the optimization space becomes just too big.
<br>>> People that can spare the time to find the optimal settings
<br>>for a larger
<br>>> application space are more then welcome to try it out on the
<br>>source code
<br>>> made available.
<br>>>
<br>>> regards, Martin
<br>>> >
<br>>> > Thanks in advance!
<br>>> >
<br>>> > Harm van Beek
<br>>> > Netherlands Forensic Institute
<br>>> >
<br>>> >
<br>>---------------------------------------------------------------
<br>>----------
<br>>> -----
<br>>> > Let Crystal Reports handle the reporting - Free Crystal
<br>>Reports 2008 30-
<br>>> Day
<br>>> > trial. Simplify your report design, integration and
<br>>deployment - and
<br>>> focus on
<br>>> > what you do best, core application coding. Discover what's new with
<br>>> > Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>>> > _______________________________________________
<br>>> > MonetDB-users mailing list
<br>>> > <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>>> > <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>>
<br>>>
<br>>>
<br>>---------------------------------------------------------------
<br>>------------
<br>>> ---
<br>>> Let Crystal Reports handle the reporting - Free Crystal
<br>>Reports 2008 30-Day
<br>>> trial. Simplify your report design, integration and
<br>>deployment - and focus
<br>>> on
<br>>> what you do best, core application coding. Discover what's new with
<br>>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>>> _______________________________________________
<br>>> MonetDB-users mailing list
<br>>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=3" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>>
<br>>> No virus found in this incoming message.
<br>>> Checked by AVG - www.avg.com
<br>>> Version: 8.5.426 / Virus Database: 270.14.83/2526 - Release
<br>>Date: 11/25/09
<br>>> 19:43:00
<br>>
<br>>
<br>>---------------------------------------------------------------
<br>>---------------
<br>>Let Crystal Reports handle the reporting - Free Crystal
<br>>Reports 2008 30-Day
<br>>trial. Simplify your report design, integration and deployment
<br>>- and focus on
<br>>what you do best, core application coding. Discover what's new with
<br>>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>>_______________________________________________
<br>>MonetDB-users mailing list
<br>><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=4" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>
</div><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26542888&i=5" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26541543[ monetdb-Bugs-2903072 ] M5/GDK: assertion decref in gdk_bbp.mx2009-11-27T04:02:46Z2009-11-27T04:02:46ZSourceForge.netBugs item #2903072, was opened at 2009-11-24 13:17
<br>Message generated for change (Comment added) made by mlkersten
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903072&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903072&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: Core
<br>Group: MonetDB5 CVS Head
<br>Status: Open
<br>Resolution: None
<br>Priority: 5
<br>Private: No
<br>Submitted By: Fabian (mr-meltdown)
<br>Assigned to: Martin Kersten (mlkersten)
<br>Summary: M5/GDK: assertion decref in gdk_bbp.mx
<br><br>Initial Comment:
<br>side effect of quicktest (running all tests in All file through mclient), I found this crash:
<br><br>stderr of test 'tst272` in directory 'src/mal` itself:
<br><br># mserver5 --dbfarm=/tmp/tmp.wSFC1a0iC6/server/mal/src_mal/dbfarm --dbname=mTests_src_mal --set mapi_port=54000 --set monet_daemon=yes
<br>#warning: please don't forget to set your vault key!
<br>#(see /export/scratch/stripe/fabian/monetdb/current/program-i86pc/etc/monetdb5.conf)
<br><br># mclient -i -lmal -p54000 < src/mal/Tests/tst272.mal
<br>MAPI = monetdb@localhost:54000
<br>ACTION= read_line
<br>QUERY = profiler.cleanup();
<br>ERROR = !Connection terminated
<br><br># additional server output (if any) follows
<br><br>Assertion failed: 0, file /export/scratch/mirror/fabian/monetdb/current/common/src/gdk/gdk_bbp.mx, line 1814, function decref
<br><br># EOT src/mal/Tests/tst272.mal
<br><br><br>stdout of test 'tst272` in directory 'src/mal` itself:
<br><br># mserver5 --dbfarm=/tmp/tmp.wSFC1a0iC6/server/mal/src_mal/dbfarm --dbname=mTests_src_mal --set mapi_port=54000 --set monet_daemon=yes
<br><br># mclient -i -lmal -p54000 < src/mal/Tests/tst272.mal
<br>#-----------------#
<br># h t # name
<br># int int # type
<br>#-----------------#
<br>[ 1, 15 ]
<br>[ 2, 4 ]
<br>[ 3, nil ]
<br>[ 4, 16 ]
<br>[ 5, 25 ]
<br>[ 6, 36 ]
<br>[ 7, 49 ]
<br>[ 8, 64 ]
<br>[ 9, nil ]
<br>[ nil, nil ]
<br>[ nil, 81 ]
<br>#---------------------------------#
<br># h t t t # name
<br># int int int int # type
<br>#---------------------------------#
<br>[ 1, 15, 15, 15 ]
<br>[ 2, 4, 4, 4 ]
<br>[ 3, nil, nil, nil ]
<br>[ 4, 16, 16, 16 ]
<br>[ 5, 25, 25, 25 ]
<br>[ 6, 36, 36, 36 ]
<br>[ 7, 49, 49, 49 ]
<br>[ 8, 64, 64, 64 ]
<br>[ 9, nil, nil, nil ]
<br>#~BeginVariableOutput~#
<br>#~EndVariableOutput~#
<br><br># additional server output (if any) follows
<br># MonetDB server v5.17.0, based on kernel v1.35.0
<br># Serving database 'mTests_src_mal', using 1 thread
<br># Compiled for x86_64-pc-solaris2.11/64bit with 64bit OIDs dynamically linked
<br># Detected 2.000 GiB (2147020800 bytes) main-memory.
<br># Copyright (c) 1993-July 2008 CWI.
<br># Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information
<br># Listening for connection requests on mapi:monetdb://127.0.0.1:54000/
<br><br># EOT src/mal/Tests/tst272.mal
<br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Martin Kersten (mlkersten)
<br>Date: 2009-11-27 13:02
<br><br>Message:
<br>Patch added to mal_profiler. No additional test needed.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903072&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903072&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26541543&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26540781[ monetdb-Bugs-2900365 ] SQL: concurrent tracelog()2009-11-27T02:52:45Z2009-11-27T02:52:45ZSourceForge.netBugs item #2900365, was opened at 2009-11-19 11:46
<br>Message generated for change (Comment added) made by vzzzbx
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: SQL/Core
<br>Group: SQL "stable"
<br>Status: Open
<br>Resolution: None
<br>Priority: 5
<br>Private: No
<br>Submitted By: Wouter Alink (vzzzbx)
<br>Assigned to: Martin Kersten (mlkersten)
<br>Summary: SQL: concurrent tracelog()
<br><br>Initial Comment:
<br>When running 2 mclient instances simultaneously, the 1st mclient can run tracelog(), the second cannot (the error message is "SELECT: '(null)' does not return a table").
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>0 tuples
<br>sql>
<br><br>On another prompt:
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>SELECT: '(null)' does not return a table
<br>sql>
<br><br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-27 11:52
<br><br>Message:
<br>oops, wrong bug, ignore previous message please.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-27 11:51
<br><br>Message:
<br>I indeed reported it, without declaring the tracelog() function (as it
<br>isn't documented).
<br><br>would there be any objections to defining the tracelog() function by
<br>default? (does it have to be defined for each database, or can it be done
<br>once globally?)
<br>I can imagine that this function is not needed for many users, so perhaps
<br>the behaviour should stay as is.
<br><br>If the documentation clearly mentions that the function should first be
<br>declared before it can be used, i'm fine with closing this bug.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Martin Kersten (mlkersten)
<br>Date: 2009-11-27 11:42
<br><br>Message:
<br>This report has not been confirmed for initial state. Having multiple
<br>clients over the same database scheme and announced the definition of the
<br>tracelog function once seems to work as expected.
<br><br>Where there more processes running, e.g. updating the performance tables?
<br><br>Tracing the kernel is a global activity. For interpretation of timings
<br>this is a prerequisite.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-19 16:42
<br><br>Message:
<br>when running trace using JDBC, multiple connections to the same database do
<br>seem to be able to query tracelog(). (if no mclient is running)
<br>however, they are not isolated from each other; work done in one
<br>transaction, can be found in a call to tracelog() in another transaction.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540781&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26540766[ monetdb-Bugs-2900365 ] SQL: concurrent tracelog()2009-11-27T02:51:35Z2009-11-27T02:51:35ZSourceForge.netBugs item #2900365, was opened at 2009-11-19 11:46
<br>Message generated for change (Comment added) made by vzzzbx
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: SQL/Core
<br>Group: SQL "stable"
<br>Status: Open
<br>Resolution: None
<br>Priority: 5
<br>Private: No
<br>Submitted By: Wouter Alink (vzzzbx)
<br>Assigned to: Martin Kersten (mlkersten)
<br>Summary: SQL: concurrent tracelog()
<br><br>Initial Comment:
<br>When running 2 mclient instances simultaneously, the 1st mclient can run tracelog(), the second cannot (the error message is "SELECT: '(null)' does not return a table").
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>0 tuples
<br>sql>
<br><br>On another prompt:
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>SELECT: '(null)' does not return a table
<br>sql>
<br><br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-27 11:51
<br><br>Message:
<br>I indeed reported it, without declaring the tracelog() function (as it
<br>isn't documented).
<br><br>would there be any objections to defining the tracelog() function by
<br>default? (does it have to be defined for each database, or can it be done
<br>once globally?)
<br>I can imagine that this function is not needed for many users, so perhaps
<br>the behaviour should stay as is.
<br><br>If the documentation clearly mentions that the function should first be
<br>declared before it can be used, i'm fine with closing this bug.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Martin Kersten (mlkersten)
<br>Date: 2009-11-27 11:42
<br><br>Message:
<br>This report has not been confirmed for initial state. Having multiple
<br>clients over the same database scheme and announced the definition of the
<br>tracelog function once seems to work as expected.
<br><br>Where there more processes running, e.g. updating the performance tables?
<br><br>Tracing the kernel is a global activity. For interpretation of timings
<br>this is a prerequisite.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-19 16:42
<br><br>Message:
<br>when running trace using JDBC, multiple connections to the same database do
<br>seem to be able to query tracelog(). (if no mclient is running)
<br>however, they are not isolated from each other; work done in one
<br>transaction, can be found in a call to tracelog() in another transaction.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540766&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26540650[ monetdb-Bugs-2900365 ] SQL: concurrent tracelog()2009-11-27T02:42:54Z2009-11-27T02:42:54ZSourceForge.netBugs item #2900365, was opened at 2009-11-19 11:46
<br>Message generated for change (Comment added) made by mlkersten
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: SQL/Core
<br>Group: SQL "stable"
<br>Status: Open
<br>Resolution: None
<br>Priority: 5
<br>Private: No
<br>Submitted By: Wouter Alink (vzzzbx)
<br>Assigned to: Martin Kersten (mlkersten)
<br>Summary: SQL: concurrent tracelog()
<br><br>Initial Comment:
<br>When running 2 mclient instances simultaneously, the 1st mclient can run tracelog(), the second cannot (the error message is "SELECT: '(null)' does not return a table").
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>0 tuples
<br>sql>
<br><br>On another prompt:
<br><br>$ mclient -lsql -p51234 -daap
<br>Welcome to mclient, the MonetDB/SQL interactive terminal
<br>Database: MonetDB v5.16.0, 'aap'
<br>Type \q to quit, \? for a list of available commands
<br>auto commit mode: on
<br>sql>SELECT * FROM tracelog();
<br>SELECT: '(null)' does not return a table
<br>sql>
<br><br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Martin Kersten (mlkersten)
<br>Date: 2009-11-27 11:42
<br><br>Message:
<br>This report has not been confirmed for initial state. Having multiple
<br>clients over the same database scheme and announced the definition of the
<br>tracelog function once seems to work as expected.
<br><br>Where there more processes running, e.g. updating the performance tables?
<br><br>Tracing the kernel is a global activity. For interpretation of timings
<br>this is a prerequisite.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Wouter Alink (vzzzbx)
<br>Date: 2009-11-19 16:42
<br><br>Message:
<br>when running trace using JDBC, multiple connections to the same database do
<br>seem to be able to query tracelog(). (if no mclient is running)
<br>however, they are not isolated from each other; work done in one
<br>transaction, can be found in a call to tracelog() in another transaction.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2900365&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540650&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26540472Re: other than bug things...2009-11-27T02:25:36Z2009-11-27T02:25:36ZMartin KerstenHello Wouter,
<br><br>Thank you for this positive message. It made my day :))
<br><br>We would like to learn more about applications where
<br>MonetDB is working as planned. Giving the steady
<br>increasing monthly downloads there must be more of
<br>such cases in the field.
<br><br>regards, Martin
<br><br>Wouter Alink wrote:
<div class='shrinkable-quote'><br>> Hello Developers,
<br>>
<br>> Usually I try to report bugs (and try to find the worst of the
<br>> system), but just this morning I noticed my mserver5 instance
<br>> (Aug2009), which I have been intensively querying over the last days
<br>> with ten-thousands of reasonably complex queries and sometimes with
<br>> more than 20 queries in parallel. It just passed the 4800 minutes of
<br>> actual cpu time (= 80 hours of hard work) and still going strong. I
<br>> thought it was worth mentioning!
<br>>
<br>> Cheers,
<br>> Wouter
<br>>
<br>> p.s. roberto, actually it's your mserver instance...
