[bug #24300] ipmiconsole cannot connect to various Intel servers

[INVALID.NOREPLY]

Follow-up Comment #4, bug #24300 (project freeipmi):

I will try that Monday or Tuesday.


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?24300>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[INVALID.NOREPLY]

Follow-up Comment #5, bug #24300 (project freeipmi):

I tried freeIPMI 0.3.0, and ipmipower works:

  $ ipmipower -h xiv3-ipmi -R 2.0 --intel-2-0-session
  > stat

But that also works in freeIPMI 0.7.2:

  $ ipmipower -h xiv3-ipmi --driver-type=LAN_2_0 --authent=NONE
--workaround=intel20
  > stat

ipmi-chassis 0.7.2 continues to fail with k_g invalid.

I said before that ipmiconsole 0.7.2 fails with k_g_invalid, but today I see
it really fails with "password" invalid, as for the other reporter.  I'm not
specifying a userid or password (and the BMC is fine with that with
Ipmitool).

On 0.3.0, ipmiconsole --intel-2-0-session hangs for about a minute, then
fails with "BMC Error".


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?24300>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan,

Thanks.  That means I've just ported something incorrectly over to the
other tools.  I'll audit some more to try and figure it out.  Because
you are using a NULL username/password, there's probably just some
corner case in there.

Thanks,
Al

On Tue, 2008-12-09 at 01:10 +0000, Bryan Henderson wrote:

> Follow-up Comment #5, bug #24300 (project freeipmi):
>
> I tried freeIPMI 0.3.0, and ipmipower works:
>
>   $ ipmipower -h xiv3-ipmi -R 2.0 --intel-2-0-session
>   > stat
>
> But that also works in freeIPMI 0.7.2:
>
>   $ ipmipower -h xiv3-ipmi --driver-type=LAN_2_0 --authent=NONE
> --workaround=intel20
>   > stat
>
> ipmi-chassis 0.7.2 continues to fail with k_g invalid.
>
> I said before that ipmiconsole 0.7.2 fails with k_g_invalid, but today I see
> it really fails with "password" invalid, as for the other reporter.  I'm not
> specifying a userid or password (and the BMC is fine with that with
> Ipmitool).
>
> On 0.3.0, ipmiconsole --intel-2-0-session hangs for about a minute, then
> fails with "BMC Error".
>
>
>     _______________________________________________________
>
> Reply to this item at:
>
>   <http:// savannah.gnu.org/bugs/?24300>
>
> _______________________________________________
>   Message sent via/by Savannah
>   http:// savannah.gnu.org/
>
>

--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan, Andrew,

I looked more closely and I think I found the Intel 2.0 authentication
workaround issues that prohibited ipmi-chassis from working.  I also
gave the "activate payload" command an intel specific tweak so that
ipmiconsole can hopefully now connect for a SOL session.  

So hopefully this release will fix them.  Could you give the tar.gz
below a shot?  Also, could you tell me what motherboards you are testing
against?  So I can add it into the documentation.

http://ftp.zresearch.com/pub/freeipmi/qa-
release/freeipmi-0.7.4.beta1.tar.gz

Thanks,
Al

P.S.  Remember to use the "intel20" workaround flag.

On Tue, 2008-12-09 at 01:10 +0000, Bryan Henderson wrote:

> Follow-up Comment #5, bug #24300 (project freeipmi):
>
> I tried freeIPMI 0.3.0, and ipmipower works:
>
>   $ ipmipower -h xiv3-ipmi -R 2.0 --intel-2-0-session
>   > stat
>
> But that also works in freeIPMI 0.7.2:
>
>   $ ipmipower -h xiv3-ipmi --driver-type=LAN_2_0 --authent=NONE
> --workaround=intel20
>   > stat
>
> ipmi-chassis 0.7.2 continues to fail with k_g invalid.
>
> I said before that ipmiconsole 0.7.2 fails with k_g_invalid, but today I see
> it really fails with "password" invalid, as for the other reporter.  I'm not
> specifying a userid or password (and the BMC is fine with that with
> Ipmitool).
>
> On 0.3.0, ipmiconsole --intel-2-0-session hangs for about a minute, then
> fails with "BMC Error".
>
>
>     _______________________________________________________
>
> Reply to this item at:
>
>   <http:// savannah.gnu.org/bugs/?24300>
>
> _______________________________________________
>   Message sent via/by Savannah
>   http:// savannah.gnu.org/
>
>

--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Bryan Henderson-2]

'ipmiconsole' works great in freeipmi-0.7.4.beta1.

'ipmi-chassis' fails a different way:

  $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
      --driver-type=LAN_2_0 --workaround=intel20 --debug

  [time passes]

  ipmi_ctx_open_outofband_2_0: session timeout

Attaching the debug output.


The motherboard is Intel S5000PSL.

--
Bryan Henderson                                   San Jose, California

_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan,

Cool.  that's good news that 'ipmiconsole' now works.  Not to look into
my ipmi-chassis still isn't working on the intels.  There's some
wierdness I'm seeing in the debug out.  I'm still taking a look.

