Nabble - Distcc

PATCH: packaging/RedHat/init.d/distcc

2009-12-16T04:29:43Z

Hi,

A silly patch - in packaging/RedHat/init.d/distcc the executable is
once refereed to using the variable EXEC, and once as with the
explicit expected name.
Change it to always use the EXEC variable.

Tal

$ svn diff
Index: packaging/RedHat/init.d/distcc
===================================================================
--- packaging/RedHat/init.d/distcc (revision 710)
+++ packaging/RedHat/init.d/distcc (working copy)
@@ -103,7 +103,7 @@

RETVAL=0

-[ -x /usr/bin/distccd ] || exit 0
+[ -x $EXEC ] || exit 0

ALLOWFILE=""
if test -f $CONFIGDIR/$DOMAIN/clients.allow; then
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

include_server -v switch

2009-12-14T04:34:30Z

From include_server.1.man, what does "--verify" do?
Could someone explain?

- S
_________________________________________________________________
Hotmail: Trusted email with Microsoft’s powerful SPAM protection.
http://clk.atdmt.com/GBL/go/177141664/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Disable compression

2009-12-14T04:31:46Z

How can compression be disabled?
"192.168.1.20,cpp" in /usr/local/etc/distcc/hosts isn't working.

We have GigE and would like to test build times without it.

- S

_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
http://clk.atdmt.com/GBL/go/177141665/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distccd only compiling locally

2009-12-14T04:01:51Z

>>>
>>> 2009/12/14 STR . :
>>>> distccd[3052] (dcc_nofork_parent) waiting to accept connection
>>>> distccd[3052] (dcc_check_client) connection from 192.168.1.10:49110
>>>> distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
>>>> distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4
>>>
>>> For some reason the client is connecting and then immediately
>>> disconnecting. You probably need to look at the client log to find
>>> out why.
>>>
>>

>
> Maybe your makefile doesn't actually respect $CC?
>

Martin,

Correct, Rules.mk was adding its own TARGET_CC, distcc is being invoked with some errors, but I think I can hack it now.

Thanks!

-S
_________________________________________________________________
Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
http://clk.atdmt.com/GBL/go/171222985/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distccd only compiling locally

2009-12-14T00:53:10Z

> $ pump make -j 20 CC=distcc
> __________Using distcc-pump from /usr/local/bin
> __________Found 2 available distcc servers

Try

pump make -j20 CC="distcc gcc" CXX="distcc g++"

--
Key Fingerprint
4588 F931 A6E7 C7D5 A302 3064 12EB E87F 9B28 A9D2
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distccd only compiling locally

2009-12-13T23:32:19Z

2009/12/14 STR . <strykar@...>:

>
>>
>> 2009/12/14 STR . :
>>> distccd[3052] (dcc_nofork_parent) waiting to accept connection
>>> distccd[3052] (dcc_check_client) connection from 192.168.1.10:49110
>>> distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
>>> distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4
>>
>> For some reason the client is connecting and then immediately
>> disconnecting. You probably need to look at the client log to find
>> out why.
>>
>
>
>
> Martin,
>
> This is odd, the client log isn't writing anything for the make job.
> I can see distcc updating DISTCC_LOG when I do "distcc --show-hosts" or "distcc --version".
>
> But when I do "pump make -j20 CC=distcc" it does not send anything to the remote machine and does all the compilation locally. It also does not update DISTCC_LOG for the make command.

Maybe your makefile doesn't actually respect $CC?

>
>
> $ distcc --version
> distcc 3.1 i686-pc-linux-gnu
> (protocols 1, 2 and 3) (default port 3632)
> built Dec 9 2009 14:07:11
>
> $ echo $DISTCC_LOG
> /tmp/distcc.log
>
> $ echo $DISTCC_VERBOSE
> 1
>
>
> t$ pump make -j10 CC=distcc
> __________Using distcc-pump from /usr/local/bin
> __________Found 3 available distcc servers
> **snipped output***
> Make successful.
> __________Shutting down distcc-pump include server
> Include server timing. Elapsed: 381.1s User: 0.1s System: 0.0s User + System: 0.1s
>
>
>
> # cat /tmp/distcc.log
> distcc[25557] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[25557] (dcc_recursion_safeguard) safeguard level=0
> distcc[25557] (main) compiler name is "distcc"
> distcc[25557] (dcc_parse_hosts_file) load hosts from /home/avid/.distcc/hosts
> distcc[25557] (dcc_parse_hosts) found localhost token "localhost"
> distcc[25557] (dcc_parse_hosts) found tcp token "192.168.1.20"
> distcc[25557] (dcc_parse_hosts) found tcp token "192.168.1.21"
> distcc[25557] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
> distcc[25557] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
> distcc[14721] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[14721] (dcc_recursion_safeguard) safeguard level=0
> distcc[14721] (main) compiler name is "distcc"
> distcc[14721] (dcc_parse_hosts_file) load hosts from /home/avid/.distcc/hosts
> distcc[14721] (dcc_parse_hosts) found localhost token "localhost"
> distcc[14721] (dcc_parse_hosts) found tcp token "192.168.1.20"
> distcc[14721] (dcc_parse_hosts) found tcp token "192.168.1.21"
> distcc[14721] (dcc_exit) exit: code 0; self: 0.000000 user 0.000999 sys; children: 0.000000 user 0.000000 sys
> distcc[14721] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
> distcc[7865] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[7865] (dcc_recursion_safeguard) safeguard level=0
> distcc[7865] (main) compiler name is "distcc"
> distcc[7865] (dcc_get_hostlist) read hosts from environment
> distcc[7865] (dcc_parse_hosts) found tcp token "192.168.1.20"
> distcc[7865] (dcc_parse_hosts) found tcp token "192.168.1.21"
> distcc[7865] (dcc_exit) exit: code 0; self: 0.000999 user 0.000000 sys; children: 0.000000 user 0.000000 sys
> distcc[7865] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
> distcc[7866] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[7866] (dcc_recursion_safeguard) safeguard level=0
> distcc[7866] (main) compiler name is "distcc"
> distcc[7866] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
> distcc[7866] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
> distcc[7870] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[7870] (dcc_recursion_safeguard) safeguard level=0
> distcc[7870] (main) compiler name is "distcc"
> distcc[7870] (dcc_exit) exit: code 0; self: 0.000999 user 0.000000 sys; children: 0.000000 user 0.000000 sys
> distcc[7870] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
> distcc[7871] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
> distcc[7871] (dcc_recursion_safeguard) safeguard level=0
> distcc[7871] (main) compiler name is "distcc"
> distcc[7871] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
> distcc[7871] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
>
>
> distccd was started as:
> distccd -j 8 -P /var/tmp/distcc.pid --user=nobody -a 192.168.1.0/24 --log-file /var/log/distccd.log --log-level=debug --stats --daemon
>
>
> The remote machine's distccd log is the same as before:
> ~# tail -f /var/log/distccd.log
> distccd[23640] (dcc_detach) setsid to session 23640
> distccd[23640] (dcc_log_daemon_started) preforking daemon started (3.1 i686-pc-linux-gnu, built Dec 9 2009 14:07:11)
> distccd[23640] (dcc_create_kids) up to 1 children
> distccd[23640] (dcc_create_kids) up to 2 children
> distccd[23640] (dcc_create_kids) up to 3 children
> distccd[23640] (dcc_create_kids) up to 4 children
> distccd[23640] (dcc_create_kids) up to 5 children
> distccd[23640] (dcc_create_kids) up to 6 children
> distccd[23640] (dcc_create_kids) up to 7 children
> distccd[23640] (dcc_create_kids) up to 8 children
> distccd[23640] (dcc_listen_by_addr) listening on 0.0.0.0:3633
> distccd[23640] (dcc_stats_server) HTTP server started on port 3633
> distccd[23641] (dcc_check_client) connection from 192.168.1.10:56587
> distccd[23642] (dcc_check_client) connection from 192.168.1.10:45563
> distccd[23642] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23641] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23642] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23642] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23641] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23642] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23641] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23641] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23641] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23641] (dcc_job_summary) client: 192.168.1.10:56587 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
> distccd[23642] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23642] (dcc_job_summary) client: 192.168.1.10:45563 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
> distccd[23643] (dcc_check_client) connection from 192.168.1.10:56590
> distccd[23643] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23644] (dcc_check_client) connection from 192.168.1.10:45566
> distccd[23644] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23643] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23644] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23643] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23644] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23643] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23644] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23643] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23643] (dcc_job_summary) client: 192.168.1.10:56590 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
> distccd[23644] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23644] (dcc_job_summary) client: 192.168.1.10:45566 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
> distccd[23646] (dcc_check_client) connection from 192.168.1.10:39635
> distccd[23645] (dcc_check_client) connection from 192.168.1.10:55005
> distccd[23646] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23645] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[23646] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23645] (dcc_readx) ERROR: unexpected eof on fd5
> distccd[23646] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23645] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
> distccd[23646] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23645] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
> distccd[23645] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23645] (dcc_job_summary) client: 192.168.1.10:55005 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
> distccd[23646] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
> distccd[23646] (dcc_job_summary) client: 192.168.1.10:39635 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
>
>
>
> - S
> _________________________________________________________________
> Hotmail: Trusted email with Microsoft’s powerful SPAM protection.
> http://clk.atdmt.com/GBL/go/177141664/direct/01/
>

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: FW: distccd only compiling locally