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> Monetdb-developers mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540472&i=0" target="_top" rel="nofollow">Monetdb-developers@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-developers" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-developers</a></div><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-developers mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540472&i=1" target="_top" rel="nofollow">Monetdb-developers@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-developers" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-developers</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-developers-f4365.html" embed="fixTarget[4365]" target="_top" >monetdb-developers</a></p>tag:old.nabble.com,2006:post-26540274other than bug things...2009-11-27T02:09:51Z2009-11-27T02:09:51ZWouter Alink-3Hello Developers,
<br><br>Usually I try to report bugs (and try to find the worst of the
<br>system), but just this morning I noticed my mserver5 instance
<br>(Aug2009), which I have been intensively querying over the last days
<br>with ten-thousands of reasonably complex queries and sometimes with
<br>more than 20 queries in parallel. It just passed the 4800 minutes of
<br>actual cpu time (= 80 hours of hard work) and still going strong. I
<br>thought it was worth mentioning!
<br><br>Cheers,
<br>Wouter
<br><br>p.s. roberto, actually it's your mserver instance...
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-developers mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26540274&i=0" target="_top" rel="nofollow">Monetdb-developers@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-developers" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-developers</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-developers-f4365.html" embed="fixTarget[4365]" target="_top" >monetdb-developers</a></p>tag:old.nabble.com,2006:post-26539859Re: Fatal error: aborted too deep recursion2009-11-27T01:25:59Z2009-11-27T01:25:59ZPeter Boncz-2Hi Harm,
<br><br>The 'too deep recursion' error is typically triggered by very large queries that
<br>send the pathfinder compiler into a very deep (and time consuming) search. The
<br>error is given to prevent a crash due to lack of stack space. Recently, I do
<br>recall that the limits were tightened as there are some operating systems in our
<br>nightly test farm that give their processes little stack space, and could still
<br>cause pathfinder to crash due to lack of stack space.
<br><br>So, it depends. The limit when to break off search is arbitrary and based
<br>currently on the lowest common denominator among the systems we used. We could
<br>potentially increase it (and possibly fix the settings on our testing machines
<br>to allow more stack space). We could even introduce a user-controller limit.
<br><br>However, this phenomenon typically happens only on strange queries (e.fg doing
<br>tens or hundreds XPath steps after each other).
<br><br>Could you share what query produced such an error for you?
<br><br>Best,
<br><br>Peter Boncz
<br><br><div class='shrinkable-quote'><br>> -----Original Message-----
<br>> From: Martin Kersten [mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26539859&i=0" target="_top" rel="nofollow">Martin.Kersten@...</a>]
<br>> Sent: Thursday, November 26, 2009 4:28 PM
<br>> To: Communication channel for MonetDB users
<br>> Subject: Re: [MonetDB-users] Fatal error: aborted too deep recursion
<br>>
<br>> Harm van Beek(DT) wrote:
<br>> > During the development of our tool, using Xquery on MonetDB, we ran onto
<br>> > the following error:
<br>> >
<br>> > ERROR = !fatal error: aborted too deep recursion
<br>> >
<br>> >>From previous posts (I guess) I can conclude that this relates to an
<br>> > Xquery that simply is too big (too many operators) to create a good plan
<br>> > for it.
<br>> All systems have limits that either cause an emergency stop. In this case,
<br>> the operating system has such a limit.
<br>> >
<br>> > I tried several MonetDB releases (May2009-SP2, Aug2009-SP2, Nov2009-DB0)
<br>> > on Red Hat Enterprise Linux Server release 5.4 (Tikanga), all showing
<br>> > the same result.
<br>> >
<br>> > Does anyone have any directions on how to solve this?
<br>> Yes. Programming cleverness comes into play. Especially if you use a tool
<br>> to generate queries, then that tool should be a little more conservative
<br>> in the kind of queries generated.
<br>>
<br>> > Can we reconfigure the compiler to fix this (or recompile it with
<br>> > options)?
<br>> No. At some point the optimization space becomes just too big.
<br>> People that can spare the time to find the optimal settings for a larger
<br>> application space are more then welcome to try it out on the source code
<br>> made available.
<br>>
<br>> regards, Martin
<br>> >
<br>> > Thanks in advance!
<br>> >
<br>> > Harm van Beek
<br>> > Netherlands Forensic Institute
<br>> >
<br>> > -------------------------------------------------------------------------
<br>> -----
<br>> > Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-
<br>> Day
<br>> > trial. Simplify your report design, integration and deployment - and
<br>> focus on
<br>> > what you do best, core application coding. Discover what's new with
<br>> > Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> > _______________________________________________
<br>> > MonetDB-users mailing list
<br>> > <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26539859&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> > <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>
<br>>
<br>> ---------------------------------------------------------------------------
<br>> ---
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus
<br>> on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26539859&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>
<br>> No virus found in this incoming message.
<br>> Checked by AVG - www.avg.com
<br>> Version: 8.5.426 / Virus Database: 270.14.83/2526 - Release Date: 11/25/09
<br>> 19:43:00
</div><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26539859&i=3" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26539042[ monetdb-Bugs-1681244 ] Mapi ports and patfhinder/pftijah2009-11-27T00:07:15Z2009-11-27T00:07:15ZSourceForge.netBugs item #1681244, was opened at 2007-03-15 09:25
<br>Message generated for change (Settings changed) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1681244&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1681244&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: Mapi
<br>Group: MonetDB4 CVS Head
<br>>Status: Closed
<br>>Resolution: Fixed
<br>Priority: 5
<br>Private: No
<br>Submitted By: Jan Rittinger (tsheyar)
<br>Assigned to: Stefan Manegold (stmane)
<br>Summary: Mapi ports and patfhinder/pftijah
<br><br>Initial Comment:
<br>When I have a Mserver (with module pathfinder loaded) running somewhere and want to start a complete unrelated Mserver (with pathfinder (and pftijah) I get the following Mserver crashes:
<br><br>Mserver --dbinit="module(pathfinder);"
<br># Monet Database Server V4.17.1
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>!ERROR: MAPIlisten:binding to stream socket (50000) failed
<br>!OS: Address already in use
<br>MonetDB>01/Jan/1970 01:00:00 shttpd_open_port: bind(50001): Address already in use
<br><br>Mserver --dbinit="module(pathfinder); module(pftijah);"
<br># Monet Database Server V4.17.1
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>!ERROR: MAPIlisten:binding to stream socket (50000) failed
<br>!OS: Address already in use
<br>!WARNING: TBL_M_directloaded: pftijah relative refcounts are wrong (2!=3)!
<br>!WARNING: TBL_M_directloaded: pftijah relative refcounts are wrong (2!=3)!
<br>Mserver: /local_tmp/MonetDB4/src/monet/monet_tbl.mx:1740: TBL_M_decref: Assertion `p' failed.
<br>Aborted
<br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 09:07
<br><br>Message:
<br>Bug report and test are covered by
<br>ID: 1809586 "PF: busy port crashes Mserver"
<br><a href="https://sourceforge.net/tracker/?func=detail&aid=1809586&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=1809586&group_id=56967&atid=482468</a><br>"
<br>We accept the fact that starting a MonetDB/XQury server (`Mserver
<br>--dbinit="module(pathfinder);"`)
<br>quits instantly (with a proper error message)
<br>in case either the MAPIPORT or the XQUERYPORT are not available
<br>(the server does not make much sense without access to these ports).
<br>The default ports can easily be overwritten on the server commandline:
<br>--set mapi_port=<portno> --set xrpc_port=<portno>
<br>"
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-08-02 22:57
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>...as always a test in CVS could (have) answer(ed) the question...
<br><br>If nobody is faster, I'll try to add one tomorrow (CEST).
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Peter Boncz (boncz)
<br>Date: 2007-08-02 22:45
<br><br>Message:
<br>Logged In: YES
<br>user_id=591107
<br>Originator: NO
<br><br>is this bug still reproducable, given the changes we made in tijah module
<br>loading before the .18 release?
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Sjoerd Mullender (sjoerd)
<br>Date: 2007-03-15 10:54
<br><br>Message:
<br>Logged In: YES
<br>user_id=43607
<br>Originator: NO
<br><br>I agree with Jan, there is a genuine bug here.
<br><br>In one of my tries I even got a segfault.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Jan Rittinger (tsheyar)
<br>Date: 2007-03-15 10:32
<br><br>Message:
<br>Logged In: YES
<br>user_id=993208
<br>Originator: YES
<br><br>Stefan,
<br><br>with this feature I expect only an error message (like the one in the
<br>first line) and the side effect that the module is not loaded. However a
<br>Mserver crash (with a misleading error information) does not seem
<br>reasonable to me:
<br><br>Mserver
<br># Monet Database Server V4.17.1
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>MonetDB>module(pathfinder);
<br>!ERROR: MAPIlisten:binding to stream socket (50000) failed
<br>!OS: Address already in use
<br>!WARNING: TBL_M_directloaded: _pftijah relative refcounts are wrong
<br>(1!=2)!
<br>01/Jan/1970 01:00:00 shttpd_open_port: bind(50001): Address already in
<br>use
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-03-15 10:24
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>Jan
<br><br>this is not a bug but a feature:
<br><br>Loading module pathfinder (just like loading module sql) implicitly
<br>triggers the start of the respective mapi listener so that clients can
<br>connect. The default port (mapi_port) used for these listeners is 50000.
<br>Obviously, starting two programs that (try to) claim the same port/socket
<br>does/can not work.
<br>If you want/need to run two (or more) Mservers with MAPI enabled (via
<br>module(pathfinder);, module(sql);, or module(mapi);mil_start();) on the
<br>same machine, you need to make sure that they use different mapi_port's by
<br>starting (at least) all but one with specifying a different mapi_port,
<br>e.g., via `Mserver --set mapi_port=<mapi_port_no>`; in case of pathfinder
<br>you might also want/need to set a different xrpc_port, i.e. `Mserver --set
<br>mapi_port=<mapi_port_no> --set xprc_port=<xrpc_port_no>`.
<br><br>Stefan
<br><br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1681244&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1681244&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26539042&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26538990[ monetdb-Bugs-1893448 ] PF: xrpcConvert2 fails on Windows2009-11-27T00:00:08Z2009-11-27T00:00:08ZSourceForge.netBugs item #1893448, was opened at 2008-02-14 11:05
<br>Message generated for change (Comment added) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1893448&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1893448&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: PF/runtime
<br>Group: Pathfinder "stable"
<br>>Status: Closed
<br>>Resolution: Fixed
<br>Priority: 3
<br>Private: No
<br>Submitted By: Stefan Manegold (stmane)
<br>Assigned to: Ying Zhang (yingying)
<br>Summary: PF: xrpcConvert2 fails on Windows
<br><br>Initial Comment:
<br><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert.err.00.html</a><br>========
<br>MAPI = monetdb@TURELUUR:32113
<br>QUERY = import module namespace xrpc = "xrpc-fun" at
<br>ERROR = !ERROR: Document is empty
<br> !ERROR: during RPC call to "TURELUUR:49525"
<br> !ERROR: shredder_parse: XML input not well-formed.
<br> !ERROR: response2bat: invalid XRPC response received
<br> !ERROR: CMDhttp_post: operation failed.
<br>MAPI = monetdb@TURELUUR:32113
<br>QUERY = import module namespace xrpc = "xrpc-fun" at
<br>ERROR = !ERROR: doLoopLiftedRPC: execution failed at all destinations
<br>========
<br><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.32.32.d-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html</a><br>========
<br>MAPI = monetdb@TURELUUR:32113
<br>QUERY = import module namespace xrpc = "bar" at
<br>ERROR = !parse error: syntax error, unexpected IntegerLiteral, expecting ) on line 7, column 36 (next token is `4')
<br> !parse error: XQuery parsing failed
<br> !parse error: syntax error, unexpected IntegerLiteral, expecting ) on line 7, column 36 (next token is `4')
<br> !parse error: error parsing module from C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.27057\\pathfinder\\NT\\mTests\\tests\\XRpc/xrpcConvert2.xrpcfun2.xq
<br>========
<br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 09:00
<br><br>Message:
<br>Test appaers to work fine also on Windows (at least) since 2009.11.16;
<br>cf.,
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.64.d.1-Windows6.0/tests_XRpc/xrpcConvert2.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.64.d.1-Windows6.0/tests_XRpc/xrpcConvert2.err.00.html</a><br><br>Closing.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-07 14:29
<br><br>Message:
<br>See also
<br>ID: 2893815 "PF: xrpcConvert tests segfault since 2009/10/08"
<br><a href="https://sourceforge.net/tracker/?func=detail&aid=2893815&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=2893815&group_id=56967&atid=482468</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-02-14 18:31
<br><br>Message:
<br>(only) the xrpcConvert2 test still fails on Windows:
<br><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.32.32.d.1-Windows5.1/tests_XRpc/xrpcConvert2.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.32.d.1-Windows5.2/tests_XRpc/xrpcConvert2.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTestsg103/Int.64.32.d.1-Windows5.2/tests_XRpc/xrpcConvert2.err.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Ying Zhang (yingying)
<br>Date: 2008-09-05 18:35
<br><br>Message:
<br>Logged In: YES
<br>user_id=341633
<br>Originator: NO
<br><br>Some status info about this bug:
<br><br>The path matching problem has been fixed in milprint_summer.c, version
<br>1.419.2.5, thanks to Sjoerd's debugger on windows.
