Well you can fix this one by applying by hand the attached patch to
ipmi_domain.cpp around line 540. But there are many others that were fixed
since then (This one was fixed 2 years ago!) and I am not going to support
such an old version.
Good luck!
Pierre
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jason
Barrett
Sent: Wednesday, December 12, 2007 22:53
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
OK, so for product and support reasons I'm told that we cannot upgrade any
element that's considered part of the core OS, including OpenHPI. We have
opened a ticket with MontaVista (the OS vendor), but in the meantime, I'm
trying to understand as best I can what is happening and why this hardware
configuration is causing a failure so I can try to plan a way around it.
While I start to plumb the depths of the code on SourceForge, does anyone
have a summary of this problem and/or a suggested workaround? Any
assistance is much appreciated.
Good night all,
Jason
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jason
Barrett
Sent: Wednesday, December 12, 2007 11:03 AM
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
Pierre,
Gotcha. That's probably what we'll do unless someone gives me a good reason
not to. Just asking out of due diligence, but are there any workarounds
available for this bug in lieu of an upgrade?
Thanks,
Jason
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Pierre
Sangouard
Sent: Wednesday, December 12, 2007 10:58 AM
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
Jason,
If you are to upgrade, I would recommend moving to the latest stable version
- that is 2.10.1.
Pierre
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jason
Barrett
Sent: Wednesday, December 12, 2007 15:48
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
Thanks, Pierre, let me see what I have to do to upgrade from the version
bundled with the MV Linux OS.
Jason
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Pierre
Sangouard
Sent: Wednesday, December 12, 2007 3:41 AM
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
You've hit bug #1365565 which was fixed in 2.2.2.
Pierre
_____
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jason
Barrett
Sent: Wednesday, December 12, 2007 03:46
To: [email protected]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
Thanks, Michael, that sheds a little light on the IPMI results. So is that
Shelf FRU device at 0x10 bogus? I'm curious why the code doesn't just
ignore it, then.
The shelf manager is Intel-based.
Thanks,
Jason
_____
From: [EMAIL PROTECTED] on behalf of
[EMAIL PROTECTED]
Sent: Tue 12/11/2007 6:50 PM
To: [email protected]
Cc: [email protected];
[EMAIL PROTECTED]
Subject: Re: [Openhpi-devel] Runtime error in 'hpitree' opening session
The shelf manager at 0x20 is the BMC, the shelf manager at 0x10 is the
physical shelf manager.
Usually ATCA chassis FRU data is only on the BMC at 0x20.
Is this a PPS or Intel based shelf manager?
Michael Thompson
Principal Engineer
Pentair Electronic Packaging
(Schroff & Electronic Solutions)
170 Commerce Drive
Warwick, RI 02886
Phone: +1 401-535-4869
FAX: +1 401-535-4951
Email: [EMAIL PROTECTED]
"Jason Barrett" <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
12/11/2007 06:14 PM
Please respond to
[email protected]
To
<[email protected]>
cc
Subject
[Openhpi-devel] Runtime error in 'hpitree' opening session
Good evening,
I'm running 'hpitree' on a blade in an ATCA chassis using the ipmidirect
plugin and RMCP. I'm getting the following error at the point where it
finds the second of two Shelf FRU information modules (the first at IPMB
address 0x20, the second at 0x10):
hpitree: ../../../plugins/ipmidirect/ipmi_fru_info.cpp:156: cIpmiFruInfo*
cIpmiFruInfoContainer::NewFruInfo(unsigned int, unsigned int,
SaHpiEntityTypeT, unsigned int, tIpmiAtcaSiteType, unsigned int): Assertion
`fru_id == 0' failed.
I'm using v.2.0.4, which is bundled with MontaVista Linux 4.0. Here's my
openhpi.conf file:
plugin libipmidirect
handler libipmidirect {
entity_root = "{SYSTEM_CHASSIS,7}"
name = "lan" # RMCP
addr = "localhost" # ipaddress
addr = "111.11.11.111" # IP changed to protect the innocent :-)
# port = "623" # RMCP port
auth_type = "md2" # none, md2, md5 or straight
auth_level = "operator" # none, callback, user, operator or admin
username = "root"
password = "thisisntreallymypassword"
logflags = "file stdout"
logfile = "log"
logfile_max = 10
}
The chassis has one physical shelf manager and five board slots, of which
four are populated, including this blade at 0x8a.
I'm trying to understand a little better how this plugin works. Why is the
code asserting fru_id == 0? That value designates an IPMC proper, correct?
Would the Shelf FRU information module have a different FRU device ID?
Using raw IPMI through impish, I see a system-relative ShMC at (240.0) and a
device-relative ShMC at (0.16.240.0). This appears to be the one it's
polling when it fails (0x10=16d). Both ShMC records appear to have
associated records of type 'ATCA Shelf FRU) at (0.32.242.0) and (0.16.242.0)
respectively.
The last few entries in the 'hpitree' output are:
17:42:12.101 found ATCA Board at 0x8a.
17:42:12.101 >cmd 0x08 0c 0f 00 GetAddressInfo (05) 00 00 03 06 00
17:42:12.104 <rsp 0x08 0c 0f 00 GetAddressInfo (08) c9 00 01 86 ff 00
03 00
17:42:12.104 checking for Power Entry Module.
17:42:12.104 >cmd 0x09 0c 0f 00 GetAddressInfo (05) 00 00 03 01 01
17:42:12.108 <rsp 0x09 0c 0f 00 GetAddressInfo (08) c9 00 12 00 00 51
3d 98
17:42:12.108 checking for Shelf FRU Information.
17:42:12.108 >cmd 0x0a 0c 0f 00 GetAddressInfo (05) 00 00 03 01 02
17:42:12.111 <rsp 0x0a 0c 0f 00 GetAddressInfo (08) 00 00 10 20 ff 00
01 02
17:42:12.111 found Shelf FRU Information at 0x20.
17:42:12.111 >cmd 0x0b 0c 0f 00 GetAddressInfo (05) 00 00 03 02 02
17:42:12.114 <rsp 0x0b 0c 0f 00 GetAddressInfo (08) 00 00 08 10 ff 02
02 02
17:42:12.114 found Shelf FRU Information at 0x10.
Thanks,
Jason
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________------
-------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php_______________________________
________________
Openhpi-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openhpi-devel
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
- // add slot for initial scan
- NewFruInfo( rsp.m_data[3], rsp.m_data[5], entity, j + 1,
- (tIpmiAtcaSiteType)i,
m_atca_site_property[i].m_property );
+ // add MC for initial scan (FRU 0)
+ if (rsp.m_data[5] == 0)
+ NewFruInfo( rsp.m_data[3], rsp.m_data[5], entity, j + 1,
+ (tIpmiAtcaSiteType)i,
m_atca_site_property[i].m_property );
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Openhpi-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openhpi-devel