2009-12-13T22:55:36Z

>
> This is odd, the client log isn't writing anything for the make job.
> I can see distcc updating DISTCC_LOG when I do "distcc --show-hosts" or "distcc --version".
>
> But when I do "pump make -j20 CC=distcc" it does not send anything to the remote machine and does all the compilation locally. It also does not update DISTCC_LOG for the make command.
>
>
> $ distcc --version
> distcc 3.1 i686-pc-linux-gnu
> (protocols 1, 2 and 3) (default port 3632)
> built Dec 9 2009 14:07:11
>
> $ echo $DISTCC_LOG
> /tmp/distcc.log
>
> $ echo $DISTCC_VERBOSE
> 1
>
>
> t$ pump make -j10 CC=distcc
> __________Using distcc-pump from /usr/local/bin
> __________Found 3 available distcc servers
> **snipped output***
> Make successful.
> __________Shutting down distcc-pump include server
> Include server timing. Elapsed: 381.1s User: 0.1s System: 0.0s User + System: 0.1s
>

tcpdump of the session:

~# tcpdump -nnvvXSs 1514 dst tb2 and dst port 3632
tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 1514 bytes
12:14:32.363859 IP (tos 0x0, ttl 64, id 59773, offset 0, flags [DF], proto TCP (6), length 60)
192.168.1.10.40047> 192.168.1.20.3632: Flags [S], cksum 0x8350 (correct), seq 1535381683, win 5840, options [mss 1460,sackOK,TS val 432863694 ecr 0,nop,wscale 6], length 0
0x0000: 4500 003c e97d 4000 4006 cdcf c0a8 010a E..<.}@.@.......
0x0010: c0a8 0114 9c6f 0e30 5b84 10b3 0000 0000 .....o.0[.......
0x0020: a002 16d0 8350 0000 0204 05b4 0402 080a .....P..........
0x0030: 19cc f9ce 0000 0000 0103 0306 ............
12:14:32.363991 IP (tos 0x0, ttl 64, id 59774, offset 0, flags [DF], proto TCP (6), length 52)
192.168.1.10.40047> 192.168.1.20.3632: Flags [.], cksum 0x8015 (correct), seq 1535381684, ack 1847275248, win 92, options [nop,nop,TS val 432863694 ecr 3279939550], length 0
0x0000: 4500 0034 e97e 4000 4006 cdd6 c0a8 010a E..4.~@.@.......
0x0010: c0a8 0114 9c6f 0e30 5b84 10b4 6e1b 2ef0 .....o.0[...n...
0x0020: 8010 005c 8015 0000 0101 080a 19cc f9ce ...\............
0x0030: c37f e7de ....
12:14:32.364329 IP (tos 0x0, ttl 64, id 59775, offset 0, flags [DF], proto TCP (6), length 52)
192.168.1.10.40047> 192.168.1.20.3632: Flags [F.], cksum 0x8014 (correct), seq 1535381684, ack 1847275248, win 92, options [nop,nop,TS val 432863694 ecr 3279939550], length 0
0x0000: 4500 0034 e97f 4000 4006 cdd5 c0a8 010a E..4..@.@.......
0x0010: c0a8 0114 9c6f 0e30 5b84 10b4 6e1b 2ef0 .....o.0[...n...
0x0020: 8011 005c 8014 0000 0101 080a 19cc f9ce ...\............
0x0030: c37f e7de ....
12:14:32.364890 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto TCP (6), length 52)
192.168.1.10.40047> 192.168.1.20.3632: Flags [.], cksum 0x8011 (correct), seq 1535381685, ack 1847275249, win 92, options [nop,nop,TS val 432863695 ecr 3279939551], length 0
0x0000: 4500 0034 0000 4000 4006 b755 c0a8 010a E..4..@.@..U....
0x0010: c0a8 0114 9c6f 0e30 5b84 10b5 6e1b 2ef1 .....o.0[...n...
0x0020: 8010 005c 8011 0000 0101 080a 19cc f9cf ...\............
0x0030: c37f e7df ....

_________________________________________________________________
Hotmail: Free, trusted and rich email service.
http://clk.atdmt.com/GBL/go/171222984/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

FW: distccd only compiling locally

2009-12-13T22:30:32Z

>
> 2009/12/14 STR . :
>> distccd[3052] (dcc_nofork_parent) waiting to accept connection
>> distccd[3052] (dcc_check_client) connection from 192.168.1.10:49110
>> distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
>> distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4
>
> For some reason the client is connecting and then immediately
> disconnecting. You probably need to look at the client log to find
> out why.
>

Resent to mailing list.

Martin,

This is odd, the client log isn't writing anything for the make job.
I can see distcc updating DISTCC_LOG when I do "distcc --show-hosts" or "distcc --version".

But when I do "pump make -j20 CC=distcc" it does not send anything to the remote machine and does all the compilation locally. It also does not update DISTCC_LOG for the make command.

$ distcc --version
distcc 3.1 i686-pc-linux-gnu
(protocols 1, 2 and 3) (default port 3632)
built Dec 9 2009 14:07:11

$ echo $DISTCC_LOG
/tmp/distcc.log

$ echo $DISTCC_VERBOSE
1

t$ pump make -j10 CC=distcc
__________Using distcc-pump from /usr/local/bin
__________Found 3 available distcc servers
**snipped output***
Make successful.
__________Shutting down distcc-pump include server
Include server timing. Elapsed: 381.1s User: 0.1s System: 0.0s User + System: 0.1s

# cat /tmp/distcc.log
distcc[25557] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[25557] (dcc_recursion_safeguard) safeguard level=0
distcc[25557] (main) compiler name is "distcc"
distcc[25557] (dcc_parse_hosts_file) load hosts from /home/avid/.distcc/hosts
distcc[25557] (dcc_parse_hosts) found localhost token "localhost"
distcc[25557] (dcc_parse_hosts) found tcp token "192.168.1.20"
distcc[25557] (dcc_parse_hosts) found tcp token "192.168.1.21"
distcc[25557] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
distcc[25557] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
distcc[14721] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[14721] (dcc_recursion_safeguard) safeguard level=0
distcc[14721] (main) compiler name is "distcc"
distcc[14721] (dcc_parse_hosts_file) load hosts from /home/avid/.distcc/hosts
distcc[14721] (dcc_parse_hosts) found localhost token "localhost"
distcc[14721] (dcc_parse_hosts) found tcp token "192.168.1.20"
distcc[14721] (dcc_parse_hosts) found tcp token "192.168.1.21"
distcc[14721] (dcc_exit) exit: code 0; self: 0.000000 user 0.000999 sys; children: 0.000000 user 0.000000 sys
distcc[14721] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
distcc[7865] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[7865] (dcc_recursion_safeguard) safeguard level=0
distcc[7865] (main) compiler name is "distcc"
distcc[7865] (dcc_get_hostlist) read hosts from environment
distcc[7865] (dcc_parse_hosts) found tcp token "192.168.1.20"
distcc[7865] (dcc_parse_hosts) found tcp token "192.168.1.21"
distcc[7865] (dcc_exit) exit: code 0; self: 0.000999 user 0.000000 sys; children: 0.000000 user 0.000000 sys
distcc[7865] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
distcc[7866] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[7866] (dcc_recursion_safeguard) safeguard level=0
distcc[7866] (main) compiler name is "distcc"
distcc[7866] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
distcc[7866] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
distcc[7870] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[7870] (dcc_recursion_safeguard) safeguard level=0
distcc[7870] (main) compiler name is "distcc"
distcc[7870] (dcc_exit) exit: code 0; self: 0.000999 user 0.000000 sys; children: 0.000000 user 0.000000 sys
distcc[7870] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files
distcc[7871] (dcc_trace_version) distcc 3.1 i686-pc-linux-gnu; built Dec 9 2009 14:07:11
distcc[7871] (dcc_recursion_safeguard) safeguard level=0
distcc[7871] (main) compiler name is "distcc"
distcc[7871] (dcc_exit) exit: code 0; self: 0.000000 user 0.000000 sys; children: 0.000000 user 0.000000 sys
distcc[7871] (dcc_cleanup_tempfiles_inner) deleted 0 temporary files

distccd was started as:
distccd -j 8 -P /var/tmp/distcc.pid --user=nobody -a 192.168.1.0/24 --log-file /var/log/distccd.log --log-evel=debug --stats --daemon

The remote machine's distccd log is the same as before:
~# tail -f /var/log/distccd.log
distccd[23640] (dcc_detach) setsid to session 23640
distccd[23640] (dcc_log_daemon_started) preforking daemon started (3.1 i686-pc-linux-gnu, built Dec 9 2009 4:07:11)
distccd[23640] (dcc_create_kids) up to 1 children
distccd[23640] (dcc_create_kids) up to 2 children
distccd[23640] (dcc_create_kids) up to 3 children
distccd[23640] (dcc_create_kids) up to 4 children
distccd[23640] (dcc_create_kids) up to 5 children
distccd[23640] (dcc_create_kids) up to 6 children
distccd[23640] (dcc_create_kids) up to 7 children
distccd[23640] (dcc_create_kids) up to 8 children
distccd[23640] (dcc_listen_by_addr) listening on 0.0.0.0:3633
distccd[23640] (dcc_stats_server) HTTP server started on port 3633
distccd[23641] (dcc_check_client) connection from 192.168.1.10:56587
distccd[23642] (dcc_check_client) connection from 192.168.1.10:45563
distccd[23642] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23641] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23642] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23642] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23641] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23642] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23641] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23641] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23641] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23641] (dcc_job_summary) client: 192.168.1.10:56587 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[23642] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23642] (dcc_job_summary) client: 192.168.1.10:45563 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[23643] (dcc_check_client) connection from 192.168.1.10:56590
distccd[23643] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23644] (dcc_check_client) connection from 192.168.1.10:45566
distccd[23644] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23643] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23644] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23643] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23644] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23643] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23644] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23643] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23643] (dcc_job_summary) client: 192.168.1.10:56590 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[23644] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23644] (dcc_job_summary) client: 192.168.1.10:45566 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[23646] (dcc_check_client) connection from 192.168.1.10:39635
distccd[23645] (dcc_check_client) connection from 192.168.1.10:55005
distccd[23646] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23645] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[23646] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23645] (dcc_readx) ERROR: unexpected eof on fd5
distccd[23646] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23645] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[23646] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23645] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[23645] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23645] (dcc_job_summary) client: 192.168.1.10:55005 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[23646] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[23646] (dcc_job_summary) client: 192.168.1.10:39635 OTHER exit:0 sig:0 core:0 ret:108 time:0ms