<br><br>Sjoerd and I had a look at the problem with xrpcConvert2, and
<br>unfortunetely, we could not reproduce the error with his version compiled
<br>with debugging on.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2008-05-29 15:43
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: YES
<br><br>With the code base for the upcoming Jun2008 release (XQuery_0-24 branch),
<br>the error for the first xrpcConvert test no seems to indicate some path
<br>matching / '\' quoting problem when checking for "allowed" ("trusted")
<br>module paths/locations:
<br><br>========
<br>MAPI = monetdb@TURELUUR:37285
<br>QUERY = import module namespace xf = "xrpc-fun" at
<br>"C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.20696\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
<br>ERROR = !ERROR: during RPC call to "TURELUUR:45558"
<br> !ERROR: response2bat: invalid response from TURELUUR45558
<br> !ERROR: !ERROR: during RPC call to "TURELUUR:45558
<br> !ERROR: !ERROR: HTTP Error Code : 403 Forbidde
<br> !ERROR: !ERROR: SOAP Fault Code : env:Sende
<br> !ERROR: !ERROR: SOAP Fault Reason
<br> !ERROR: !ERROR: Called module not allowed
<br> !ERROR: !ERROR: 'C:cygwinvar mpmonet.Int.32.32.d.20696pathfinder
<br>estsXRpcTests/xrpcfun.xq' not i
<br> !ERROR: !ERROR: '<a href="http://monetdb.cwi.nl'" target="_top" rel="nofollow">http://monetdb.cwi.nl'</a><br> !ERROR: !ERROR:
<br>'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\MonetDB4\.Int.32.32.d-Windows5.1\share\MonetDB\xrpc\export'
<br> !ERROR: !ERROR: '<a href="http://127.0.0.1:45558'" target="_top" rel="nofollow">http://127.0.0.1:45558'</a><br> !ERROR: !ERROR:
<br>'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\pathfinder\tests\XRpc\Tests'
<br> !ERROR: !ERROR:
<br>'C:\cygwin\var\tmp\monet.Int.32.32.d.20696\pathfinder\NT\mTests\tests\XRpc'
<br> !ERROR: !ERROR: CMDhttp_post: operation failed
<br> !ERROR: CMDhttp_post: operation failed.
<br>MAPI = monetdb@TURELUUR:37285
<br>QUERY = import module namespace xf = "xrpc-fun" at
<br>"C:\\cygwin\\var\\tmp\\monet.Int.32.32.d.20696\\pathfinder\\tests\\XRpc\\Tests/xrpcfun.xq";
<br>ERROR = !ERROR: doLoopLiftedRPC: execution failed at all destinations
<br>========
<br><br>For xprcConvert2, the problem is still the same as initially reported.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Ying Zhang (yingying)
<br>Date: 2008-02-16 22:16
<br><br>Message:
<br>Logged In: YES
<br>user_id=341633
<br>Originator: NO
<br><br>Lower the priority, as I don't know how to debug on windows.
<br>I will ask our windows expert for some help.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2008-02-16 01:44
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: YES
<br><br>Jennie,
<br><br>since the release is / will be based on the code base of 2008-02-13 (as
<br>tagged in CVS), there is no urgency to fix this before the release.
<br><br>Stefan
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Ying Zhang (yingying)
<br>Date: 2008-02-15 23:26
<br><br>Message:
<br>Logged In: YES
<br>user_id=341633
<br>Originator: NO
<br><br>I took a look at the errors, but could not get a clue without debugging.
<br><br>How urgent is this problem for the release? If urgent, I will ask Sjoerd
<br>for some debugging help next week, otherwise, I would like to lower the
<br>priority to solve it later.
<br><br>Jennie
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1893448&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1893448&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26538990&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26538950[ monetdb-Bugs-2903122 ] M5: --threads option exists? (monetdb5.conf documentation)2009-11-26T23:54:37Z2009-11-26T23:54:37ZSourceForge.netBugs item #2903122, was opened at 2009-11-24 14:41
<br>Message generated for change (Settings changed) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: Core
<br>Group: MonetDB Common CVS Head
<br>>Status: Closed
<br>Resolution: Invalid
<br>Priority: 5
<br>Private: No
<br>Submitted By: Stefan de Konink (skinkie)
<br>Assigned to: Stefan Manegold (stmane)
<br>Summary: M5: --threads option exists? (monetdb5.conf documentation)
<br><br>Initial Comment:
<br>/opt/monetdb-head/bin/mserver5 --dbfarm=/home/skinkie/monetdb --dbname=monitor --dbinit="include sql;" --threads 1
<br># MonetDB server v5.17.0, based on kernel v1.35.0
<br># Serving database 'monitor', using 2 threads
<br># Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
<br># Detected 4.000 GiB (4295331840 bytes) main-memory.
<br># Copyright (c) 1993-July 2008 CWI.
<br># Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information
<br>#warning: please don't forget to set your vault key!
<br>#(see /opt/monetdb-head/etc/monetdb5.conf)
<br>#01 1 140683233527553 sp = 140734711916164
<br>#01 nrofthreads 0
<br># Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
<br># MonetDB/SQL module v2.35.0 loaded
<br>#01 BBPdir: writing BBP.dir (527 bats).
<br>#WARNING: could not open file: /home/skinkie/monetdb/1
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 08:54
<br><br>Message:
<br>Closing --- see my previous comment.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-24 15:32
<br><br>Message:
<br>a)
<br>yes, the --threads option exist, but it does not take any argument; as
<br>monetdb5.conf says, "--threads" is merely a short cut for
<br>"--debug=$[1+128]", i.e., for setting debug bits 1 (thread-specific debug
<br>output) and 128 (PARMASK = Thread management).
<br>(As Fabian said, if you want to set the number of threads, you need to us
<br>"--set gdk_nr_threads=1"; cf. e.g.,
<br><a href="http://monetdb.cwi.nl/projects/monetdb/Development/Releases/Nov2008/index.html#Issues" target="_top" rel="nofollow">http://monetdb.cwi.nl/projects/monetdb/Development/Releases/Nov2008/index.html#Issues</a>)
<br><br>b)
<br>The error is perfectly fine and correct:
<br>Your mserver5 commanline specifies three option with arguments
<br>(--dbfarm=/home/skinkie/monetdb, --dbname=monitor, --dbinit="include
<br>sql;"), one without argument (--threads), and (the name of) one
<br>(MAL-)script to be executed (1) --- see also `mserver5 --help` --- the
<br>latter does not seem to exist in your working directory (you didn't specify
<br>an absolute path), and MonetDB informs you about that.
<br><br>IMHO all fine and correct.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Fabian (mr-meltdown)
<br>Date: 2009-11-24 14:48
<br><br>Message:
<br>there is just bogus in the monetdb5.conf file (misleading to say the
<br>least)
<br><br>there is no overwriting going on here, as the database properly reports to
<br>be serving "monitor", it simply complains that it can't find a file "1" to
<br>open. The --thread option seems to enable some magical debug output, and
<br>not do what you expect.
<br><br>Try --set gdk_nr_threads=1
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26538950&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26538948[ monetdb-Bugs-2903122 ] M5: --threads option exists? (monetdb5.conf documentation)2009-11-26T23:54:22Z2009-11-26T23:54:22ZSourceForge.netBugs item #2903122, was opened at 2009-11-24 14:41
<br>Message generated for change (Comment added) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: Core
<br>Group: MonetDB Common CVS Head
<br>Status: Open
<br>Resolution: Invalid
<br>Priority: 5
<br>Private: No
<br>Submitted By: Stefan de Konink (skinkie)
<br>>Assigned to: Stefan Manegold (stmane)
<br>Summary: M5: --threads option exists? (monetdb5.conf documentation)
<br><br>Initial Comment:
<br>/opt/monetdb-head/bin/mserver5 --dbfarm=/home/skinkie/monetdb --dbname=monitor --dbinit="include sql;" --threads 1
<br># MonetDB server v5.17.0, based on kernel v1.35.0
<br># Serving database 'monitor', using 2 threads
<br># Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked
<br># Detected 4.000 GiB (4295331840 bytes) main-memory.
<br># Copyright (c) 1993-July 2008 CWI.
<br># Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information
<br>#warning: please don't forget to set your vault key!
<br>#(see /opt/monetdb-head/etc/monetdb5.conf)
<br>#01 1 140683233527553 sp = 140734711916164
<br>#01 nrofthreads 0
<br># Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
<br># MonetDB/SQL module v2.35.0 loaded
<br>#01 BBPdir: writing BBP.dir (527 bats).
<br>#WARNING: could not open file: /home/skinkie/monetdb/1
<br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 08:54
<br><br>Message:
<br>Closing --- see my previous comment.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-24 15:32
<br><br>Message:
<br>a)
<br>yes, the --threads option exist, but it does not take any argument; as
<br>monetdb5.conf says, "--threads" is merely a short cut for
<br>"--debug=$[1+128]", i.e., for setting debug bits 1 (thread-specific debug
<br>output) and 128 (PARMASK = Thread management).
<br>(As Fabian said, if you want to set the number of threads, you need to us
<br>"--set gdk_nr_threads=1"; cf. e.g.,
<br><a href="http://monetdb.cwi.nl/projects/monetdb/Development/Releases/Nov2008/index.html#Issues" target="_top" rel="nofollow">http://monetdb.cwi.nl/projects/monetdb/Development/Releases/Nov2008/index.html#Issues</a>)
<br><br>b)
<br>The error is perfectly fine and correct:
<br>Your mserver5 commanline specifies three option with arguments
<br>(--dbfarm=/home/skinkie/monetdb, --dbname=monitor, --dbinit="include
<br>sql;"), one without argument (--threads), and (the name of) one
<br>(MAL-)script to be executed (1) --- see also `mserver5 --help` --- the
<br>latter does not seem to exist in your working directory (you didn't specify
<br>an absolute path), and MonetDB informs you about that.
<br><br>IMHO all fine and correct.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Fabian (mr-meltdown)
<br>Date: 2009-11-24 14:48
<br><br>Message:
<br>there is just bogus in the monetdb5.conf file (misleading to say the
<br>least)
<br><br>there is no overwriting going on here, as the database properly reports to
<br>be serving "monitor", it simply complains that it can't find a file "1" to
<br>open. The --thread option seems to enable some magical debug output, and
<br>not do what you expect.
<br><br>Try --set gdk_nr_threads=1
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903122&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26538948&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26538918[ monetdb-Bugs-2903111 ] Wrong error / no check for new database dir2009-11-26T23:52:25Z2009-11-26T23:52:25ZSourceForge.netBugs item #2903111, was opened at 2009-11-24 14:29
<br>Message generated for change (Comment added) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903111&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903111&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: Core
<br>Group: MonetDB Common CVS Head
<br>>Status: Closed
<br>>Resolution: Fixed
<br>Priority: 5
<br>Private: No
<br>Submitted By: Stefan de Konink (skinkie)
<br>>Assigned to: Stefan Manegold (stmane)
<br>Summary: Wrong error / no check for new database dir
<br><br>Initial Comment:
<br>Reproduce with:
<br>create (as root) a dbfarm directory
<br>switch to the normal user
<br>start monet with this dbfarm directory and (thus) a new database
<br><br>The error that arrises is:
<br>!FATAL: GDKlockHome: could not move to /home/skinkie/monetdb/monitor/
<br><br>It seems MonetDB did not check if the directory could not be created.
<br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 08:52
<br><br>Message:
<br>===================================================================
<br>2009/11/27 - stmane: MonetDB/src/gdk/gdk_utils.mx,1.255.2.6
<br><br>fixing ID: 2903111 "Wrong error / no check for new database dir"
<br><a href="https://sourceforge.net/tracker/?func=detail&aid=2903111&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=2903111&group_id=56967&atid=482468</a><br><br>give more detailed error message, whether we fail to create
<br>/dbfarm/dbname, or whether we fail to move (cd) to it
<br>===================================================================
<br><br>Closing w/o test, as this is merely an error message convenience, and
<br>making a portable test with platform-invariant output would be an
<br>overkill.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Fabian (mr-meltdown)
<br>Date: 2009-11-24 15:38
<br><br>Message:
<br>The bug here is that the message is misleading. MonetDB normally creates
<br>the directory for the dbname (monitor), however that failed (it must be),
<br>but apparently MonetDB didn't complain about that. It only fails once it
<br>tries to set its lock.
<br><br>We should complain earlier when we can't create the dbname dir, if it
<br>doesn't exist already.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-24 15:36
<br><br>Message:
<br>MonetDB does (try to) "move" (`cd`) to the dbfarm/dbname.
<br>If this fails for whatever reason (e.g., because the user isn't allowed to
<br>do so by filesystem access rights), this fails and MonetDB returns an
<br>error.
<br>Perfectly reasonable with me ...
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan de Konink (skinkie)
<br>Date: 2009-11-24 14:32
<br><br>Message:
<br>Fabian hinted also to write this is a permission error. Hence the dbfarm
<br>dir was owned by root, and not by the user that tried to start the
<br>database.
<br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903111&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2903111&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26538918&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26538805[ monetdb-Bugs-1607210 ] XQ: server-side compilation crash (member benchmark)2009-11-26T23:36:12Z2009-11-26T23:36:12ZSourceForge.netBugs item #1607210, was opened at 2006-12-02 01:27
<br>Message generated for change (Settings changed) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1607210&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1607210&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: PF/runtime
<br>Group: Pathfinder CVS Head
<br>>Status: Closed
<br>Resolution: Fixed
<br>Priority: 8
<br>Private: No
<br>Submitted By: Peter Boncz (boncz)
<br>Assigned to: Peter Boncz (boncz)
<br>Summary: XQ: server-side compilation crash (member benchmark)
<br><br>Initial Comment:
<br>Philippe Michiels reported at dagstuhl that the simple CHILD test of the MemBeR benchmark crashes when run in MapiClient -lx; crashing during the *compilation* process
<br><br>It works with pf standalone, translating to MIL explicitly
<br><br><a href="http://ilps.science.uva.nl/Resources/MemBeR/CHILD.html" target="_top" rel="nofollow">http://ilps.science.uva.nl/Resources/MemBeR/CHILD.html</a><br><br>The query used in this micro-benchmark navigates downward from the root, performing n successive child navigation steps, where n is to be varied.