Al

On Tue, 2008-12-09 at 23:31 +0000, Bryan Henderson wrote:

> 'ipmiconsole' works great in freeipmi-0.7.4.beta1.
>
> 'ipmi-chassis' fails a different way:
>
>   $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
>       --driver-type=LAN_2_0 --workaround=intel20 --debug
>
>   [time passes]
>
>   ipmi_ctx_open_outofband_2_0: session timeout
>
> Attaching the debug output.
>
>
> The motherboard is Intel S5000PSL.
>
> --
> Bryan Henderson                                   San Jose, California

--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan,

I think I found the bug.  It was a combination of a non-intel workaround
bug and an intel workaround bug.  Hopefully this one works now for ipmi-
chassis :-)

http://ftp.zresearch.com/pub/freeipmi/qa-
release/freeipmi-0.7.4.beta2.tar.gz

Thanks for all the help.  Without an Intel motherboard, my ability to do
this is somewhat limited.

Al

On Tue, 2008-12-09 at 23:31 +0000, Bryan Henderson wrote:

> 'ipmiconsole' works great in freeipmi-0.7.4.beta1.
>
> 'ipmi-chassis' fails a different way:
>
>   $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
>       --driver-type=LAN_2_0 --workaround=intel20 --debug
>
>   [time passes]
>
>   ipmi_ctx_open_outofband_2_0: session timeout
>
> Attaching the debug output.
>
>
> The motherboard is Intel S5000PSL.
>
> --
> Bryan Henderson                                   San Jose, California

--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Bryan Henderson-2]

With freeipmi-0.7.4.beta2.tar.gz, ipmi-chassis on Intel fails differently
from any of the earlier versions:

  $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
      --driver-type=LAN_2_0 --workaround=intel20 --debug

  ipmi_ctx_open_outofband_2_0: bad rmcpplus status code

Debug output attached.

--
Bryan Henderson                                   San Jose, California

_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan,

Thanks.  We're getting closer.  The status code of 0xF means the
integrity check value passed in via RAKP 3 was bad.  So I have something
to point me to the problem.  I'll let you know when I have another RPM
to try out.

Thanks,
Al

On Thu, 2008-12-11 at 23:32 +0000, Bryan Henderson wrote:

> With freeipmi-0.7.4.beta2.tar.gz, ipmi-chassis on Intel fails differently
> from any of the earlier versions:
>
>   $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
>       --driver-type=LAN_2_0 --workaround=intel20 --debug
>
>   ipmi_ctx_open_outofband_2_0: bad rmcpplus status code
>
> Debug output attached.
>
> --
> Bryan Henderson                                   San Jose, California

_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Hey Bryan,

Ok, I think I found (hopefully the final remaining) problem.  Could you
give it a shot?

http://ftp.zresearch.com/pub/freeipmi/qa-release/freeipmi-0.7.4.beta3.tar.gz

Thanks,
Al

On Thu, 2008-12-11 at 23:32 +0000, Bryan Henderson wrote:

> With freeipmi-0.7.4.beta2.tar.gz, ipmi-chassis on Intel fails differently
> from any of the earlier versions:
>
>   $ ipmi-chassis --get-status --host=xiv4-ipmi --authentication-type=NONE \
>       --driver-type=LAN_2_0 --workaround=intel20 --debug
>
>   ipmi_ctx_open_outofband_2_0: bad rmcpplus status code
>
> Debug output attached.
>
> --
> Bryan Henderson                                   San Jose, California

--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Bryan Henderson-2]

Looks like that got it.  Thanks!

--
Bryan Henderson                                   San Jose, California


_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[Al Chu11]

Cool.  Thanks for all your help.

I'll probably release 0.7.4 sometime early next week.

Thanks,
Al

On Sat, 2008-12-13 at 00:07 +0000, Bryan Henderson wrote:
> Looks like that got it.  Thanks!
>
--
Albert Chu
chu11@...
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[INVALID.NOREPLY]

Update of bug #24300 (project freeipmi):

                  Status:                    None => Fixed                  
             Assigned to:                    None => chu11                  
             Open/Closed:                    Open => Closed                


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?24300>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[INVALID.NOREPLY]

Follow-up Comment #6, bug #24300 (project freeipmi):

How did this get changed from Open with "BMC Error" to fixed, without code
changes being implemented?


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?24300>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel

[INVALID.NOREPLY]

Follow-up Comment #7, bug #24300 (project freeipmi):

Sorry, never documented that I fixed them.  I think we dealt with it in
e-mail.  It was fixed in the 0.7.4 release.

0.7.4 - 12/15/08
----------------
<snip>
o Fix Intel IPMI 2.0 workarounds in all tools/libraries.
<snip>

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?24300>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/



_______________________________________________
Freeipmi-devel mailing list
Freeipmi-devel@...
http://lists.gnu.org/mailman/listinfo/freeipmi-devel