- S
_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
http://clk.atdmt.com/GBL/go/177141665/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: gcc's -x handling is complex

2009-12-13T18:02:25Z

2009/12/3 Hinz, Daniel W <Daniel.Hinz@...>:

> So, I turned on logging and found this line after all of my compile
> invocations:
>
>
>
> distcc[5070] (dcc_scan_args) gcc's -x handling is complex; running locally
>
>
>
> Sure enough, the person who wrote the script that invokes distcc (pump mode)
> does include –x on the command line. The code base is a mixture of c and
> c++, the with the switch set appropriately.
>
>
>
> Are these two pieces of information related and what should I learn from it?

Yes, because of the way your makefile uses -x, this job can't be
distributed. You can either change the script to avoid this, or
change distcc to understand it, or live with it.

>
>
>
> Also, I have not found a definition of ‘Blocked’ in any of the documentation
> I’ve found so far. What does that indicate?

Waiting for a place to run the job.

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distccd only compiling locally

2009-12-13T17:58:16Z

2009/12/14 STR . <strykar@...>:
> distccd[3052] (dcc_nofork_parent) waiting to accept connection
> distccd[3052] (dcc_check_client) connection from 192.168.1.10:49110
> distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
> distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4

For some reason the client is connecting and then immediately
disconnecting. You probably need to look at the client log to find
out why.

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

distccd only compiling locally

2009-12-13T14:26:26Z

Hi,

I'm having an issue similar to http://www.mail-archive.com/distcc@.../msg03593.html
In my case distccd is doing all the compiling locally and none on the remote machine.

To test, I'd setup distccd on two machines, A and B, having 2 x dual-core core Opterons and 4 gigs of ECC RAM.
They're running Slackware Linux 13, stock kernel 2.6.29.6-smp i686.
These machines don't do anything else and have dual 1000M ethernet links.

$ ls -al /usr/bin/gcc
lrwxrwxrwx 1 root root 9 2009-09-18 03:06 /usr/bin/gcc -> gcc-4.3.3*

$ gcc -v
Reading specs from /usr/lib/gcc/i486-slackware-linux/4.3.3/specs
Target: i486-slackware-linux
Configured with: ../gcc-4.3.3/configure --prefix=/usr --libdir=/usr/lib --enable-shared --enable-bootstrap --enable-languages=ada,c,c++,fortran,java,objc --enable-threads=posix --enable-checking=release --with-system-zlib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-libssp --with-gnu-ld --verbose --with-arch=i486 --target=i486-slackware-linux --build=i486-slackware-linux --host=i486-slackware-linux
Thread model: posix
gcc version 4.3.3 (GCC)

USER_TIME_QUOTA in /usr/local/lib/python2.6/site-packages/include_server/basics.py = 20 seconds

///
Machine A (Localhost)

DISTCC_POTENTIAL_HOSTS = Machine B's IPs
INCLUDE_SERVER_ARGS = --time --statistics --stat_reset_triggers="*"

# distccd --wizard --user nobody -a 192.168.1.0/24
distccd[11881] (dcc_discard_root) discarded root privileges, changed to uid=99 gid=99
distccd[11881] (main) chdir to /tmp
distccd[11881] (dcc_setup_daemon_path) daemon's PATH is /usr/local/sbin:/usr/sbin:/sbin:/usr/local/bin:/usr/bin:/bin:/usr/games:/usr/lib/java/bin
distccd[11881] (dcc_listen_by_addr) listening on 0.0.0.0:3632
distccd[11881] (dcc_defer_accept) TCP_DEFER_ACCEPT turned on
distccd[11881] (dcc_standalone_server) 4 CPUs online on this server
distccd[11881] (dcc_standalone_server) allowing up to 6 active jobs
distccd[11881] (dcc_standalone_server) not detaching
distccd[11881] (dcc_new_pgrp) already a process group leader
distccd[11881] (dcc_log_daemon_started) non-forking daemon started (3.1 i686-pc-linux-gnu, built Dec 9 2009 14:07:11)
distccd[11881] (dcc_nofork_parent) waiting to accept connection

$ pump make -j 20 CC=distcc
__________Using distcc-pump from /usr/local/bin
__________Found 2 available distcc servers

***snipped output
Make successful.

__________Shutting down distcc-pump include server
Include server timing. Elapsed: 378.9s User: 0.1s System: 0.0s User + System: 0.1s
///

///
Machine B (Remote host)

DISTCC_POTENTIAL_HOSTS = Machine A's IPs + 127.0.0.1
INCLUDE_SERVER_ARGS = --time --statistics --stat_reset_triggers="*"

# distccd --wizard --user nobody -a 192.168.1.0/24
distccd[3052] (dcc_discard_root) discarded root privileges, changed to uid=99 gid=99
distccd[3052] (main) chdir to /tmp
distccd[3052] (dcc_setup_daemon_path) daemon's PATH is /usr/local/sbin:/usr/sbin:/sbin:/usr/local/bin:/usr/bin:/bin:/usr/games:/usr/lib/java/bin
distccd[3052] (dcc_listen_by_addr) listening on 0.0.0.0:3632
distccd[3052] (dcc_defer_accept) TCP_DEFER_ACCEPT turned on
distccd[3052] (dcc_standalone_server) 4 CPUs online on this server
distccd[3052] (dcc_standalone_server) allowing up to 6 active jobs
distccd[3052] (dcc_standalone_server) not detaching
distccd[3052] (dcc_new_pgrp) already a process group leader
distccd[3052] (dcc_log_daemon_started) non-forking daemon started (3.1 i686-pc-linux-gnu, built Dec 9 2009 14:07:11)
distccd[3052] (dcc_nofork_parent) waiting to accept connection
distccd[3052] (dcc_check_client) connection from 192.168.1.10:49110
distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4
distccd[3052] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[3052] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[3052] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[3052] (dcc_job_summary) client: 192.168.1.10:49110 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[3052] (dcc_nofork_parent) waiting to accept connection
distccd[3052] (dcc_check_client) connection from 192.168.1.10:38121
distccd[3052] (dcc_check_address) match client 0xa01a8c0, value 0x1a8c0, mask 0xffffff
distccd[3052] (dcc_readx) ERROR: unexpected eof on fd4
distccd[3052] (dcc_r_token_int) ERROR: read failed while waiting for token "DIST"
distccd[3052] (dcc_r_request_header) ERROR: client did not provide distcc magic fairy dust
distccd[3052] (dcc_cleanup_tempfiles_inner) deleted 3 temporary files
distccd[3052] (dcc_job_summary) client: 192.168.1.10:38121 OTHER exit:0 sig:0 core:0 ret:108 time:0ms
distccd[3052] (dcc_nofork_parent) waiting to accept connection
///

The jobs are run locally, and the compilation was successful, it's just not distributed.
I don't see any jobs on the remote machine and the CPUs are idle.
What's up with that?

-S
_________________________________________________________________
Hotmail: Free, trusted and rich email service.
http://clk.atdmt.com/GBL/go/171222984/direct/01/
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

gcc's -x handling is complex

2009-12-02T10:32:55Z

I’m new to distcc and have not gotten things to work, yet. Running distccmon-text I see:

31028 Blocked [0]

31033 Blocked [0]

31037 Blocked [0]

31050 Blocked [0]

31285 Compile localhost[0]

31290 Compile localhost[1]

31292 Compile localhost[2]

31336 Compile localhost[3]

So, I turned on logging and found this line after all of my compile invocations:

distcc[5070] (dcc_scan_args) gcc's -x handling is complex; running locally

Sure enough, the person who wrote the script that invokes distcc (pump mode) does include –x on the command line. The code base is a mixture of c and c++, the with the switch set appropriately.

Are these two pieces of information related and what should I learn from it?

Also, I have not found a definition of ‘Blocked’ in any of the documentation I’ve found so far. What does that indicate?

Thanks,

Dan Hinz

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: Hi, I report one error information DISTCC made in CYGWIN and used to compile in windriver platform

2009-12-01T18:52:21Z

Ok, thanks for you kind reply. I have already solved this problem. Thanks anyway.

Best regards

Kevin

From: ext Fergus Henderson [mailto:fergus@...]
Sent: Wednesday, December 02, 2009 7:27 AM
To: Zhang, Kevin W. (NSN - CN/Hangzhou)
Cc: distcc@...
Subject: Re: [distcc] Hi, I report one error information DISTCC made in CYGWIN and used to compile in windriver platform

Try setting DISTCC_VERBOSE=1... that should give some more clues.