<br>For every integer n, the corresponding query, named CHILD(n), is:
<br><br>/t1/t2/.../tn
<br><br>THE BUG ONLY OCCURS WITH n>15
<br>AND WITH SERVER SIDE COMPILATION
<br><br>may well be some buffer overflow..
<br><br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-27 08:36
<br><br>Message:
<br>After some more fixes (see below), the test now appears to work fine on all
<br>platforms except
<br><br>- timeout on
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Windows6.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d.1-Windows6.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d.1-Windows6.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d.1-Windows6.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br>This needs to be analyzed separately.
<br><br>- abort too deep recursion on
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br>This occurs with numerous tests on that platform and needs to be analyzed
<br>separately.
<br><br>Closing.
<br><br>===================================================================
<br>2009/11/23 - stmane: pathfinder/runtime/pathfinder.mx,1.462.4.13
<br><br>try even more conservative stack limit (PFmaxstack)
<br>for too deep recusion check (PFrecursion_fence())
<br>to check whether this helps to prevent the remaining segfaults with test
<br>tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210.*
<br>on some platforms; cf.
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-Debian4.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-Debian4.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Gentoo2.0.1/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Gentoo2.0.1/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br>===================================================================
<br>2009/11/24 - stmane:
<br>
<br>pathfinder/tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210_s00.xq,1.4.22.1
<br>
<br>pathfinder/tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210_s02.xq,1.4.22.1
<br><br>after
<br>"
<br>try even more conservative stack limit (PFmaxstack)
<br>for too deep recusion check (PFrecursion_fence())
<br>to check whether this helps to prevent the remaining segfaults with test
<br>tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210.*
<br>on some platforms; cf.
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-Debian4.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-Debian4.0/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Gentoo2.0.1/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Gentoo2.0.1/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.32.32.d.1-SunOS5.11/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br>"
<br>this test does not segfault any more,
<br>but runs into "too deep recursion" on the mentioned platforms
<br>earlier than on others; hence, we shrink parts of the test
<br>a bit.
<br>===================================================================
<br>2009/11/26 - stmane:
<br>
<br>pathfinder/tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210_s00.xq,1.4.22.2
<br>
<br>pathfinder/tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210_s02.xq,1.4.22.2
<br><br>shrink queries even more
<br>to avoid "aborted too deep recursion"
<br>also on 32-bit Debian4.0 & 32-bit SunOS5.11
<br>===================================================================
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-21 15:35
<br><br>Message:
<br>===================================================================
<br>2009/11/21 - stmane: pathfinder/runtime/pathfinder.mx,1.462.4.12
<br><br>use more conservative stack limit (PFmaxstack)
<br>for too deep recusion check (PFrecursion_fence())
<br>to prevent segfault (due to too deep recursion and
<br>consequent stack overflow) in tests
<br>tests/BugTracker/Tests/collection_management.SF-1726599.*
<br>tests/BugTracker/Tests/crash_on_concatenated_query.SF-1730547.*
<br>cf.,
<br><a href="https://sourceforge.net/tracker/?func=detail&aid=1607210&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=1607210&group_id=56967&atid=482468</a><br><a href="https://sourceforge.net/tracker/?func=detail&aid=1730547&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=1730547&group_id=56967&atid=482468</a><br>===================================================================
<br><br>Report can be closed once nightly testing confirms the fix for all
<br>platforms.
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-16 22:18
<br><br>Message:
<br>re-opened as the respective test fails on all platform with timeout or
<br>segfault at least since 2009.03.25
<br>cf.,
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/GNU.64.64.d.1-Fedora10/tests_BugTracker/server-side_compilation_crash.SF-1607210.out.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2008-02-10 14:12
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>Appears to work fine on most platforms, again;
<br>exceptions:
<br><br>Segfault on RedHat (on Itanium):
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d-RedHat4WS/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.64.d-RedHat4WS/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><br>Timeout on Windows:
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Stable/pathfinder/.mTests103/Int.64.32.d-Windows5.2/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2008-01-22 14:58
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>re-opened as it fails, again, since 2008.01.12, i.e., after checkins on
<br>2008.01.11, now with
<br><br>ERROR = !fatal error: aborted too deep recursion
<br><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.32.32.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Peter Boncz (boncz)
<br>Date: 2007-06-05 18:37
<br><br>Message:
<br>Logged In: YES
<br>user_id=591107
<br>Originator: YES
<br><br>fixed
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-06-04 21:19
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>See also
<br>[ 1730547 ] XQ: Mserver crashes on concatenated query
<br><a href="http://sourceforge.net/tracker/index.php?func=detail&aid=1730547&group_id=56967&atid=482468" target="_top" rel="nofollow">http://sourceforge.net/tracker/index.php?func=detail&aid=1730547&group_id=56967&atid=482468</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-05-29 10:24
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>See also
<br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.err.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugTracker/server-side_compilation_crash.SF-1607210.out.00.html</a><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-05-28 12:33
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>Added test in
<br>pathfinder/tests/BugTracker/Tests/server-side_compilation_crash.SF-1607210*
<br><br>As opposed to
<br>[ 1207048 ] XQuery: crashtest
<br><a href="http://sourceforge.net/tracker/index.php?func=detail&aid=1207048&group_id=56967&atid=482468" target="_top" rel="nofollow">http://sourceforge.net/tracker/index.php?func=detail&aid=1207048&group_id=56967&atid=482468</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.out.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.out.00.html</a><br><a href="http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.err.00.html" target="_top" rel="nofollow">http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora6/tests_BugDay_2005-11-09_0.9.3/xquery_crashtest.SF-1207048.err.00.html</a><br><br>it still crashes with a segfault, i.e.,
<br>the new fence does not catch the stack overflow in this case, yet.
<br><br>Not analysed, yet.
<br><br>Re-opend to keep us reminded to have another look at this one.
<br><br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Peter Boncz (boncz)
<br>Date: 2007-05-25 23:55
<br><br>Message:
<br>Logged In: YES
<br>user_id=591107
<br>Originator: YES
<br><br>fixed using a fence
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2006-12-11 16:10
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>I am pretty sure, this is a problem with recursive burg-generated code
<br>running out of stack space, which in unfortunately out of out stack space
<br>control, and hence IMHO very hard to prevent/fix...
<br>With "pf" standalone, it happens as well, just with a longer path
<br>expression (29 child steps in my case):
<br><br>$ n=28 ; i=0 ; q='doc("MyDoc.xml")' ; while [ $i -lt $n ] ; do i=$[i+1] ;
<br>q="$q`printf "/t%02d" $i`" ; done ; echo $q | tee /tmp/q.xq ; pf /tmp/q.xq
<br>>/dev/null
<br>doc("MyDoc.xml")/t01/t02/t03/t04/t05/t06/t07/t08/t09/t10/t11/t12/t13/t14/t15/t16/t17/t18/t19/t20/t21/t22/t23/t24/t25/t26/t27/t28
<br><br>$ n=29 ; i=0 ; q='doc("MyDoc.xml")' ; while [ $i -lt $n ] ; do i=$[i+1] ;
<br>q="$q`printf "/t%02d" $i`" ; done ; echo $q | tee /tmp/q.xq ; pf /tmp/q.xq
<br>>/dev/null
<br>doc("MyDoc.xml")/t01/t02/t03/t04/t05/t06/t07/t08/t09/t10/t11/t12/t13/t14/t15/t16/t17/t18/t19/t20/t21/t22/t23/t24/t25/t26/t27/t28/t29
<br>Segmentation fault
<br><br>$ gdb --args pf /tmp/q.xq
<br>GNU gdb Red Hat Linux (6.3.0.0-1.84rh)
<br>Copyright 2004 Free Software Foundation, Inc.
<br>GDB is free software, covered by the GNU General Public License, and you
<br>are
<br>welcome to change it and/or distribute copies of it under certain
<br>conditions.
<br>Type "show copying" to see the conditions.
<br>There is absolutely no warranty for GDB. Type "show warranty" for
<br>details.
<br>This GDB was configured as "x86_64-redhat-linux-gnu"...Using host
<br>libthread_db library "/lib64/libthread_db.so.1".
<br><br>(gdb) r
<br>Starting program:
<br>/net/corona.ins.cwi.nl/export/scratch0/manegold/Monet/Testing/Stable/.GNU.64.32.d.--disable-optimize_--enable-debug_--enable-assert.PREFIX./PATHFINDER/bin/pf
<br>/tmp/q.xq
<br>[Thread debugging using libthread_db enabled]
<br>[New Thread 46912496328800 (LWP 30516)]
<br><br>Program received signal SIGSEGV, Segmentation fault.
<br>[Switching to Thread 46912496328800 (LWP 30516)]
<br>0x0000000000483666 in reduce (p=0x2aaaaaae0b90, goalnt=50) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:874
<br>874 rule = PFfs_rule (STATE_LABEL (p), goalnt);
<br>(gdb) li
<br>869 PFpnode_t *kids[MAX_KIDS]; /* leaf nodes of this rule */
<br>870 bool topdown; /* is this a top-down rule? */
<br>871 PFcnode_t *c; /* temporary helper variable */
<br>872
<br>873 /* determine rule that matches for this non-terminal */
<br>874 rule = PFfs_rule (STATE_LABEL (p), goalnt);
<br>875
<br>876 /* PFinfo (OOPS_NOTICE, "match for rule %i", rule); */
<br>877 assert (rule);
<br>878
<br>(gdb) up
<br>#1 0x00000000004837d9 in reduce (p=0x2aaaaaae0b90, goalnt=49) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:929
<br>929 reduce (kids[i], nts[i]);
<br>(gdb) li
<br>924 /*
<br>925 * Recursively invoke compilation. This means bottom-up
<br>compilation.
<br>926 */
<br>927 if (!topdown)
<br>928 for (unsigned short i = 0; nts[i]; i++)
<br>929 reduce (kids[i], nts[i]);
<br>930
<br>931 /*
<br>932 PFinfo (OOPS_NOTICE, "processing rule %i", rule);
<br>933 */
<br>(gdb) bt
<br>#0 0x0000000000483666 in reduce (p=0x2aaaaaae0b90, goalnt=50) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:874
<br>#1 0x00000000004837d9 in reduce (p=0x2aaaaaae0b90, goalnt=49) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:929
<br>[...]
<br>#680 0x00000000004837d9 in reduce (p=0x2aaaaaae5f00, goalnt=1) at #680
<br>0x00000000004837d9 in reduce (p=0x2aaaaaae5f00, goalnt=1) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:929
<br>#681 0x0000000000491eaf in PFfs (r=0x2aaaaaae5f00) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:4209
<br>#682 0x00000000004065da in PFcompile (url=0x7ffffffb672e "/tmp/q.xq",
<br>pfout=0x38ecf318c0, status=0x731f40) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/compile.c:381
<br>#683 0x0000000000405f11 in main (argc=2, argv=0x7ffffffb4e78) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/main.c:905
<br>(gdb)
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:929
<br>#681 0x0000000000491eaf in PFfs (r=0x2aaaaaae5f00) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/core/fs.brg:4209
<br>#682 0x00000000004065da in PFcompile (url=0x7ffffffb672e "/tmp/q.xq",
<br>pfout=0x38ecf318c0, status=0x731f40) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/compile.c:381
<br>#683 0x0000000000405f11 in main (argc=2, argv=0x7ffffffb4e78) at
<br>/ufs/manegold/_/scratch0/Monet/Testing/Stable/pathfinder/compiler/main.c:905
<br>(gdb)
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2006-12-02 08:38
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>How does it crash?
<br>Is there any error message or exit status?
<br>Does it work for stand-alone pf also for n >> 15?
<br><br>Might well be a stack-overflow in some burg generated code...
<br><br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1607210&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1607210&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26538805&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26531039Re: Fatal error: aborted too deep recursion2009-11-26T07:28:10Z2009-11-26T07:28:10ZMartin KerstenHarm van Beek(DT) wrote:
<br>> During the development of our tool, using Xquery on MonetDB, we ran onto
<br>> the following error:
<br>>
<br>> ERROR = !fatal error: aborted too deep recursion
<br>>
<br>>>From previous posts (I guess) I can conclude that this relates to an
<br>> Xquery that simply is too big (too many operators) to create a good plan
<br>> for it.
<br>All systems have limits that either cause an emergency stop. In this case,
<br>the operating system has such a limit.
<br>>
<br>> I tried several MonetDB releases (May2009-SP2, Aug2009-SP2, Nov2009-DB0)
<br>> on Red Hat Enterprise Linux Server release 5.4 (Tikanga), all showing
<br>> the same result.
<br>>
<br>> Does anyone have any directions on how to solve this?
<br>Yes. Programming cleverness comes into play. Especially if you use a tool
<br>to generate queries, then that tool should be a little more conservative
<br>in the kind of queries generated.
<br><br>> Can we reconfigure the compiler to fix this (or recompile it with
<br>> options)?
<br>No. At some point the optimization space becomes just too big.
<br>People that can spare the time to find the optimal settings for a larger
<br>application space are more then welcome to try it out on the source code
<br>made available.
<br><br>regards, Martin
<div class='shrinkable-quote'><br>>
<br>> Thanks in advance!