On Wed, Sep 16, 2009 at 6:03 AM, Zhang, Kevin W. (NSN - CN/Hangzhou) <kevin.w.zhang@...> wrote:

Hi, dear distcc experts:

How are you? dear all. Sorry to trouble you. I have one question to report to you and pls give me some advice about this. I'm now using distcc to try to build my source codes in windriver platform. My operating system is WINDOWS 2003 server. I use CYGWIN 1.7beta to make distcc source code. Distcc version is 3.1(both client and server). Its downloading page is here:

http://code.google.com/p/distcc/downloads/list
Name is distcc-3.1.tar.bz2

Following the distcc installation guide, firstly I use make.exe in cygwin to create distcc.exe, distccd.exe and distccmon-text.exe etc. executable files. Now I am working in windriver vxworks-6.7 version environment. To set client environment in WINDOWS system.

DISTCC_HOSTS=127.0.0.1
DISTCC_DIR=c:\apps\distcc\distcc_dir
TMPDIR=c:\apps\distcc\distcc_tmp_dir

run distccd in the same machine
set DISTCC_BIN=C:\APPS\distcc
set DISTCC_DIR=C:\APPS\distcc\distcc_dir
set TMPDIR=%DISTCC_BIN%\distcc_tmp_dir
distccd --allow 10.0.0.1/8 --port 3699 --daemon --no-detach --log-file=%DISTCC_BIN%/distccd_vx67.log --nice 6 --verbose

Then in client I input this command:
make CPU=SIMNU TOOL=gnu CC="distcc cc$(TOOLENV)" -j2

note: CPU, TOOL and CC is MARCO of vxworks-6.7. It has the nearly same effect as
make -j2 CC=distcc

So error information popups like this following, which tells me ccpentium.exe compiler(a transplanting compiler of gcc in windriver platform) no input file. The precompilation process failed to pass. Don't mention sending preprocess results to remote machine. I think it's possibly caused by these two lines

distcc ccpentium -mtune=i486 -march=i486 -ansi -O2 -nostdlib -fno-builtin -fno-defer-pop -Wall -IC:/APPS/WindRiver/vxworks-6.7/target/h -IC:/APPS/WindRiver/vxworks-6.7/target/h/wrn/coreip   -DCPU=SIMNT -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL -DINET -D_WRS_LIB_BUILD -I.   -D_VSB_CONFIG_FILE=\"D:/Workspace/kevin/vsb_test/h/config/vsbConfig.h\"   -fno-zero-initialized-in-bss   -MD -c -o D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o aimMmuShow.c

I don't know what's the reason for? I have a look at FAQ of distcc. I notice -MD may not work rightly by compiler gcc 3.0 or later. Detailed information is as

http://distcc.samba.org/faq.html
Problems with gcc -MD
So I don't know if it's the root cause. Because it's difficult to modify windriver makefile to replace -MD with -MF, I can't verify if my idear is right. The gcc version of Windriver is using is 4.1.2. So it's later of 3.0.

Another reason I doubt the command line length is a little long. So distcc only intercept front part of options and .c file name at the end is lost. So it tells me no input file.

So what's the definite reason for this strange thing. I remove distcc prefix, then everything is OK. Compilation is no problem.

Build Started in Project 'vsb_test':   2009-09-16 18:24:28
Platform: Wind River VxWorks 6.7
Command: make "CC=distcc cc$(TOOLENV)"
Working Directory: D:/Workspace/kevin/vsb_test
cd C:/APPS/WindRiver/vxworks-6.7/target/src && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _VSB_CONFIG_ADDEDCFLAGS= _WRS_CONFIG_APP_TOOL=gnu VSB_DIR=D:/Workspace/kevin/vsb_test BSPNAME=simpc _WRS_CONFIG_FP=hard   ADDED_CFLAGS+=-I. DUAL_BUILD=false

make.exe[1]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
cd arch && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[2]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
cd simnt && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[3]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objarch
updating D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/libarch.a
cd math && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[4]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt/math'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objarch
updating D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/libarch.a
make.exe[4]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt/math'
make.exe[3]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt'
cd doc && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[3]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/doc'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
make.exe[3]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/doc'
make.exe[2]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch'
cd aim && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[2]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/aim'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim
distcc ccpentium -mtune=i486 -march=i486 -ansi -O2 -nostdlib -fno-builtin -fno-defer-pop -Wall -IC:/APPS/WindRiver/vxworks-6.7/target/h -IC:/APPS/WindRiver/vxworks-6.7/target/h/wrn/coreip   -DCPU=SIMNT -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL -DINET -D_WRS_LIB_BUILD -I.   -D_VSB_CONFIG_FILE=\"D:/Workspace/kevin/vsb_test/h/config/vsbConfig.h\"   -fno-zero-initialized-in-bss   -MD -c -o D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o aimMmuShow.c

ccpentium.exe: no input files
distcc[668] ERROR: compile (null) on localhost failed
make.exe[2]: *** [D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o] Error 1
make.exe[2]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/aim'
make.exe[1]: *** [aim] Error 2
make.exe[1]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src'
C:\APPS\WindRiver\utilities-1.0\x86-win32\bin\make.exe: *** [_target_gnuSIMNT_fooC_/APPS/WindRiver/vxworks-6.7/target/src] Error 2

Build Failed in Project 'vsb_test' (Process Exit Value was 2):   2009-09-16 18:24:29   (Elapsed Time: 00:01)

Best regards
Kevin in NSN

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc failed to parse "-x c" and "-x c++" compiler flags and to pass -nostdinc to distcc server

2009-12-01T16:57:39Z

Martin Pool wrote:

If I pass -x c to
  distcc.exe arm-none-symbianelf-g++.exe ... -x c ...
I get
arm-none-symbianelf-g++.exe: warning: `-x c' after last input file has no
effect
arm-none-symbianelf-g++.exe: no input files


What's in the ...?  If you are putting -x c after the source files,
the error is reasonable.  If you're doing it correctly and distcc's
messing up the order, that would be a bug.

-x c is before source file, actually the source file is last.

With -x c++ there is no warning, but I am not sure if it handle that.

Problem 2:
There is compiler flag -nostdinc which does not cause problems to distcc.exe
calling preprocessor, but this option is not passed to distccd server. It
seems that distcc.exe does not recognize it, so it ignores this.


-nostdinc only affects searching for include files, therefore it'd be
appropriate not to pass it to the server?

Even in non-pump mode, compiler on server side will faile to build without this flag.

I solved the problem with wrapper on client and on server side, but that's not nice.

Tihomir

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc failed to parse "-x c" and "-x c++" compiler flags and to pass -nostdinc to distcc server

2009-12-01T16:26:20Z

2009/12/2 Martin Pool <mbp@...>:
>> Problem 2:
>> There is compiler flag -nostdinc which does not cause problems to distcc.exe
>> calling preprocessor, but this option is not passed to distccd server. It
>> seems that distcc.exe does not recognize it, so it ignores this.
>
> -nostdinc only affects searching for include files, therefore it'd be
> appropriate not to pass it to the server?

Correction - I was thinking of old distcc. In pump mode, I suppose it
should be passed, and the client may need to be updated to reflect
this.

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc failed to parse "-x c" and "-x c++" compiler flags and to pass -nostdinc to distcc server

2009-12-01T16:25:49Z

2009/10/28 Tihomir Heidelberg - 9a4gl <9a4gl@...>:

> Hi,
>
> having problem with gcce compiler from CSL Arm Toolchain using distcc,
> command arguments parsing problem, description bellow...
>
> According to distcc FAQ:
>
> Compiler command line parsing problems?
>
> distcc parses the compiler command line to work out what operation is being
> invoked, and what are the input and output files. The semantics of gcc
> command lines is fairly complex, and has changed slightly over time. There
> may be some valid command lines that distcc understands differently to gcc,
> though none are known at the moment. These ought to cause the command to be
> run locally, but it is possible that it would cause a failure. Either case
> should be reported as a bug.
>
> Using:
> - Cygwin
> - distcc 2.18.3 i686-pc-cygwin (protocols 1 and 2) (default port 3632)
> built Apr 25 2005 11:10:45
> Copyright (C) 2002, 2003, 2004 by Martin Pool.
> Includes miniLZO (C) 1996-2002 by Markus Franz Xaver Johannes Oberhumer.
> As I see there is no newer version for cygwin, and on
> http://distcc.samba.org/download.html the same is available as source
>
> Using: CSL Arm Toolchain with gcce compiler for ARM, building binaries for
> Symbian OS
> arm-none-symbianelf-g++.exe (GCC) 3.4.3 (release) (CodeSourcery ARM Q1C
> 2005)
> Copyright (C) 2004 Free Software Foundation, Inc.
> This is free software; see the source for copying conditions. There is NO
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
>
> Problem 1:
> This compiler has option to tell which compiler to use, c or cpp, option is:
> -x <language>            Specify the language of the following input files
>                            Permissible languages include: c c++ assembler
> none
>                            'none' means revert to the default behavior of
>                            guessing the language based on the file's
> extension
> If I pass -x c to
> distcc.exe arm-none-symbianelf-g++.exe ... -x c ...
> I get
> arm-none-symbianelf-g++.exe: warning: `-x c' after last input file has no
> effect
> arm-none-symbianelf-g++.exe: no input files

What's in the ...? If you are putting -x c after the source files,
the error is reasonable. If you're doing it correctly and distcc's
messing up the order, that would be a bug.

>
> With -x c++ there is no warning, but I am not sure if it handle that.
>
> Problem 2:
> There is compiler flag -nostdinc which does not cause problems to distcc.exe
> calling preprocessor, but this option is not passed to distccd server. It
> seems that distcc.exe does not recognize it, so it ignores this.

-nostdinc only affects searching for include files, therefore it'd be
appropriate not to pass it to the server?

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc for ada

2009-12-01T16:22:32Z

2009/12/2 Martin Pool <mbp@...>:
> I thought you were scanning the moderation queue so I only just did it
> this morning - I'll check that you're a moderator.

I had not made you a moderator - sorry.

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc for ada

2009-12-01T16:19:12Z

2009/12/2 Fergus Henderson <fergus@...>:
> For the record... email propagation on the distcc@... mail list
> appears to have been delayed by about three months.
> Luckily Tim noticed this and instead posted his patch to
> distcc-patches@..., where I reviewed it, and gave him some
> detailed feedback.
> He has recently posted another updated version of his patch, and I hope to
> soon find the time to review that too.

hi,

I thought you were scanning the moderation queue so I only just did it
this morning - I'll check that you're a moderator.

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc for ada

2009-12-01T15:30:44Z

For the record... email propagation on the distcc@... mail list appears to have been delayed by about three months.
Luckily Tim noticed this and instead posted his patch to distcc-patches@..., where I reviewed it, and gave him some detailed feedback.
He has recently posted another updated version of his patch, and I hope to soon find the time to review that too.

P.S. Tim, this is cool stuff! Great work!

On Fri, Oct 9, 2009 at 12:44 PM, Tim Dossett <timothy.dossett@...> wrote:

Hello all,
I have modified distcc 3.1 to distribute ada compilation as well as c
and c++, and I'd like to contribute the modifications back to the
project. Patch is attached. Please consider this for inclusion in the distcc baseline.

Thanks,
Tim
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: Hi, I report one error information DISTCC made in CYGWIN and used to compile in windriver platform

2009-12-01T15:27:14Z

Try setting DISTCC_VERBOSE=1... that should give some more clues.

On Wed, Sep 16, 2009 at 6:03 AM, Zhang, Kevin W. (NSN - CN/Hangzhou) <kevin.w.zhang@...> wrote:

Hi, dear distcc experts:

How are you? dear all. Sorry to trouble you. I have one question to report to you and pls give me some advice about this. I'm now using distcc to try to build my source codes in windriver platform. My operating system is WINDOWS 2003 server. I use CYGWIN 1.7beta to make distcc source code. Distcc version is 3.1(both client and server). Its downloading page is here:

http://code.google.com/p/distcc/downloads/list
Name is distcc-3.1.tar.bz2

Following the distcc installation guide, firstly I use make.exe in cygwin to create distcc.exe, distccd.exe and distccmon-text.exe etc. executable files. Now I am working in windriver vxworks-6.7 version environment. To set client environment in WINDOWS system.

DISTCC_HOSTS=127.0.0.1
DISTCC_DIR=c:\apps\distcc\distcc_dir
TMPDIR=c:\apps\distcc\distcc_tmp_dir

run distccd in the same machine
set DISTCC_BIN=C:\APPS\distcc
set DISTCC_DIR=C:\APPS\distcc\distcc_dir
set TMPDIR=%DISTCC_BIN%\distcc_tmp_dir
distccd --allow 10.0.0.1/8 --port 3699 --daemon --no-detach --log-file=%DISTCC_BIN%/distccd_vx67.log --nice 6 --verbose

Then in client I input this command:
make CPU=SIMNU TOOL=gnu CC="distcc cc$(TOOLENV)" -j2

note: CPU, TOOL and CC is MARCO of vxworks-6.7. It has the nearly same effect as
make -j2 CC=distcc

So error information popups like this following, which tells me ccpentium.exe compiler(a transplanting compiler of gcc in windriver platform) no input file. The precompilation process failed to pass. Don't mention sending preprocess results to remote machine. I think it's possibly caused by these two lines

distcc ccpentium -mtune=i486 -march=i486 -ansi -O2 -nostdlib -fno-builtin -fno-defer-pop -Wall -IC:/APPS/WindRiver/vxworks-6.7/target/h -IC:/APPS/WindRiver/vxworks-6.7/target/h/wrn/coreip   -DCPU=SIMNT -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL -DINET -D_WRS_LIB_BUILD -I.   -D_VSB_CONFIG_FILE=\"D:/Workspace/kevin/vsb_test/h/config/vsbConfig.h\"   -fno-zero-initialized-in-bss   -MD -c -o D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o aimMmuShow.c

I don't know what's the reason for? I have a look at FAQ of distcc. I notice -MD may not work rightly by compiler gcc 3.0 or later. Detailed information is as

http://distcc.samba.org/faq.html
Problems with gcc -MD
So I don't know if it's the root cause. Because it's difficult to modify windriver makefile to replace -MD with -MF, I can't verify if my idear is right. The gcc version of Windriver is using is 4.1.2. So it's later of 3.0.

Another reason I doubt the command line length is a little long. So distcc only intercept front part of options and .c file name at the end is lost. So it tells me no input file.

So what's the definite reason for this strange thing. I remove distcc prefix, then everything is OK. Compilation is no problem.

Build Started in Project 'vsb_test':   2009-09-16 18:24:28
Platform: Wind River VxWorks 6.7
Command: make "CC=distcc cc$(TOOLENV)"
Working Directory: D:/Workspace/kevin/vsb_test
cd C:/APPS/WindRiver/vxworks-6.7/target/src && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _VSB_CONFIG_ADDEDCFLAGS= _WRS_CONFIG_APP_TOOL=gnu VSB_DIR=D:/Workspace/kevin/vsb_test BSPNAME=simpc _WRS_CONFIG_FP=hard   ADDED_CFLAGS+=-I. DUAL_BUILD=false

make.exe[1]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
cd arch && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[2]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
cd simnt && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[3]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objarch
updating D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/libarch.a
cd math && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[4]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt/math'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objarch
updating D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/libarch.a
make.exe[4]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt/math'
make.exe[3]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/simnt'
cd doc && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[3]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/doc'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/obj
make.exe[3]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch/doc'
make.exe[2]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/arch'
cd aim && C:/APPS/WindRiver/utilities-1.0/x86-win32/bin/make.exe CPU=SIMNT TOOL=gnu _WRS_CONFIG_FP=hard
make.exe[2]: Entering directory `C:/APPS/WindRiver/vxworks-6.7/target/src/aim'
mkdir -p D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim
distcc ccpentium -mtune=i486 -march=i486 -ansi -O2 -nostdlib -fno-builtin -fno-defer-pop -Wall -IC:/APPS/WindRiver/vxworks-6.7/target/h -IC:/APPS/WindRiver/vxworks-6.7/target/h/wrn/coreip   -DCPU=SIMNT -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL -DINET -D_WRS_LIB_BUILD -I.   -D_VSB_CONFIG_FILE=\"D:/Workspace/kevin/vsb_test/h/config/vsbConfig.h\"   -fno-zero-initialized-in-bss   -MD -c -o D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o aimMmuShow.c

ccpentium.exe: no input files
distcc[668] ERROR: compile (null) on localhost failed
make.exe[2]: *** [D:/Workspace/kevin/vsb_test/simpc/SIMNT/common/objaim/aimMmuShow.o] Error 1
make.exe[2]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src/aim'
make.exe[1]: *** [aim] Error 2
make.exe[1]: Leaving directory `C:/APPS/WindRiver/vxworks-6.7/target/src'
C:\APPS\WindRiver\utilities-1.0\x86-win32\bin\make.exe: *** [_target_gnuSIMNT_fooC_/APPS/WindRiver/vxworks-6.7/target/src] Error 2

Build Failed in Project 'vsb_test' (Process Exit Value was 2):   2009-09-16 18:24:29   (Elapsed Time: 00:01)

Best regards
Kevin in NSN

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: How to use distcc in a project using more than one cross-compiler?

2009-12-01T15:22:26Z

On Mon, Aug 3, 2009 at 5:21 AM, Caixia Jiang <cjiang@...> wrote:

Hi all,

I want to use distcc to compile a project in such a situation:

This project uses three kinds of cross-compilers. And the cross-compiler paths are set absolutely in the config files, which invalidates the command line setting ‘CC=distcc CROSS-COMPILER-PATH”’.

I want to know that how I can use distcc in the above situation? Look forward to your reply.

Try using "distcc <CROSS-COMPILER-PATH>" rather than "<CROSS-COMPILER-PATH>" in your config files.
Or if that doesn't work, use "masquerading" - see the distcc man page section on masquerading.

Thanks && Best wishes,

Caixia

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: Generating build for multi platforms

2009-12-01T15:19:09Z

On Wed, Oct 21, 2009 at 4:20 AM, Eliakim Cohen <Eliakim@...> wrote:

Hi,

Our build is very long (12 hours) for verey platform (Sun Solaris, HP UX) etc.

We would like to implement DISTCC to speed up the build.

We build number of platforms every night (Sun Solaris, HP UX, Linux Etc).