<br>>
<br>> Harm van Beek
<br>> Netherlands Forensic Institute
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26531039&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a></div><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26531039&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26530909Fatal error: aborted too deep recursion2009-11-26T07:18:52Z2009-11-26T07:18:52ZHarm van Beek(DT)During the development of our tool, using Xquery on MonetDB, we ran onto
<br>the following error:
<br><br>ERROR = !fatal error: aborted too deep recursion
<br><br>>From previous posts (I guess) I can conclude that this relates to an
<br>Xquery that simply is too big (too many operators) to create a good plan
<br>for it.
<br><br>I tried several MonetDB releases (May2009-SP2, Aug2009-SP2, Nov2009-DB0)
<br>on Red Hat Enterprise Linux Server release 5.4 (Tikanga), all showing
<br>the same result.
<br><br>Does anyone have any directions on how to solve this?
<br>Can we reconfigure the compiler to fix this (or recompile it with
<br>options)?
<br><br>Thanks in advance!
<br><br>Harm van Beek
<br>Netherlands Forensic Institute
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26530909&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26530531Re: COPY INTO memory issues2009-11-26T06:52:15Z2009-11-26T06:52:15ZMartin KerstenAndreas Streichardt wrote:
<br>> On Wednesday 25 November 2009 19:29:02 Martin Kersten wrote:
<br>>> Andreas Streichardt wrote:
<br>>>> Hi,
<br>>> Hi Andreas,
<br>>
<br>> Hi Martin,
<br>>
<br>> thanks for your reply.
<br>>
<br><br><div class='shrinkable-quote'><br>>>>
<br>>>> My first attempt was:
<br>>>>
<br>>>> sql>copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>>>> ',','\n' null as '';
<br>>>> MAPI = monetdb@localhost:50000
<br>>>> ACTION= read_line
<br>>>> QUERY = copy into md_data FROM '/home/mop/performance/hans.csv'
<br>>>> delimiters ',','\n' null as '';
<br>>>> ERROR = Connection terminated
<br>>> did you kill the connection or the server?
<br>>> if the server crashed, then the input file would be interesting to look at
<br>>> for use.
<br>>
<br>> Server crashed when there wasn't any mem left (only have 1 Gig Swap).
<br>>
<br>>>> This was of course the quick and dirty way. What happened in the
<br>>>> background was that monetdb consumed all memory. My system is a debian
<br>>>> lenny 64bit. Server has 4GB and decent CPUs. Why does it need all my
<br>>>> memory here?
<br>>> Quote from the manual:" It is strongly adviced to announce the number of
<br>>> records to be inserted. It avoids guessing by the server and subsequent
<br>>> re-allocation of table space, which involves potential copying."
<br>>>
<br>>> Furthermore, MonetDB uses whatever memory there is. The tables constructed
<br>>> are memory mapped files.
<br>>
<br>> Does that mean that the maximum size of a table is limited by the available
<br>> RAM? More precisely: I can't work on a 100 million rows*100 columns table if i
<br>> only have 1GB Ram?
</div>This is a recurring question from people with different database experiences,
<br>and certainly needs a short chapter in the document.
<br><br>- No, the databases are *not* limited to the amount of physical memory (RAM)
<br> (the largest one running in our lab is 4TB on a single PC)
<br>- Yes, the temporary tables may be allocated in swap space, which calls for a
<br>larger setting then most
<br>- Yes, you can work with huge database
<br>- Yes, ideally the hot set you are working with fits in memory (RAM+swap)
<br><div class='shrinkable-quote'><br>>
<br>>>> I then did the same but loaded only parts of the file (1000000 rows).
<br>>>>
<br>>>> sql>copy 1000000 records into md_data FROM
<br>>>> '/home/mop/performance/hans.csv' delimiters ',','\n' null as '' ;
<br>>> Same issue. It will have to reallocate space, which may involve copying.
<br>>> (at least on the released versions)
<br>>
<br>> Hmm...Isn't the "1000000" records stuff the hint about the number of records
<br>> you mentioned earlier? If not how do i announce the number of records?
<br>> Couldn't find anything else here:
</div>The issue is mostly gone in the next release, which uses the memory mapped
<br>files for all non-trival sizes.
<br><div class='shrinkable-quote'><br>>
<br>> <a href="http://monetdb.cwi.nl/projects/monetdb/SQL/Documentation/Data-" target="_top" rel="nofollow">http://monetdb.cwi.nl/projects/monetdb/SQL/Documentation/Data-</a><br>> Manipulation.html
<br>>
<br>>>> This worked but the server used ~2-3GB of my RAM during the process and
<br>>>> what's even more important it used the same amount even though the copy
<br>>>> process finished. Why does it use the RAM even though it is not used
<br>>>> anymore?
<br>>> Tables are retained in memory until not needed and flushed.
<br>>> Persistent tables therefore likely stay around
<br>>
<br>> I still wonder how and if i can control how much memory monetdb is allowed to
</div>you can not.
<br><br>> use. When does monetdb decide that a table is not used anymore? My biggest
<br>> concern currently is that monetdb simply crashes due to memory outages (which
<br>> i already hit during loading ;) ). Are there any settings to limit the memory
<br>> usage or does monetdb require to have everything in RAM? Does it mean i need a
<br>> 8GB RAM Server if i want to work on a 2,1million rows*200 columns table?
<br>MonetDB will be happily sharing the 8 GB with the other processes.
<br>It may mean that your OS kicks in at some point if it feels there is an
<br>uneven balance of resource requests.
<br><br>Running a database server on a box in isolation is not uncommon for
<br>real-life applications.
<br><br>For those with hard memory constraints: we are experimenting with
<br>virtual engines to assess the impact and control the resources.
<br><div class='shrinkable-quote'><br>>
<br>>>> When i tried a:
<br>>>>
<br>>>> sql>SELECT COUNT(*) FROM md_data;
<br>>>>
<br>>>> it hang completely. Same when trying to establish a new connection.
<br>>> this is not what i would expect. Did you accidently ran the query as
<br>>> another parallel user? or the same session as the copy command?
<br>>
<br>> I think (!) it was the same session. But a new connection didn't work either.
</div>I think it is related to the low swap space situation.
<br>And a hard one to debug. Perhaps the next release (Nov2009) is already better,
<br>because we spent quite some effort to harden the code further
<br>using the Coverity tools
<div class='shrinkable-quote'><br>>
<br>>>> I stopped the server and restarted again. Everything was fine again. I
<br>>> good :)
<br>>>
<br>>> Memory leaks has been addressed in the Aug release, we are not aware of
<br>>> serious ones and would need to re-do the precise experiment to confirm it.
<br>>> What you see is memory fragmentation due to re-allocation of the BATs in
<br>>> memory.
<br>>
<br>> My problem is: How do i fix that?
</div>you can not.
<br>> I am currently inserting in 10k steps and i am hitting the memory problems
<br>> again. What should i do to insert my data? Stopping and starting the database
<br>> for each step? :S
<br>Enlarge the OS swap space is the road to take.
<div class='shrinkable-quote'><br>>
<br>> Interestingly i am seeing a big gap in loading speed somewhere in between:
<br>>
<br>> 880000
<br>> [ 10000 ]
<br>> Timer 4916.922 msec
<br>> 890000
<br>> [ 10000 ]
<br>> Timer 4477.197 msec
<br>> 900000
<br>> [ 10000 ]
<br>> Timer 4535.896 msec
<br>> 910000
<br>> [ 10000 ]
<br>> Timer 5708.848 msec
<br>> 920000
<br>> [ 10000 ]
<br>> Timer 19675.302 msec
<br>> 930000
<br>> [ 10000 ]
<br>> Timer 28901.041 msec
<br>> 940000
<br>> [ 10000 ]
<br>> Timer 28013.104 msec
<br>> 950000
<br>> [ 10000 ]
<br>> Timer 28235.442 msec
<br>> 960000
<br>> [ 10000 ]
<br>> Timer 28741.978 msec
<br>> 970000
<br>> [ 10000 ]
<br>> Timer 29581.496 msec
<br>>
<br>> The number indicates at which offset i am currently (~2.1 million rows in
<br>> total). It is not yet swapping.
</div>indeed after the memory is filled, it starts moving dirty pages out.
<br>>
<br>>> I think your expectations are not yet aligned with what MonetDB does
<br>>> differently from other systems.
<br>>
<br>> Hmm...What does it do differently? ;) I simply wonder how to insert 2.1
<br>The architecture of MonetDB is completely different from the row-store
<br>you are familiar with. It is documented in tens of papers, a glimps
<br>of the differences is summarised in the recent paper mentioned
<br>in thehome page
<br>"Database Architecture Evolution: Mammals Flourished long before Dinosaurs
<br>became Extinct "
<div class='shrinkable-quote'><br>> million rows currently :S (data layout see above). I could also post my quick
<br>> and dirty generation script here if that helps.
<br>>
<br>> Let me know if you need any more details!
<br>>
<br>> Thanks,
<br>>
<br>> Andreas Streichardt
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26530531&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a></div><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26530531&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26528933Re: COPY INTO memory issues2009-11-26T04:41:04Z2009-11-26T04:41:04ZAndreas Streichardt-2On Wednesday 25 November 2009 19:29:02 Martin Kersten wrote:
<br>> Andreas Streichardt wrote:
<br>> > Hi,
<br>>
<br>> Hi Andreas,
<br><br>Hi Martin,
<br><br>thanks for your reply.
<br><div class='shrinkable-quote'><br>> Thank you for taking the time to experiment and more importantly to report
<br>> back. There has been a lot of work on the loader, but its importance for
<br>> users to get started has our full attention.
<br>> An external evaluation of MonetDB using extensive copying is reported in:
<br>> <a href="http://www.mysqlperformanceblog.com/2009/09/29/quick-comparison-of-myisam-i" target="_top" rel="nofollow">http://www.mysqlperformanceblog.com/2009/09/29/quick-comparison-of-myisam-i</a><br>> nfobright-and-monetdb/ and
<br>> <a href="http://www.mysqlperformanceblog.com/2009/10/02/analyzing-air-traffic-perfor" target="_top" rel="nofollow">http://www.mysqlperformanceblog.com/2009/10/02/analyzing-air-traffic-perfor</a><br>> mance-with-infobright-and-monetdb/ and upto date information on the same
<br>> experiment in
<br>> <a href="http://www.cwi.nl/~mk/ontimeReport" target="_top" rel="nofollow">http://www.cwi.nl/~mk/ontimeReport</a></div><br>One of the very few blogs i read regulary ;)
<br><div class='shrinkable-quote'><br>> > i had a look at monetdb a while (a year or so) ago and stumbled upon a
<br>> > few crashes/memory issues back then. Apart from that it made a very
<br>> > promising impression to me.
<br>> >
<br>> > Today i thought i could give it a try again to see if these issues have
<br>> > been fixed and found out that copy into still seems to have memory issues
<br>> > and i wonder if i am the only one who hits such issues.
<br>>
<br>> First and important question is: Which version exactly did you use?
<br>> mserver5 --version
</div><br>MonetDB server v5.14.4 (64-bit), based on kernel v1.32.4 (64-bit oids)
<br>Copyright (c) 1993-July 2008 CWI
<br>Copyright (c) August 2008-2009 MonetDB B.V., all rights reserved
<br>Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information
<br>Configured for prefix: /usr
<br>Libraries:
<br> libpcre: 7.6 2008-01-28 (compiled with 7.6)
<br> openssl: OpenSSL 0.9.8g 19 Oct 2007 (compiled with OpenSSL 0.9.8g 19 Oct
<br>2007)
<br> libxml2: 2.6.32 (compiled with 2.6.32)
<br>Compiled by: root@debian64 (x86_64-pc-linux-gnu)
<br>Compilation: cc -O2 -Wall -O2 -O6 -fomit-frame-pointer
<br>-finline-functions -falign-loops=4 -falign-jumps=4 -falign-functions=4 -
<br>fexpensive-optimizations -funroll-loops -frerun-cse-after-
<br>loop -frerun-loop-opt -ftree-vectorize
<br>Linking : ld -IPA -m elf_x86_64
<br><br>(I am simply using the precompiled debs)
<br><div class='shrinkable-quote'><br>> > I am experimeting with dummy data created by a script.
<br>> >
<br>> > Data is created the following way:
<br>> >
<br>> > 20 smallints: Values 1-10.
<br>> > 100 tinyints: Values 0-1
<br>> > 20 ints: Values 1-10000
<br>> > 40 varchars: md5 of some random value
<br>> > 20 floats: rand(1000000)/1000000
<br>> >
<br>> > I created a CSV file with ~2.1 million rows and then tried to insert the
<br>> > data.
<br>> >
<br>> > My first attempt was:
<br>> >
<br>> > sql>copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>> > ',','\n' null as '';
<br>> > MAPI = monetdb@localhost:50000
<br>> > ACTION= read_line
<br>> > QUERY = copy into md_data FROM '/home/mop/performance/hans.csv'
<br>> > delimiters ',','\n' null as '';
<br>> > ERROR = Connection terminated
<br>>
<br>> did you kill the connection or the server?
<br>> if the server crashed, then the input file would be interesting to look at
<br>> for use.
</div><br>Server crashed when there wasn't any mem left (only have 1 Gig Swap).
<br>
<div class='shrinkable-quote'><br>> > This was of course the quick and dirty way. What happened in the
<br>> > background was that monetdb consumed all memory. My system is a debian
<br>> > lenny 64bit. Server has 4GB and decent CPUs. Why does it need all my
<br>> > memory here?
<br>>
<br>> Quote from the manual:" It is strongly adviced to announce the number of
<br>> records to be inserted. It avoids guessing by the server and subsequent
<br>> re-allocation of table space, which involves potential copying."
<br>>
<br>> Furthermore, MonetDB uses whatever memory there is. The tables constructed
<br>> are memory mapped files.
</div><br>Does that mean that the maximum size of a table is limited by the available
<br>RAM? More precisely: I can't work on a 100 million rows*100 columns table if i
<br>only have 1GB Ram?
<br><br>> > I then did the same but loaded only parts of the file (1000000 rows).
<br>> >
<br>> > sql>copy 1000000 records into md_data FROM
<br>> > '/home/mop/performance/hans.csv' delimiters ',','\n' null as '' ;
<br>>
<br>> Same issue. It will have to reallocate space, which may involve copying.
<br>> (at least on the released versions)
<br><br>Hmm...Isn't the "1000000" records stuff the hint about the number of records
<br>you mentioned earlier? If not how do i announce the number of records?
<br>Couldn't find anything else here:
<br><br><a href="http://monetdb.cwi.nl/projects/monetdb/SQL/Documentation/Data-" target="_top" rel="nofollow">http://monetdb.cwi.nl/projects/monetdb/SQL/Documentation/Data-</a><br>Manipulation.html
<br><br>> > This worked but the server used ~2-3GB of my RAM during the process and
<br>> > what's even more important it used the same amount even though the copy
<br>> > process finished. Why does it use the RAM even though it is not used
<br>> > anymore?
<br>>
<br>> Tables are retained in memory until not needed and flushed.
<br>> Persistent tables therefore likely stay around
<br><br>I still wonder how and if i can control how much memory monetdb is allowed to
<br>use. When does monetdb decide that a table is not used anymore? My biggest
<br>concern currently is that monetdb simply crashes due to memory outages (which
<br>i already hit during loading ;) ). Are there any settings to limit the memory
<br>usage or does monetdb require to have everything in RAM? Does it mean i need a
<br>8GB RAM Server if i want to work on a 2,1million rows*200 columns table?
<br><br>> > When i tried a:
<br>> >
<br>> > sql>SELECT COUNT(*) FROM md_data;
<br>> >
<br>> > it hang completely. Same when trying to establish a new connection.
<br>>
<br>> this is not what i would expect. Did you accidently ran the query as
<br>> another parallel user? or the same session as the copy command?
<br><br>I think (!) it was the same session. But a new connection didn't work either.
<br><br>> > I stopped the server and restarted again. Everything was fine again. I
<br>>
<br>> good :)
<br>>
<br>> Memory leaks has been addressed in the Aug release, we are not aware of
<br>> serious ones and would need to re-do the precise experiment to confirm it.
<br>> What you see is memory fragmentation due to re-allocation of the BATs in
<br>> memory.
<br><br>My problem is: How do i fix that?
<br>I am currently inserting in 10k steps and i am hitting the memory problems
<br>again. What should i do to insert my data? Stopping and starting the database
<br>for each step? :S
<br><br>Interestingly i am seeing a big gap in loading speed somewhere in between:
<br><br>880000
<br>[ 10000 ]
<br>Timer 4916.922 msec
<br>890000
<br>[ 10000 ]
<br>Timer 4477.197 msec
<br>900000
<br>[ 10000 ]
<br>Timer 4535.896 msec
<br>910000
<br>[ 10000 ]
<br>Timer 5708.848 msec
<br>920000
<br>[ 10000 ]
<br>Timer 19675.302 msec
<br>930000
<br>[ 10000 ]
<br>Timer 28901.041 msec
<br>940000
<br>[ 10000 ]
<br>Timer 28013.104 msec
<br>950000
<br>[ 10000 ]
<br>Timer 28235.442 msec
<br>960000
<br>[ 10000 ]
<br>Timer 28741.978 msec
<br>970000
<br>[ 10000 ]
<br>Timer 29581.496 msec
<br><br>The number indicates at which offset i am currently (~2.1 million rows in
<br>total). It is not yet swapping.
<br><br>> I think your expectations are not yet aligned with what MonetDB does
<br>> differently from other systems.
<br><br>Hmm...What does it do differently? ;) I simply wonder how to insert 2.1
<br>million rows currently :S (data layout see above). I could also post my quick
<br>and dirty generation script here if that helps.
<br><br>Let me know if you need any more details!
<br>
<br>Thanks,
<br><br> Andreas Streichardt
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26528933&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26518529Re: COPY INTO memory issues2009-11-25T11:07:26Z2009-11-25T11:07:26Zjoehl> > This wasn't the fact. Again monetdb consumed all my memory and when loading
<br>> > finished it indicated that it was still doing something (CPU @ ~20%).
<br>> log records have to be written, your OS may kick in to flush dirty pages,
<br>> reloading swapped out processes, ....
<br><br>The following might not apply to Linux, but still: under windows, suddenly flushing dirty pages from large memory mapped files can severly degrade the system performance ("stalled"). I know windows has settings to limit the amount of dirty pages that may accumulate without flushing. Maybe Linux has something similar and maybe this helps here.
<br><br>Please keep me in the loop if you try this route.
<br><br><br>Jens
<br><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26518529&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26517940Re: COPY INTO memory issues2009-11-25T10:29:02Z2009-11-25T10:29:02ZMartin KerstenAndreas Streichardt wrote:
<br>> Hi,
<br>Hi Andreas,
<br><br>Thank you for taking the time to experiment and more importantly to report back.
<br>There has been a lot of work on the loader, but its importance for users to
<br>get started has our full attention.
<br>An external evaluation of MonetDB using extensive copying is reported in:
<br><a href="http://www.mysqlperformanceblog.com/2009/09/29/quick-comparison-of-myisam-infobright-and-monetdb/" target="_top" rel="nofollow">http://www.mysqlperformanceblog.com/2009/09/29/quick-comparison-of-myisam-infobright-and-monetdb/</a><br>and
<br><a href="http://www.mysqlperformanceblog.com/2009/10/02/analyzing-air-traffic-performance-with-infobright-and-monetdb/" target="_top" rel="nofollow">http://www.mysqlperformanceblog.com/2009/10/02/analyzing-air-traffic-performance-with-infobright-and-monetdb/</a><br>and upto date information on the same experiment in
<br><a href="http://www.cwi.nl/~mk/ontimeReport" target="_top" rel="nofollow">http://www.cwi.nl/~mk/ontimeReport</a><br><br>>
<br>> i had a look at monetdb a while (a year or so) ago and stumbled upon a few
<br>> crashes/memory issues back then. Apart from that it made a very promising
<br>> impression to me.
<br>>
<br>> Today i thought i could give it a try again to see if these issues have been
<br>> fixed and found out that copy into still seems to have memory issues and i
<br>> wonder if i am the only one who hits such issues.
<br>First and important question is: Which version exactly did you use?
<br>mserver5 --version
<br><div class='shrinkable-quote'><br>>
<br>> I am experimeting with dummy data created by a script.
<br>>
<br>> Data is created the following way:
<br>>
<br>> 20 smallints: Values 1-10.
<br>> 100 tinyints: Values 0-1
<br>> 20 ints: Values 1-10000
<br>> 40 varchars: md5 of some random value
<br>> 20 floats: rand(1000000)/1000000
<br>>
<br>> I created a CSV file with ~2.1 million rows and then tried to insert the data.
<br>>
<br>> My first attempt was:
<br>>
<br>> sql>copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>> ',','\n' null as '';
<br>> MAPI = monetdb@localhost:50000
<br>> ACTION= read_line
<br>> QUERY = copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>> ',','\n' null as '';
<br>> ERROR = Connection terminated
<br>>
</div>did you kill the connection or the server?
<br>if the server crashed, then the input file would be interesting to look at
<br>for use.
<br>> This was of course the quick and dirty way. What happened in the background
<br>> was that monetdb consumed all memory. My system is a debian lenny 64bit.
<br>> Server has 4GB and decent CPUs. Why does it need all my memory here?
<br>Quote from the manual:" It is strongly adviced to announce the number of records
<br>to be inserted. It avoids guessing by the server and subsequent re-allocation of
<br>table space, which involves potential copying."
<br><br>Furthermore, MonetDB uses whatever memory there is. The tables constructed are
<br>memory mapped files.
<br><br>>
<br>> I then did the same but loaded only parts of the file (1000000 rows).
<br>>
<br>> sql>copy 1000000 records into md_data FROM '/home/mop/performance/hans.csv'
<br>> delimiters ',','\n' null as '' ;
<br>Same issue. It will have to reallocate space, which may involve copying.
<br>(at least on the released versions)
<br>>
<br>> This worked but the server used ~2-3GB of my RAM during the process and what's
<br>> even more important it used the same amount even though the copy process
<br>> finished. Why does it use the RAM even though it is not used anymore?
<br>Tables are retained in memory until not needed and flushed.
<br>Persistent tables therefore likely stay around
<br><br>>
<br>> Then i tried to insert 1000000 rows more and my expectation was that monetdb
<br>> would reuse the memory it allocated (and didn't free) in the first loading
<br>> process.
<br>>
<br>> This wasn't the fact. Again monetdb consumed all my memory and when loading
<br>> finished it indicated that it was still doing something (CPU @ ~20%).
<br>log records have to be written, your OS may kick in to flush dirty pages,
<br>reloading swapped out processes, ....
<br><br>>
<br>> When i tried a:
<br>>
<br>> sql>SELECT COUNT(*) FROM md_data;
<br>>
<br>> it hang completely. Same when trying to establish a new connection.
<br>this is not what i would expect. Did you accidently ran the query as another
<br>parallel user? or the same session as the copy command?
<br><br>>
<br>> I stopped the server and restarted again. Everything was fine again. I
<br>good :)
<br><br>> inserted the rest of the data (memory raised again but there were only 100000
<br>> rows left so there wasn't any problem this time).
<br>>
<br>> Am i the only one hitting such issues? One could argue that i should import
<br>> less records at once but even then. There seems to be some nasty memleak in
<br>Memory leaks has been addressed in the Aug release, we are not aware of serious
<br>ones and would need to re-do the precise experiment to confirm it.
<br>What you see is memory fragmentation due to re-allocation of the BATs in memory.
<br><br>> the background. Or is this a configuration issue? Or am i doing something
<br>> completely wrong?
<br><br>I think your expectations are not yet aligned with what MonetDB does differently
<br>from other systems.
<br><br>>
<br>> I am using the .debs: <a href="http://monetdb.cwi.nl/downloads/Debian/" target="_top" rel="nofollow">http://monetdb.cwi.nl/downloads/Debian/</a><br>>
<br>> Thanks in advance,
<br>thanks to you for sharing this information,
<br>Martin Kersten
<div class='shrinkable-quote'><br>>
<br>> Andreas Streichardt
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517940&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a></div><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517940&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26517278Re: 'create table as select...' from union queries doesn't work2009-11-25T09:46:49Z2009-11-25T09:46:49ZNiels NesOn Wed, Nov 25, 2009 at 09:41:29AM -0800, Alex Bo. wrote:
<div class='shrinkable-quote'><br>>
<br>> Hi,
<br>>
<br>> If I do this:
<br>>
<br>> create table t1 (f1 int);
<br>> create table t2 (f1 int);
<br>> create table t_union as (select f1 from t1 union all select f1 from t2) with
<br>> data;
<br>>
<br>> Then no table is created and no exception is thrown.
<br>> Should I open a bug for this?
<br>>
<br>>
<br>> (AugSP2 on Windows)
</div>just tested on linux Nov2009 release and that doesn't have this problem
<br><br>Niels
<div class='shrinkable-quote'><br>> --
<br>> View this message in context: <a href="http://old.nabble.com/%27create-table-as-select...%27-from-union-queries-doesn%27t-work-tp26517185p26517185.html" target="_top" rel="nofollow">http://old.nabble.com/%27create-table-as-select...%27-from-union-queries-doesn%27t-work-tp26517185p26517185.html</a><br>> Sent from the monetdb-users mailing list archive at Nabble.com.
<br>>
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517278&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a></div><br>--
<br>Niels Nes, Centrum Wiskunde & Informatica (CWI)
<br>Science Park 123, 1098 XG Amsterdam, The Netherlands
<br>room C0.02/M3.46, phone ++31 20 592-4098
<br>url: <a href="http://www.cwi.nl/~niels" target="_top" rel="nofollow">http://www.cwi.nl/~niels</a> e-mail: <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517278&i=1" target="_top" rel="nofollow">Niels.Nes@...</a>
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517278&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26517185'create table as select...' from union queries doesn't work2009-11-25T09:41:29Z2009-11-25T09:41:29ZAlex Bo.Hi,
<br><br>If I do this:
<br><br>create table t1 (f1 int);
<br>create table t2 (f1 int);
<br>create table t_union as (select f1 from t1 union all select f1 from t2) with data;
<br><br>Then no table is created and no exception is thrown.
<br>Should I open a bug for this?
<br><br><br>(AugSP2 on Windows)<p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26517544COPY INTO memory issues2009-11-25T09:37:52Z2009-11-25T09:37:52ZAndreas Streichardt-2Hi,
<br><br>i had a look at monetdb a while (a year or so) ago and stumbled upon a few
<br>crashes/memory issues back then. Apart from that it made a very promising
<br>impression to me.
<br><br>Today i thought i could give it a try again to see if these issues have been
<br>fixed and found out that copy into still seems to have memory issues and i
<br>wonder if i am the only one who hits such issues.
<br><br>I am experimeting with dummy data created by a script.
<br><br>Data is created the following way:
<br><br>20 smallints: Values 1-10.