Since DISTCC can utilize an agent platform to compile for the master platform, I was wondering whether there is a way to reuse the DISTCC/gcc intermediate compile output when compiling for the second platform. E.g, if I used DISTCC/gcc to compile for HP UX, can I use whatever was created by DISTCC/gcc for HP UX in order to start the Sun Solaris compilation from a better point than from compiling from scratch for the Sun Solaris?

Generally it is NOT possible to reuse even the preprocessor output, because conditional compilation (#if) and includes of system header files mean that it will in general be different on different platforms.

You _can_ however reuse the distcc pump include server cache... just arrange things so that you have a single command that will run (in parallel) all the different compilations for all the different platforms, e.g. using a Makefile with a target all_platforms, and then run that single command under "pump", e.g. "pump make all_platforms". This avoids the need to scan the includes of each source file more than once. The main restriction is that your build process should not overwrite any source files during the build, so if you need to run "configure" to generate a config.h which is different for each target platform, then you should do so in a different directory for each platform.

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distcc failed to parse "-x c" and "-x c++" compiler flags and to pass -nostdinc to distcc server

2009-12-01T15:10:09Z

2009/10/27 Tihomir Heidelberg - 9a4gl <9a4gl@...>

Hi,

having problem with gcce compiler from CSL Arm Toolchain using distcc, command arguments parsing problem, description bellow...

According to distcc FAQ:

Compiler command line parsing problems?

distcc parses the compiler command line to work out what operation is being invoked, and what are the input and output files. The semantics of gcc command lines is fairly complex, and has changed slightly over time. There may be some valid command lines that distcc understands differently to gcc, though none are known at the moment. These ought to cause the command to be run locally, but it is possible that it would cause a failure. Either case should be reported as a bug.

Using:
- Cygwin
- distcc 2.18.3 i686-pc-cygwin (protocols 1 and 2) (default port 3632) built Apr 25 2005 11:10:45 Copyright (C) 2002, 2003, 2004 by Martin Pool. Includes miniLZO (C) 1996-2002 by Markus Franz Xaver Johannes Oberhumer.
As I see there is no newer version for cygwin, and on http://distcc.samba.org/download.html the same is available as source

The distcc site has moved to distcc.org.
The latest version is 3.1 and it should work for Cygwin. You may need to build from the source distribution.

Using: CSL Arm Toolchain with gcce compiler for ARM, building binaries for Symbian OS
arm-none-symbianelf-g++.exe (GCC) 3.4.3 (release) (CodeSourcery ARM Q1C 2005) Copyright (C) 2004 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Problem 1:
This compiler has option to tell which compiler to use, c or cpp, option is:
-x <language>            Specify the language of the following input files
                           Permissible languages include: c c++ assembler none
                           'none' means revert to the default behavior of
                           guessing the language based on the file's extension
If I pass -x c to
distcc.exe arm-none-symbianelf-g++.exe ... -x c ... I get arm-none-symbianelf-g++.exe: warning: `-x c' after last input file has no effect arm-none-symbianelf-g++.exe: no input files
With -x c++ there is no warning, but I am not sure if it handle that.

Problem 2:
There is compiler flag -nostdinc which does not cause problems to distcc.exe calling preprocessor, but this option is not passed to distccd server. It seems that distcc.exe does not recognize it, so it ignores this.

distcc is open-source... patches welcome!

Cheers,
Fergus.

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: distributed compiles fail due to failed cleanup (?)

2009-12-01T10:18:55Z

My guess is that this may be a portability issue relating to deletion of symlinks?
It looks like distcc is trying to delete a symlink that points to a directory which is not owned by you, but for some reason this is trying to delete the symlink target and thus getting an error. I suggest you go through the distcc deletion code looking for portability issues where Solaris and Linux might have different behavior related to this.

Posted from an Android phone.

On Dec 1, 2009 11:22 AM, "Tim Flechtner" <timf@...> wrote:

hi everyone,
i'm trying to get distcc-3.1 up and running, and my distributed compiles are failing due to (i think) distccd failing to clean up the temporary directories. this is all taking place in a solaris x86 environment.

i started distccd as user timf on a host named udesktop71 (amongst others) using:
/opt/app/nonc++/distcc-3.1/bin/distccd --daemon --allow 10.100.0.0/16 --log-level=debug

i invoked distcc on a remote host (also as me) using:
DISTCC_POTENTIAL_HOSTS="udesktop71" MAKEFLAGS=-j8 pump tlsupermake -u -c

(tlsupermake is an in-house tool which recursively builds our libraries, invoking gnu make)

on udesktop71 i see in the syslog (in /var/adm/messages):
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test/libs/price failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test/libs failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home/timf/builds failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home/timf failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice] (dcc_job_summary) client: 10.100.114.11:33841 COMPILE_ERROR exit:1 sig:0 core:0 ret:0 time:1677ms g++ Price.C

(which is a nice staircase effect by the way)

i can't figure out why distccd wouldn't be able to clean up those directories and files. i'm also not sure if they are the cause of the compile failure. can anyone help? is there some other log file i can look in to get more information about what is going on?

thanks!
-tim
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

distributed compiles fail due to failed cleanup (?)

2009-12-01T08:21:57Z

hi everyone,
i'm trying to get distcc-3.1 up and running, and my distributed
compiles are failing due to (i think) distccd failing to clean up the
temporary directories. this is all taking place in a solaris x86
environment.

i started distccd as user timf on a host named udesktop71 (amongst
others) using:
/opt/app/nonc++/distcc-3.1/bin/distccd --daemon --allow 10.100.0.0/16
--log-level=debug

i invoked distcc on a remote host (also as me) using:
DISTCC_POTENTIAL_HOSTS="udesktop71" MAKEFLAGS=-j8 pump tlsupermake -u -c

(tlsupermake is an in-house tool which recursively builds our libraries,
invoking gnu make)

on udesktop71 i see in the syslog (in /var/adm/messages):
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup
/tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test/libs/price
failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup
/tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test/libs failed: Not
owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup
/tmp/distccd_FaaqzZ/export/home/timf/builds/distcc_test failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup
/tmp/distccd_FaaqzZ/export/home/timf/builds failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup
/tmp/distccd_FaaqzZ/export/home/timf failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export/home
failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ/export failed:
Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_cleanup_tempfiles_inner) cleanup /tmp/distccd_FaaqzZ failed: Not owner
Dec 1 10:14:48 udesktop71 distccd[26314]: [ID 702911 daemon.notice]
(dcc_job_summary) client: 10.100.114.11:33841 COMPILE_ERROR exit:1 sig:0
core:0 ret:0 time:1677ms g++ Price.C

(which is a nice staircase effect by the way)

i can't figure out why distccd wouldn't be able to clean up those
directories and files. i'm also not sure if they are the cause of the
compile failure. can anyone help? is there some other log file i can
look in to get more information about what is going on?

thanks!
-tim
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: pump problem on solaris 10, x86 [solved]

2009-11-30T14:25:39Z

that also worked. i suspect you are right about it being a sh/bash
incompatibility.

thanks!
-tim

Fergus Henderson wrote:

>
> I haven't seen anything like this one before.
> But maybe this is due to some incompatibility between solaris sh and bash?
> The ^J (control-J) in the middle there is the immediate cause of the
> problem.
> Try running 'bash pump ...' instead of 'pump ...'.
>
> On Tue, Nov 24, 2009 at 6:17 PM, Tim Flechtner <timf@...
> <mailto:timf@...>> wrote:
>
> hi everyone,
> i'm trying to get distcc-3.1 running on solaris x86. i am
> getting the following error, can anyone help me out? i tried
> searching through google, and read the first four quarters of
> archived list emails but didn't see anything either place that
> looked familiar. is there a better way to search?
>
> thanks!
> -tim
>
> userver43:/export/home/timf/builds/distcc_test/libs/meta $
> DISTCC_POTENTIAL_HOSTS="udesktop71 udesktop103" pump gmake
> __________Using distcc-pump from /opt/app/nonc++/distcc-3.1/bin
> /opt/app/nonc++/distcc-3.1/bin/pump: DISTCC_HOSTS=--randomize
> udesktop103,cpp,lzo^Judesktop71,cpp,lzo: is not an identifier
> __________Shutting down distcc-pump include server
> userver43:/export/home/timf/builds/distcc_test/libs/meta $
>
> __
> distcc mailing list http://distcc.samba.org/
> To unsubscribe or change options:
> https://lists.samba.org/mailman/listinfo/distcc
>
>
>
> --
> Fergus Henderson <fergus@... <mailto:fergus@...>>
>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Any x86_64 rpms somewhere?

2009-11-29T20:39:55Z

Hello,

Are there any 64 bit rpms available somewhere?

I wasn't available to build them from the tarball found here:
http://code.google.com/p/distcc/downloads/list

Regards,
Francois.
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: pump problem on solaris 10, x86

2009-11-25T07:38:13Z

I haven't seen anything like this one before.
But maybe this is due to some incompatibility between solaris sh and bash?
The ^J (control-J) in the middle there is the immediate cause of the problem.
Try running 'bash pump ...' instead of 'pump ...'.