<br>100 tinyints: Values 0-1
<br>20 ints: Values 1-10000
<br>40 varchars: md5 of some random value
<br>20 floats: rand(1000000)/1000000
<br><br>I created a CSV file with ~2.1 million rows and then tried to insert the data.
<br><br>My first attempt was:
<br><br>sql>copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>',','\n' null as '';
<br>MAPI = monetdb@localhost:50000
<br>ACTION= read_line
<br>QUERY = copy into md_data FROM '/home/mop/performance/hans.csv' delimiters
<br>',','\n' null as '';
<br>ERROR = Connection terminated
<br><br>This was of course the quick and dirty way. What happened in the background
<br>was that monetdb consumed all memory. My system is a debian lenny 64bit.
<br>Server has 4GB and decent CPUs. Why does it need all my memory here?
<br><br>I then did the same but loaded only parts of the file (1000000 rows).
<br><br>sql>copy 1000000 records into md_data FROM '/home/mop/performance/hans.csv'
<br>delimiters ',','\n' null as '' ;
<br><br>This worked but the server used ~2-3GB of my RAM during the process and what's
<br>even more important it used the same amount even though the copy process
<br>finished. Why does it use the RAM even though it is not used anymore?
<br><br>Then i tried to insert 1000000 rows more and my expectation was that monetdb
<br>would reuse the memory it allocated (and didn't free) in the first loading
<br>process.
<br><br>This wasn't the fact. Again monetdb consumed all my memory and when loading
<br>finished it indicated that it was still doing something (CPU @ ~20%).
<br><br>When i tried a:
<br><br>sql>SELECT COUNT(*) FROM md_data;
<br><br>it hang completely. Same when trying to establish a new connection.
<br><br>I stopped the server and restarted again. Everything was fine again. I
<br>inserted the rest of the data (memory raised again but there were only 100000
<br>rows left so there wasn't any problem this time).
<br><br>Am i the only one hitting such issues? One could argue that i should import
<br>less records at once but even then. There seems to be some nasty memleak in
<br>the background. Or is this a configuration issue? Or am i doing something
<br>completely wrong?
<br><br>I am using the .debs: <a href="http://monetdb.cwi.nl/downloads/Debian/" target="_top" rel="nofollow">http://monetdb.cwi.nl/downloads/Debian/</a><br><br>Thanks in advance,
<br><br> Andreas Streichardt
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26517544&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26513568[ monetdb-Bugs-1809586 ] PF: busy port crashes Mserver2009-11-25T06:28:04Z2009-11-25T06:28:04ZSourceForge.netBugs item #1809586, was opened at 2007-10-08 18:23
<br>Message generated for change (Comment added) made by stmane
<br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1809586&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1809586&group_id=56967</a><br><br>Please note that this message will contain a full copy of the comment thread,
<br>including the initial issue submission, for this request,
<br>not just the latest update.
<br>Category: PF/runtime
<br>Group: (zombie: Pathfinder 0.20)
<br>Status: Closed
<br>Resolution: None
<br>Priority: 5
<br>Private: No
<br>Submitted By: Jan Rittinger (tsheyar)
<br>Assigned to: Nobody/Anonymous (nobody)
<br>Summary: PF: busy port crashes Mserver
<br><br>Initial Comment:
<br>Currently (probably not after a restart anymore) the port 50001 on my PC is in use.
<br><br>This causes MonetDB to crash if I load 'module(pathfinder)' whereas loading only 'module(mapi)' does not kill the Mserver:
<br><br>$> Mserver --set='mapi_port=50000' --dbinit='module(mapi);'
<br># MonetDB Server v4.21.0
<br># based on GDK v1.21.0
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>MonetDB>
<br>$> Mserver --set='mapi_port=50001' --dbinit='module(mapi);'
<br># MonetDB Server v4.21.0
<br># based on GDK v1.21.0
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>!ERROR: MAPIlisten:binding to stream socket (50001) failed
<br>!OS: Address already in use
<br>MonetDB>
<br>$> Mserver --set='mapi_port=50000' --dbinit='module(pathfinder);'
<br># MonetDB Server v4.21.0
<br># based on GDK v1.21.0
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br># PF/Tijah module v0.3.0 loaded. <a href="http://dbappl.cs.utwente.nl/pftijah" target="_top" rel="nofollow">http://dbappl.cs.utwente.nl/pftijah</a><br># MonetDB/XQuery module v0.21.0 loaded
<br># XRPC administrative console at <a href="http://localhost:50001/admin" target="_top" rel="nofollow">http://localhost:50001/admin</a><br>01/Jan/1970 01:00:00 shttpd_open_port: bind(50001): Address already in use
<br>$> Mserver --set='mapi_port=50001' --dbinit='module(pathfinder);'
<br># MonetDB Server v4.21.0
<br># based on GDK v1.21.0
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for i686-suse-linux/32bit with 32bit OIDs; dynamically linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br># PF/Tijah module v0.3.0 loaded. <a href="http://dbappl.cs.utwente.nl/pftijah" target="_top" rel="nofollow">http://dbappl.cs.utwente.nl/pftijah</a><br>!ERROR: MAPIlisten:binding to stream socket (50001) failed
<br>!OS: Address already in use
<br>$>
<br><br>----------------------------------------------------------------------
<br><br>>Comment By: Stefan Manegold (stmane)
<br>Date: 2009-11-25 15:28
<br><br>Message:
<br>===================================================================
<br>2009/11/25 - stmane: testing/src/Mfilter.py.in,1.5.4.3
<br> pathfinder/tests/BugTracker/Tests/port_busy.SF-1809586.py,1.2.6.3
<br>
<br>pathfinder/tests/BugTracker/Tests/port_busy.SF-1809586.stable.err,1.2.6.1
<br>
<br>pathfinder/tests/BugTracker/Tests/port_busy.SF-1809586.stable.out,1.2.6.1
<br><br>slightly extended test and approved stable output for
<br>ID: 1809586 "PF: busy port crashes Mserver"
<br><a href="https://sourceforge.net/tracker/?func=detail&aid=1809586&group_id=56967&atid=482468" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&aid=1809586&group_id=56967&atid=482468</a><br>and added extra Mfilter rule to cope with variable output
<br>due to changing port numbers.
<br><br>We accept the fact that starting a MonetDB/XQury server (`Mserver
<br>--dbinit="module(pathfinder);"`)
<br>quits instantly (with a proper error message)
<br>in case either the MAPIPORT or the XQUERYPORT are not available
<br>(the server does not make much sense without access to these ports).
<br>The default ports can easily be overwritten on the server commandline:
<br>--set mapi_port=<portno> --set xrpc_port=<portno>
<br>===================================================================
<br><br><br>----------------------------------------------------------------------
<br><br>Comment By: Peter Boncz (boncz)
<br>Date: 2009-11-12 10:35
<br><br>Message:
<br>requested to remove this test
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Sjoerd Mullender (sjoerd)
<br>Date: 2009-07-31 10:45
<br><br>Message:
<br>Added test in pathfinder/tests/BugTracker/Tests: port_busy.SF-1809586.
<br>No correct stable output yet.
<br><br>----------------------------------------------------------------------
<br><br>Comment By: Stefan Manegold (stmane)
<br>Date: 2007-10-08 18:33
<br><br>Message:
<br>Logged In: YES
<br>user_id=572415
<br>Originator: NO
<br><br>loading module sql_server also crashed the Msever instantly if the
<br>requested mapi_port is busy:
<br><br>$ Mserver --dbinit='module(sql_server);'
<br># MonetDB Server v4.20.0
<br># based on GDK v1.20.0
<br># Copyright (c) 1993-2007, CWI. All rights reserved.
<br># Compiled for x86_64-redhat-linux-gnu/64bit with 64bit OIDs; dynamically
<br>linked.
<br># Visit <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a> for further information.
<br>!ERROR: MAPIlisten:binding to stream socket (50000) failed
<br>!OS: Address already in use
<br>$
<br><br><br>----------------------------------------------------------------------
<br><br>You can respond by visiting:
<br><a href="https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1809586&group_id=56967" target="_top" rel="nofollow">https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1809586&group_id=56967</a><br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>Monetdb-bugs mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26513568&i=0" target="_top" rel="nofollow">Monetdb-bugs@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-bugs" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-bugs</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-bugs-f4363.html" embed="fixTarget[4363]" target="_top" >monetdb-bugs</a></p>tag:old.nabble.com,2006:post-26504918create function2009-11-24T15:06:48Z2009-11-24T15:06:48Zpphillips99@gmail.comThank you for the find piped help to the src documentation.
<br><br>I can make the following function using mclient from AUG2009SP2 well:
<br><br>create function f(i integer) returns bool
<br>begin
<br>return true;
<br>end;
<br><br>It seems that the create function is failing only in the SQL editor for
<br>SQuirreL client 3.0.2. I am using the compiled jdbc jar of
<br>MonetDB-AUG2009SP2 and have tried to use a compiled sourceball version
<br>of jar from 11NOV2009 current branch.
<br><br>both provide this error:
<br><br>Error: syntax error, unexpected $end, expecting WHILE in: "create
<br>function f(i integer) returns bool
<br>begin
<br>return true;"
<br>SQLState: null
<br>ErrorCode: 0
<br>Error occured in:
<br>create function f(i integer) returns bool
<br>begin
<br>return true
<br><br>Should this then be a question for SQuirrel Client?
<br><br>Thanks
<br>Paul
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26504918&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26504154Re: temporary tables2009-11-24T14:09:11Z2009-11-24T14:09:11ZNiels NesOn Tue, Nov 24, 2009 at 02:04:15PM -0800, Alex Bo. wrote:
<br>>
<br>> Hi,
<br>>
<br>> How do temporary tables behave?
<br>> Are they only stored in memory or are they written to the disk too like
<br>> normal tables?
<br>>
<br>> What performance should I expect from using "create temporary table..." and
<br>> then querying it in comparison to having a nested query?
<br><br>Temporary tables are if needed written to disk. Performance all
<br>depend on the queries. If the nested queries can be properly optimized
<br>then they should perform better than temporary tables. When lots
<br>of work goes in to the query filling the temporary table then the later
<br>may indeed be faster.
<br><br>Niels
<div class='shrinkable-quote'><br>>
<br>> Thanks.
<br>>
<br>> --
<br>> View this message in context: <a href="http://old.nabble.com/temporary-tables-tp26504075p26504075.html" target="_top" rel="nofollow">http://old.nabble.com/temporary-tables-tp26504075p26504075.html</a><br>> Sent from the monetdb-users mailing list archive at Nabble.com.
<br>>
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26504154&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a></div><br>--
<br>Niels Nes, Centrum Wiskunde & Informatica (CWI)
<br>Science Park 123, 1098 XG Amsterdam, The Netherlands
<br>room C0.02/M3.46, phone ++31 20 592-4098
<br>url: <a href="http://www.cwi.nl/~niels" target="_top" rel="nofollow">http://www.cwi.nl/~niels</a> e-mail: <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26504154&i=1" target="_top" rel="nofollow">Niels.Nes@...</a>
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26504154&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26504075temporary tables2009-11-24T14:04:15Z2009-11-24T14:04:15ZAlex Bo.Hi,
<br><br>How do temporary tables behave?
<br>Are they only stored in memory or are they written to the disk too like normal tables?
<br><br>What performance should I expect from using "create temporary table..." and then querying it in comparison to having a nested query?
<br><br>Thanks.
<br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26499062Re: Database Failure2009-11-24T08:29:21Z2009-11-24T08:29:21ZMatthew Jones-5I tend to look at software problems before I look at hardware problems... I'm much more inclined to believe that the deadlock was a mutex issue and the timing of it happening caused the transaction log record file to be truncated. Hits on memory from cosmic rays? That's rather grandiose... I think you should go re-read Occam's Razor.<br>
<br clear="all">73,<br>Matthew W. Jones (KI4ZIB)<br><a href="http://matburt.net" target="_top" rel="nofollow">http://matburt.net</a><br>
<br><br><div class="gmail_quote">On Tue, Nov 24, 2009 at 9:55 AM, Sam Mason <span dir="ltr"><<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26499062&i=0" target="_top" rel="nofollow">sam@...</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="im">On Tue, Nov 24, 2009 at 09:42:08AM -0500, Matthew Jones wrote:<br>
> Memory tests were run and no problems were reported. With MonetDB+bad<br>
> memory I would expect to lose some data and no data was lost here.<br>
<br>
</div>You lost the contents of one file didn't you? Hits on memory from<br>
cosmic rays cause random single bit flips at a small but measurable<br>
rate--though ECC memory should help. These could cause something in<br>
Linux's page buffer to go awry and hence ending up with an empty file<br>
when unexpected. The bug rate of normal software means that that<br>
these sorts of errors are almost completely masked, however when code<br>
matures these sorts of errors start to become important. I'm not sure<br>
if Monet is stable enough to think about this yet, but if you've got a<br>
stable access pattern that only hits exactly the same code paths then it<br>
shouldn't matter about the remaining bugs.<br>
<br>
Higher assurance software tends to include checksums and other simple<br>
invariants that are checked at various places in code to make sure that<br>
errors like this aren't propagated too far.<br>
<font color="#888888"><br>
--<br>
</font><div><div></div><div class="h5"> Sam <a href="http://samason.me.uk/" target="_blank" rel="nofollow">http://samason.me.uk/</a><br>
<br>
------------------------------------------------------------------------------<br>
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day<br>
trial. Simplify your report design, integration and deployment - and focus on<br>
what you do best, core application coding. Discover what's new with<br>
Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_blank" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>
_______________________________________________<br>
MonetDB-users mailing list<br>
<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26499062&i=1" target="_top" rel="nofollow">MonetDB-users@...</a><br>
<a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_blank" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>
</div></div></blockquote></div><br>
<br />------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br />_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26499062&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26498134Re: Mx documentation2009-11-24T07:38:21Z2009-11-24T07:38:21ZStefan ManegoldOn Tue, Nov 24, 2009 at 09:15:43AM -0600, <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26498134&i=0" target="_top" rel="nofollow">pphillips99@...</a> wrote:
<div class='shrinkable-quote'><br>> Stefan,
<br>>
<br>> Thank you again.
<br>>
<br>> I would like to use CREATE FUNCTION and have been successful with this
<br>>
<br>> create function a (x int, y int)
<br>> returns int
<br>> return x*y;
<br>>
<br>> yet I try to do more complete versions with begin/end like this (and
<br>> many similar variations) without success.
<br>>
<br>> create function b (x int)
<br>> returns int
<br>> begin
<br>> declare retval int;
<br>> select someval into retval from mytbl where id=x;
<br>> return retval;
<br>> end;
<br>>
<br>> I am confused with trying to correctly learn the documentation. I
<br>> searched src/tests for examples but they are only easy like my first
<br>> that works.
</div><br>The following files might contain some more helpful examples:
<br>========
<br>$ find sql -type f -name \*.sql | xargs grep -il 'create function' | xargs grep -ilw begin | xargs grep -ilw end
<br>sql/src/sql/skyserver.sql
<br>sql/src/test/BugTracker-2009/Tests/subquery_returns_table_crash.SF-2513620.sql
<br>sql/src/test/BugTracker-2009/Tests/function_call_not_handled.SF-2864179.sql
<br>sql/src/test/BugTracker-2009/Tests/coercion.sql
<br>sql/src/test/BugTracker-2009/Tests/binop.SF.2499537.sql
<br>sql/src/test/BugTracker-2009/Tests/error_while_drop_func_returns_table_type.SF-2604573.sql
<br>sql/src/test/BugTracker-2009/Tests/set_and_arg_null.SF-2812729.sql
<br>sql/src/test/function_syntax.sql
<br>sql/src/test/zones.sql
<br>sql/src/test/Via-m/create-viam.sql
<br>sql/src/test/psm.sql
<br>sql/src/test/BugTracker/Tests/decl_sec_tab_in_func_body.SF-1907858.sql
<br>sql/src/test/BugTracker/Tests/schema_from_declared_table.SF-1983341.sql
<br>sql/src/test/BugTracker/Tests/drop_function.SF-1910396.sql
<br>sql/src/test/BugTracker/Tests/sql_prepare_prob.SF-1909638.sql
<br>sql/src/test/BugTracker/Tests/table_function_column_crash.SF-1913572.sql
<br>sql/src/test/BugTracker/Tests/variable_names.SF-1838816.sql
<br>sql/src/test/BugTracker/Tests/bat_descriptor_problem.SF-1997643.sql
<br>sql/src/test/BugTracker/Tests/view_creation_using_func.SF-1714814.sql
<br>sql/src/test/BugTracker/Tests/order_by_over_table_from_another_function.SF-1959822.sql
<br>sql/src/test/BugTracker/Tests/is_not_null_syntax_error.SF-1959456.sql
<br>sql/src/test/BugTracker/Tests/orderby_in_function.SF-1892708.sql
<br>sql/src/test/BugTracker/Tests/mdb_starts_with_sql_debug_64.SF-1999354.sql
<br>sql/src/test/BugTracker/Tests/coalesce_equal_coalesce.SF-1907905.sql
<br>sql/src/test/crashme.sql
<br>sql/src/test/pg_regress/sql/triggers.sql
<br>sql/src/test/pg_regress/sql/copy2.sql
<br>sql/src/test/pg_regress/sql/portals.sql
<br>sql/src/test/pg_regress/sql/stats.sql
<br>sql/src/test/pg_regress/sql/rangefuncs.sql
<br>sql/src/test/pg_regress/sql/plpgsql.sql
<br>sql/src/test/BugTracker-2008/Tests/col_from_another_inside_function_body.SF-2420768.sql
<br>sql/src/test/Skyserver/Skyserver_functions_v6.sql
<br>sql/src/test/Skyserver/Skyserver_functions.sql
<br>sql/src/test/Triggers/triggers_schema_operations.sql
<br>sql/src/test/DropCascade/Tests/drop_table_cascade.sql
<br>sql/src/test/DropCascade/Tests/drop_function_cascade.sql
<br>sql/src/test/DropCascade/Tests/drop_view_cascade.sql
<br>sql/src/test/DropCascade/Tests/drop_columns_cascade.sql
<br>sql/src/test/BugDay_2005-10-06_2.9.3/Tests/slow_counting_1.SF-923709.sql
<br>sql/src/test/Dependencies/dependency_DBobjects.sql
<br>sql/src/test/Dependencies/dependency_functions.sql
<br>sql/src/test/bugs/Tests/select_into.sql
<br>========
<br><br>> Separately, I try to read as much as possible from the website as you've
<br>> said and I also read the sourceball which leads me to many things like
<br>> rank() and row_number() and they seem to be working yet I don't find
<br>> many of these in the website. (That was my question about Mx.)
<br><br>Well, apparently, we've had time to implement these featured, but not yet to
<br>document them --- we're working on that ...
<br><br>> I am
<br>> trying to use things I should not? At least I would like to understand
<br>> create function as I can then try to make a lag ability.
<br><br>See above. Feel free to post concrete questions in case you run into
<br>problems.
<br><br>Stefan
<br>
<div class='shrinkable-quote'><br>>
<br>> Thank you for you're patience.
<br>> Paul
<br>>
<br>> On 11/23/2009 6:35 PM, Stefan Manegold wrote:
<br>> > Hi Paul,
<br>> >
<br>> > Mx is only used "internally" for MonetDB development; it's not meant for let
<br>> > alone required by users, and hence not distributed with the MonetDB release.
<br>> >
<br>> > All user-relevant documentation (partly indeed extracted from the Mx-based
<br>> > code base using Mx) should be available online on the MonetDB web site at
<br>> > <a href="http://www.monetdb.org/" target="_top" rel="nofollow">http://www.monetdb.org/</a><br>> > aka.
<br>> > <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a><br>> >
<br>> > Stefan
<br>> >
<br>> > ps: In case you are indeed interested in the documentation of Mx itself, you
<br>> > find it at
<br>> > <a href="http://old-www.cwi.nl/themes/ins1/publications/docs/KeScBeBo:UNPUBLISHED:96.pdf" target="_top" rel="nofollow">http://old-www.cwi.nl/themes/ins1/publications/docs/KeScBeBo:UNPUBLISHED:96.pdf</a><br>> >
<br>> > pps: In case you are indeed interested in Mx itself, you find the source in
<br>> > the "buildtools" in the MonetDB CVS repository on SourceForge; cf.,
<br>> > <a href="https://sourceforge.net/projects/monetdb/develop" target="_top" rel="nofollow">https://sourceforge.net/projects/monetdb/develop</a><br>> > <a href="http://monetdb.cvs.sourceforge.net/viewvc/monetdb/buildtools/Mx/" target="_top" rel="nofollow">http://monetdb.cvs.sourceforge.net/viewvc/monetdb/buildtools/Mx/</a><br>> >
<br>> >
<br>> > On Mon, Nov 23, 2009 at 06:06:50PM -0600, Paul Phillips [ gmail ] wrote:
<br>> >> Can someone tell where the Mx utility is for creating the documentation?
<br>> >> I have seen references in sources to this command and searches are
<br>> >> trouble with only two letters like mx.
<br>> >>
<br>> >> Is this documentation different/more than the nl website?
<br>> >>
<br>> >> Thank you.
<br>> >> Paul
<br>> >>
<br>> >> ------------------------------------------------------------------------------
<br>> >> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> >> trial. Simplify your report design, integration and deployment - and focus on
<br>> >> what you do best, core application coding. Discover what's new with
<br>> >> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> >> _______________________________________________
<br>> >> MonetDB-users mailing list
<br>> >> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26498134&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> >> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>> >>
<br>> >
<br>>
<br>> ------------------------------------------------------------------------------
<br>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>> trial. Simplify your report design, integration and deployment - and focus on
<br>> what you do best, core application coding. Discover what's new with
<br>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>> _______________________________________________
<br>> MonetDB-users mailing list
<br>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26498134&i=2" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>
</div><br>--
<br>| Dr. Stefan Manegold | mailto:<a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26498134&i=3" target="_top" rel="nofollow">Stefan.Manegold@...</a> |
<br>| CWI, P.O.Box 94079 | <a href="http://www.cwi.nl/~manegold/" target="_top" rel="nofollow">http://www.cwi.nl/~manegold/</a> |
<br>| 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 |
<br>| The Netherlands | Fax : +31 (20) 592-4312 |
<br><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26498134&i=4" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>tag:old.nabble.com,2006:post-26497720Re: Mx documentation2009-11-24T07:15:43Z2009-11-24T07:15:43Zpphillips99@gmail.comStefan,
<br><br>Thank you again.
<br><br>I would like to use CREATE FUNCTION and have been successful with this
<br><br>create function a (x int, y int)
<br>returns int
<br>return x*y;
<br><br>yet I try to do more complete versions with begin/end like this (and
<br>many similar variations) without success.
<br><br>create function b (x int)
<br>returns int
<br>begin
<br> declare retval int;
<br> select someval into retval from mytbl where id=x;
<br> return retval;
<br>end;
<br><br>I am confused with trying to correctly learn the documentation. I
<br>searched src/tests for examples but they are only easy like my first
<br>that works.
<br><br>Separately, I try to read as much as possible from the website as you've
<br>said and I also read the sourceball which leads me to many things like
<br>rank() and row_number() and they seem to be working yet I don't find
<br>many of these in the website. (That was my question about Mx.) I am
<br>trying to use things I should not? At least I would like to understand
<br>create function as I can then try to make a lag ability.
<br><br><br>Thank you for you're patience.
<br>Paul
<br><br>On 11/23/2009 6:35 PM, Stefan Manegold wrote:
<div class='shrinkable-quote'><br>> Hi Paul,
<br>>
<br>> Mx is only used "internally" for MonetDB development; it's not meant for let
<br>> alone required by users, and hence not distributed with the MonetDB release.
<br>>
<br>> All user-relevant documentation (partly indeed extracted from the Mx-based
<br>> code base using Mx) should be available online on the MonetDB web site at
<br>> <a href="http://www.monetdb.org/" target="_top" rel="nofollow">http://www.monetdb.org/</a><br>> aka.
<br>> <a href="http://monetdb.cwi.nl/" target="_top" rel="nofollow">http://monetdb.cwi.nl/</a><br>>
<br>> Stefan
<br>>
<br>> ps: In case you are indeed interested in the documentation of Mx itself, you
<br>> find it at
<br>> <a href="http://old-www.cwi.nl/themes/ins1/publications/docs/KeScBeBo:UNPUBLISHED:96.pdf" target="_top" rel="nofollow">http://old-www.cwi.nl/themes/ins1/publications/docs/KeScBeBo:UNPUBLISHED:96.pdf</a><br>>
<br>> pps: In case you are indeed interested in Mx itself, you find the source in
<br>> the "buildtools" in the MonetDB CVS repository on SourceForge; cf.,
<br>> <a href="https://sourceforge.net/projects/monetdb/develop" target="_top" rel="nofollow">https://sourceforge.net/projects/monetdb/develop</a><br>> <a href="http://monetdb.cvs.sourceforge.net/viewvc/monetdb/buildtools/Mx/" target="_top" rel="nofollow">http://monetdb.cvs.sourceforge.net/viewvc/monetdb/buildtools/Mx/</a><br>>
<br>>
<br>> On Mon, Nov 23, 2009 at 06:06:50PM -0600, Paul Phillips [ gmail ] wrote:
<br>>> Can someone tell where the Mx utility is for creating the documentation?
<br>>> I have seen references in sources to this command and searches are
<br>>> trouble with only two letters like mx.
<br>>>
<br>>> Is this documentation different/more than the nl website?
<br>>>
<br>>> Thank you.
<br>>> Paul
<br>>>
<br>>> ------------------------------------------------------------------------------
<br>>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>>> trial. Simplify your report design, integration and deployment - and focus on
<br>>> what you do best, core application coding. Discover what's new with
<br>>> Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>>> _______________________________________________
<br>>> MonetDB-users mailing list
<br>>> <a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26497720&i=0" target="_top" rel="nofollow">MonetDB-users@...</a>
<br>>> <a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br>>>
<br>>
</div><br>------------------------------------------------------------------------------
<br>Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
<br>trial. Simplify your report design, integration and deployment - and focus on
<br>what you do best, core application coding. Discover what's new with
<br>Crystal Reports now. <a href="http://p.sf.net/sfu/bobj-july" target="_top" rel="nofollow">http://p.sf.net/sfu/bobj-july</a><br>_______________________________________________
<br>MonetDB-users mailing list
<br><a href="http://old.nabble.com/user/SendEmail.jtp?type=post&post=26497720&i=1" target="_top" rel="nofollow">MonetDB-users@...</a>
<br><a href="https://lists.sourceforge.net/lists/listinfo/monetdb-users" target="_top" rel="nofollow">https://lists.sourceforge.net/lists/listinfo/monetdb-users</a><br><p>From forum: <a href="http://old.nabble.com/monetdb-users-f4367.html" embed="fixTarget[4367]" target="_top" >monetdb-users</a></p>