On Tue, Nov 24, 2009 at 6:17 PM, Tim Flechtner <timf@...> wrote:

hi everyone,
i'm trying to get distcc-3.1 running on solaris x86. i am getting the following error, can anyone help me out? i tried searching through google, and read the first four quarters of archived list emails but didn't see anything either place that looked familiar. is there a better way to search?

thanks!
-tim

userver43:/export/home/timf/builds/distcc_test/libs/meta $ DISTCC_POTENTIAL_HOSTS="udesktop71 udesktop103" pump gmake
__________Using distcc-pump from /opt/app/nonc++/distcc-3.1/bin
/opt/app/nonc++/distcc-3.1/bin/pump: DISTCC_HOSTS=--randomize udesktop103,cpp,lzo^Judesktop71,cpp,lzo: is not an identifier
__________Shutting down distcc-pump include server
userver43:/export/home/timf/builds/distcc_test/libs/meta $

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

--
Fergus Henderson <fergus@...>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: pump problem on solaris 10, x86

2009-11-25T07:35:44Z

The pump script is probably setting DISTCC_HOSTS from the output of lsdistcc.

Posted from an Android phone.

On Nov 24, 2009 7:46 PM, "Martin Pool" <mbp@...> wrote:

2009/11/25 Tim Flechtner <timf@...>:

> hi everyone, > i'm trying to get distcc-3.1 running on solaris x86. i am getting the > following...
Somewhere you must be setting DISTCC_HOSTS to a string that includes a
newline character, try taking that out?

--
Martin <http://launchpad.net/~mbp/>

__ distcc mailing list http://distcc.samba.org/ To unsubscribe or change options: https:/...

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: about distcc's scheduling

2009-11-25T02:38:10Z

Francois Berenger wrote:

> Francois Berenger wrote:
>> Fergus Henderson wrote:
>>> Have you tried using pump mode? That helps a lot.
>>>
>>> Also, if the issue is that distcc isn't running enough jobs on each
>>> host, try using the "host/JOBS" format in DISTCC_HOSTS,
>>> e.g. DISTCC_HOSTS="foo.riken.jp/20,cpp,lzo
>>> <http://foo.riken.jp/20,cpp,lzo> bar.riken.jp/40,cpp,lzo
>>> <http://bar.riken.jp/40,cpp,lzo>".
>>> This would allow up to 20 concurrent compiles on host foo.riken.jp
>>> <http://foo.riken.jp> and 40 concurrent compiles on host bar.riken.jp
>>> <http://bar.riken.jp>.
>>> (The ",cpp,lzo" part is required for pump mode.)
>>
>> Thanks a lot! I am trying this right away.
>> I'll post my speedup results when I'm done.
>
> Well, I managed to make our soft compile with:
> distcc-server-3.1-1
> distcc-3.1-1
>
> However, pump seems to be for 32bit machines only:
> titi@toto:x# pump scons -j24 bin mode=release
> __________Using distcc-pump from /usr/bin
> Traceback (most recent call last):
> File
> "/usr/lib/python2.4/site-packages/include_server/include_server.py",
> line 42, in ?
> import distcc_pump_c_extensions
> ImportError:
> /usr/lib/python2.4/site-packages/include_server/distcc_pump_c_extensions.so:
> wrong ELF class: ELFCLASS32
> __________Expected a socket at '/tmp/distcc-pump.E15442/socket'
> __________Could not start distcc-pump include server
> __________Warning: distcc-pump include server failed; running unpumped
> __________Found 0 available distcc servers
>
> Can someone publish x86_64 binary packages in the download section of
> the distcc website?
> Or at least the source RPM so I can to build for my systems.

I was unable to build the rpm myself from the tar ball.
I give up, I fought enough.
Here is the make rpm output:

Checking for unpackaged file(s): /usr/lib/rpm/check-files
/var/tmp/distcc-buildroot
error: Installed (but unpackaged) file(s) found:
/usr/lib64/python2.4/site-packages/include_server/__init__.pyo
/usr/lib64/python2.4/site-packages/include_server/basics.pyo
/usr/lib64/python2.4/site-packages/include_server/basics_test.pyo
/usr/lib64/python2.4/site-packages/include_server/c_extensions_test.pyo
/usr/lib64/python2.4/site-packages/include_server/cache_basics.pyo
/usr/lib64/python2.4/site-packages/include_server/compiler_defaults.pyo
/usr/lib64/python2.4/site-packages/include_server/compress_files.pyo
/usr/lib64/python2.4/site-packages/include_server/include_analyzer.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_memoizing_node.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_memoizing_node_test.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_test.pyo
/usr/lib64/python2.4/site-packages/include_server/include_server.pyo

/usr/lib64/python2.4/site-packages/include_server/include_server_test.pyo
/usr/lib64/python2.4/site-packages/include_server/macro_eval.pyo
/usr/lib64/python2.4/site-packages/include_server/macro_eval_test.pyo
/usr/lib64/python2.4/site-packages/include_server/mirror_path.pyo
/usr/lib64/python2.4/site-packages/include_server/mirror_path_test.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_command.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_command_test.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_file.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_file_test.pyo
/usr/lib64/python2.4/site-packages/include_server/run.pyo
/usr/lib64/python2.4/site-packages/include_server/setup.pyo
/usr/lib64/python2.4/site-packages/include_server/statistics.pyo

RPM build errors:
Installed (but unpackaged) file(s) found:
/usr/lib64/python2.4/site-packages/include_server/__init__.pyo
/usr/lib64/python2.4/site-packages/include_server/basics.pyo
/usr/lib64/python2.4/site-packages/include_server/basics_test.pyo
/usr/lib64/python2.4/site-packages/include_server/c_extensions_test.pyo
/usr/lib64/python2.4/site-packages/include_server/cache_basics.pyo
/usr/lib64/python2.4/site-packages/include_server/compiler_defaults.pyo
/usr/lib64/python2.4/site-packages/include_server/compress_files.pyo
/usr/lib64/python2.4/site-packages/include_server/include_analyzer.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_memoizing_node.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_memoizing_node_test.pyo

/usr/lib64/python2.4/site-packages/include_server/include_analyzer_test.pyo
/usr/lib64/python2.4/site-packages/include_server/include_server.pyo

/usr/lib64/python2.4/site-packages/include_server/include_server_test.pyo
/usr/lib64/python2.4/site-packages/include_server/macro_eval.pyo
/usr/lib64/python2.4/site-packages/include_server/macro_eval_test.pyo
/usr/lib64/python2.4/site-packages/include_server/mirror_path.pyo
/usr/lib64/python2.4/site-packages/include_server/mirror_path_test.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_command.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_command_test.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_file.pyo
/usr/lib64/python2.4/site-packages/include_server/parse_file_test.pyo
/usr/lib64/python2.4/site-packages/include_server/run.pyo
/usr/lib64/python2.4/site-packages/include_server/setup.pyo
/usr/lib64/python2.4/site-packages/include_server/statistics.pyo

> Thanks a lot,
> Francois.
>
>>> At some point, depending on the size and structure of the software
>>> that you're trying to compile,
>>> and the way your Makefiles are arranged, you may run into limitations
>>> in the parallelizability of your build graph.
>>> For example, if your software has only three source files, then
>>> distcc won't be able to speed up compilation by more than a factor of
>>> three at very best.
>>> Similarly, if your software has 30 files, but they are split into 10
>>> directories of three files each, and your Makefile(s) sequentialize
>>> the compilation of each directory, then you still won't get more than
>>> a factor of three speed up at very best.
>>
>> We use scons. I am not so happy with it (like when it spends 30s
>> before giving me back control of the command line when theres is
>> _nothing_ to compile because everything is already done).
>> However, it manages the dependencies far better than any Makefile I
>> could have written by hand.
>>
>> Regards,
>> Francois.
>>
>>> On Mon, Nov 23, 2009 at 8:19 PM, Francois Berenger <berenger@...
>>> <mailto:berenger@...>> wrote:
>>>
>>> Hello,
>>>
>>> Is there a way to use a pull scheduler instead of a push scheduler
>>> inside distcc?
>>>
>>> I have many compiling hosts and when they join a compiling job,
>>> they are
>>> not enough loaded, at least to my taste. Meaning: they could compile
>>> more things and I could wait less for them to finish.
>>>
>>> I think the problem is that the scheduler is pushing: the distcc
>>> server sends jobs to distcc compiling hosts.
>>>
>>> Instead, I would like this behavior: distcc compiling hosts ask and
>>> take jobs as long as they are not completely loaded with compilation
>>> jobs.
>>> The distcc server should distribute jobs to whom asks for.
>>>
>>> This should scale better.
>>>
>>> If there is no way now to do so but you can give me some hint on
>>> how to
>>> do so (where to look, what to modify, etc.), it might be helpful.
>>>
>>> Best regards,
>>> Francois.
>>>
>>> __
>>> distcc mailing list http://distcc.samba.org/
>>> To unsubscribe or change options:
>>> https://lists.samba.org/mailman/listinfo/distcc
>>>
>>>
>>>
>>>
>>> --
>>> Fergus Henderson <fergus@... <mailto:fergus@...>>
>>
>> __
>> distcc mailing list http://distcc.samba.org/
>> To unsubscribe or change options:
>> https://lists.samba.org/mailman/listinfo/distcc
>
>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: about distcc's scheduling

2009-11-25T01:13:07Z

Francois Berenger wrote:

> Fergus Henderson wrote:
>> Have you tried using pump mode? That helps a lot.
>>
>> Also, if the issue is that distcc isn't running enough jobs on each
>> host, try using the "host/JOBS" format in DISTCC_HOSTS,
>> e.g. DISTCC_HOSTS="foo.riken.jp/20,cpp,lzo
>> <http://foo.riken.jp/20,cpp,lzo> bar.riken.jp/40,cpp,lzo
>> <http://bar.riken.jp/40,cpp,lzo>".
>> This would allow up to 20 concurrent compiles on host foo.riken.jp
>> <http://foo.riken.jp> and 40 concurrent compiles on host bar.riken.jp
>> <http://bar.riken.jp>.
>> (The ",cpp,lzo" part is required for pump mode.)
>
> Thanks a lot! I am trying this right away.
> I'll post my speedup results when I'm done.

Well, I managed to make our soft compile with:
distcc-server-3.1-1
distcc-3.1-1

However, pump seems to be for 32bit machines only:
titi@toto:x# pump scons -j24 bin mode=release
__________Using distcc-pump from /usr/bin
Traceback (most recent call last):
File
"/usr/lib/python2.4/site-packages/include_server/include_server.py",
line 42, in ?
import distcc_pump_c_extensions
ImportError:
/usr/lib/python2.4/site-packages/include_server/distcc_pump_c_extensions.so:
wrong ELF class: ELFCLASS32
__________Expected a socket at '/tmp/distcc-pump.E15442/socket'
__________Could not start distcc-pump include server
__________Warning: distcc-pump include server failed; running unpumped
__________Found 0 available distcc servers

Can someone publish x86_64 binary packages in the download section of
the distcc website?
Or at least the source RPM so I can to build for my systems.

Thanks a lot,
Francois.

>> At some point, depending on the size and structure of the software
>> that you're trying to compile,
>> and the way your Makefiles are arranged, you may run into limitations
>> in the parallelizability of your build graph.
>> For example, if your software has only three source files, then distcc
>> won't be able to speed up compilation by more than a factor of three
>> at very best.
>> Similarly, if your software has 30 files, but they are split into 10
>> directories of three files each, and your Makefile(s) sequentialize
>> the compilation of each directory, then you still won't get more than
>> a factor of three speed up at very best.
>
> We use scons. I am not so happy with it (like when it spends 30s before
> giving me back control of the command line when theres is _nothing_ to
> compile because everything is already done).
> However, it manages the dependencies far better than any Makefile I
> could have written by hand.
>
> Regards,
> Francois.
>
>> On Mon, Nov 23, 2009 at 8:19 PM, Francois Berenger <berenger@...
>> <mailto:berenger@...>> wrote:
>>
>> Hello,
>>
>> Is there a way to use a pull scheduler instead of a push scheduler
>> inside distcc?
>>
>> I have many compiling hosts and when they join a compiling job,
>> they are
>> not enough loaded, at least to my taste. Meaning: they could compile
>> more things and I could wait less for them to finish.
>>
>> I think the problem is that the scheduler is pushing: the distcc
>> server sends jobs to distcc compiling hosts.
>>
>> Instead, I would like this behavior: distcc compiling hosts ask and
>> take jobs as long as they are not completely loaded with compilation
>> jobs.
>> The distcc server should distribute jobs to whom asks for.
>>
>> This should scale better.
>>
>> If there is no way now to do so but you can give me some hint on
>> how to
>> do so (where to look, what to modify, etc.), it might be helpful.
>>
>> Best regards,
>> Francois.
>>
>> __
>> distcc mailing list http://distcc.samba.org/
>> To unsubscribe or change options:
>> https://lists.samba.org/mailman/listinfo/distcc
>>
>>
>>
>>
>> --
>> Fergus Henderson <fergus@... <mailto:fergus@...>>
>
> __
> distcc mailing list http://distcc.samba.org/
> To unsubscribe or change options:
> https://lists.samba.org/mailman/listinfo/distcc

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: pump problem on solaris 10, x86 [solved]

2009-11-24T17:50:59Z

ah ha! thank you very much.

-tim

Martin Pool wrote:

> 2009/11/25 Tim Flechtner <timf@...>:
>
>> hi everyone,
>> i'm trying to get distcc-3.1 running on solaris x86. i am getting the
>> following error, can anyone help me out? i tried searching through google,
>> and read the first four quarters of archived list emails but didn't see
>> anything either place that looked familiar. is there a better way to
>> search?
>>
>> thanks!
>> -tim
>>
>> userver43:/export/home/timf/builds/distcc_test/libs/meta $
>> DISTCC_POTENTIAL_HOSTS="udesktop71 udesktop103" pump gmake
>> __________Using distcc-pump from /opt/app/nonc++/distcc-3.1/bin
>> /opt/app/nonc++/distcc-3.1/bin/pump: DISTCC_HOSTS=--randomize
>> udesktop103,cpp,lzo^Judesktop71,cpp,lzo: is not an identifier
>> __________Shutting down distcc-pump include server
>> userver43:/export/home/timf/builds/distcc_test/libs/meta $
>>
>
> Somewhere you must be setting DISTCC_HOSTS to a string that includes a
> newline character, try taking that out?
>
>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: pump problem on solaris 10, x86

2009-11-24T16:45:39Z

2009/11/25 Tim Flechtner <timf@...>:

> hi everyone,
> i'm trying to get distcc-3.1 running on solaris x86. i am getting the
> following error, can anyone help me out? i tried searching through google,
> and read the first four quarters of archived list emails but didn't see
> anything either place that looked familiar. is there a better way to
> search?
>
> thanks!
> -tim
>
> userver43:/export/home/timf/builds/distcc_test/libs/meta $
> DISTCC_POTENTIAL_HOSTS="udesktop71 udesktop103" pump gmake
> __________Using distcc-pump from /opt/app/nonc++/distcc-3.1/bin
> /opt/app/nonc++/distcc-3.1/bin/pump: DISTCC_HOSTS=--randomize
> udesktop103,cpp,lzo^Judesktop71,cpp,lzo: is not an identifier
> __________Shutting down distcc-pump include server
> userver43:/export/home/timf/builds/distcc_test/libs/meta $

Somewhere you must be setting DISTCC_HOSTS to a string that includes a
newline character, try taking that out?

--
Martin <http://launchpad.net/~mbp/>
__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

pump problem on solaris 10, x86

2009-11-24T15:17:14Z

hi everyone,
i'm trying to get distcc-3.1 running on solaris x86. i am getting the
following error, can anyone help me out? i tried searching through
google, and read the first four quarters of archived list emails but
didn't see anything either place that looked familiar. is there a
better way to search?

thanks!
-tim

userver43:/export/home/timf/builds/distcc_test/libs/meta $
DISTCC_POTENTIAL_HOSTS="udesktop71 udesktop103" pump gmake
__________Using distcc-pump from /opt/app/nonc++/distcc-3.1/bin
/opt/app/nonc++/distcc-3.1/bin/pump: DISTCC_HOSTS=--randomize
udesktop103,cpp,lzo^Judesktop71,cpp,lzo: is not an identifier
__________Shutting down distcc-pump include server
userver43:/export/home/timf/builds/distcc_test/libs/meta $

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc

Re: about distcc's scheduling

2009-11-23T20:36:20Z

Fergus Henderson wrote:

> Have you tried using pump mode? That helps a lot.
>
> Also, if the issue is that distcc isn't running enough jobs on each
> host, try using the "host/JOBS" format in DISTCC_HOSTS,
> e.g. DISTCC_HOSTS="foo.riken.jp/20,cpp,lzo
> <http://foo.riken.jp/20,cpp,lzo> bar.riken.jp/40,cpp,lzo
> <http://bar.riken.jp/40,cpp,lzo>".
> This would allow up to 20 concurrent compiles on host foo.riken.jp
> <http://foo.riken.jp> and 40 concurrent compiles on host bar.riken.jp
> <http://bar.riken.jp>.
> (The ",cpp,lzo" part is required for pump mode.)

Thanks a lot! I am trying this right away.
I'll post my speedup results when I'm done.

> At some point, depending on the size and structure of the software that
> you're trying to compile,
> and the way your Makefiles are arranged, you may run into limitations in
> the parallelizability of your build graph.
> For example, if your software has only three source files, then distcc
> won't be able to speed up compilation by more than a factor of three at
> very best.
> Similarly, if your software has 30 files, but they are split into 10
> directories of three files each, and your Makefile(s) sequentialize the
> compilation of each directory, then you still won't get more than a
> factor of three speed up at very best.

We use scons. I am not so happy with it (like when it spends 30s before
giving me back control of the command line when theres is _nothing_ to
compile because everything is already done).
However, it manages the dependencies far better than any Makefile I
could have written by hand.

Regards,
Francois.

> On Mon, Nov 23, 2009 at 8:19 PM, Francois Berenger <berenger@...
> <mailto:berenger@...>> wrote:
>
> Hello,
>
> Is there a way to use a pull scheduler instead of a push scheduler
> inside distcc?
>
> I have many compiling hosts and when they join a compiling job, they are
> not enough loaded, at least to my taste. Meaning: they could compile
> more things and I could wait less for them to finish.
>
> I think the problem is that the scheduler is pushing: the distcc
> server sends jobs to distcc compiling hosts.
>
> Instead, I would like this behavior: distcc compiling hosts ask and
> take jobs as long as they are not completely loaded with compilation
> jobs.
> The distcc server should distribute jobs to whom asks for.
>
> This should scale better.
>
> If there is no way now to do so but you can give me some hint on how to
> do so (where to look, what to modify, etc.), it might be helpful.
>
> Best regards,
> Francois.
>
> __
> distcc mailing list http://distcc.samba.org/
> To unsubscribe or change options:
> https://lists.samba.org/mailman/listinfo/distcc
>
>
>
>
> --
> Fergus Henderson <fergus@... <mailto:fergus@...>>

__
distcc mailing list http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc