I got the IPMI plugin to work in an ATCA chassis....almost. See my
email of yesterday for one issue and also, the plugin discovered the
blades OK, but had a hard time with AMC cards installed in the blades.
This particular chassis had two Intel MPCBL0040 blades with an AMC card
each - the plugin found *one* AMC card, which the 'hpitree' and 'hpitop'
commands placed off of the system chassis...the entity path was
something like,
{SYSTEM_CHASSIS, 3}{193, 6}
...where the entity ID and instance are (I believe) OEM-specific.
Currently having an issue with the 2.10.1 version of 'ipmidirect' where,
after discovering the satellite MCs for all the blades and reading their
device SDR repositories, it sends another 'Reserve Device SDR
Repositoty' command to the BMC for the blade's slave address (0x92 for
example), but to LUN 1 instead of LUN 0. Not sure why it does that, but
it fails with completion code 0xcc, which the IPMI spec says is,
"Invalid data field in request". Which is weird because that command
does not expect any data and I don't think (from the log) that openhpid
is trying to send any.
Good to know that most people use 'ipmidirect', though. I'd been
planning to use whichever one I could get working soonest and best. How
widespread is the use of OpenHPI in general, and this plugin
specifically, in the telecom industry, does anyone know? Just curious.
Jason
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Bryan
Sutula
Sent: Thursday, January 03, 2008 1:31 PM
To: [email protected]
Subject: Re: [Openhpi-devel] openhpi v 2.10.1 seg faults with libipmi
plugin
Here as well, I haven't gotten the IPMI plugin to work with ATCA
equipment either. Most people are using the ipmidirect plugin for ATCA.
It's on my longer-term list to figure out why it doesn't work and maybe
fix it. If you have time and interest to see what's wrong, I'd
appreciate any patches in this area.
Thanks,
Bryan
On Thu, 2008-01-03 at 10:02 +0530, Anand S Katti wrote:
> Hi Bryan,
>
> The Problem got resolved, I had multiple copies of the hpi library and
> was linking to the older version.
>
> But still it doesn't discover the ATCA Blades.
>
> Im running on RHEL 4.0 with 2.6 kernel; and using our own Continuous
> Computing 5U ATCA Chassis.
>
> Thanks for the response,
> Anand
>
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of
Bryan
> Sutula
> Sent: Thursday, December 27, 2007 12:52 AM
> To: [email protected]
> Subject: Re: [Openhpi-devel] openhpi v 2.10.1 seg faults with libipmi
> plugin
>
> It would be helpful if you could provide a few more specifics:
>
> - What OS and/or distribution, and version of OS?
> - What sort of hardware are you trying to access with OpenHPI?
>
> I will be mostly unavailable for the next week as well, but will try
to
> look at this when I'm back in the office next week.
>
> Thanks
> Bryan Sutula
>
>
> On Sat, 2007-12-22 at 20:58 +0530, Anand S Katti wrote:
> > Hi
> >
> >
> >
> > I tried libipmi plugin with openhpid, and I am seeing following
error
> > with it.
> >
> > It is creating 5 openhpid processes. With hpi_shell I did lsres and
it
> > doesn't list any resources either; it is empty.
> >
> >
> >
> > The behavior is same with libOpenIPMI libraries from 1.4.20 and
2.0.13
> > versions as well.
> >
> >
> >
> > It would be great if anyone tells me where im going wrong.
> >
> >
> >
> > PS: openhpid works flawless with libipmidirect plugin.
> >
> >
> >
> > Thanks
> >
> > Anand
> >
> >
> >
> > Gdb shows that the process segfaults at accept().
> >
> >
----------------------------------------------------------------------
> >
> > 290 if (stop_server) {
> >
> > (gdb)
> >
> > 294 if (servinst->Accept()) {
> >
> > (gdb) s
> >
> > sstrmsock::Accept (this=0x8beef08) at strmsock.cpp:316
> >
> > 316 socklen_t sz = sizeof(addr);
> >
> > (gdb) n
> >
> > 318 if (!fOpenS) {
> >
> > (gdb) p fOpenS
> >
> > $1 = true
> >
> > (gdb) n
> >
> > 322 sz = sizeof (struct sockaddr);
> >
> > (gdb)
> >
> > 323 s = accept(ss, (struct sockaddr *) &addr, &sz);
> >
> > (gdb)
> >
> >
> >
> > Program received signal SIGSEGV, Segmentation fault.
> >
> > [Switching to Thread -1218585680 (LWP 5561)]
> >
> > 0x00000000 in ?? ()
> >
> > (gdb) bt
> >
> > #0 0x00000000 in ?? ()
> >
> > #1 0x00281ae3 in ipmi_lock (lock=0x8be0148) at locks.c:108
> >
> > #2 0x00cdefee in _ipmi_fru_lock (fru=0x8be0148) at fru.c:201
> >
> > #3 0x00cdfedd in fetch_got_timestamp (fru=0x8bef250,
> > domain=0x8bea090, err=0,
> >
> > timestamp=167772160) at fru.c:639
> >
> > #4 0x00d1472d in atca_fru_254_get_timestamp_done (domain=0x8bea090,
> > rspi=0x8be0148)
> >
> > at oem_atca.c:3228
> >
> > #5 0x00cca41b in deliver_rsp (domain=0x8be0148,
> > rsp_handler=0x8bef3c0,
> >
> > rspi=0x8bed508) at domain.c:428
> >
> > #6 0x00ccc778 in ll_rsp_handler (ipmi=0x8be6550, orspi=0x8bef058)
at
> > domain.c:1869
> >
> > #7 0x00cca314 in ipmi_handle_rsp_item (ipmi=0x8be6550,
> > rspi=0x8bef058,
> >
> > rsp_handler=0x8be0148) at ipmi.c:1765
> >
> > #8 0x00d19cb9 in handle_payload (ipmi=0x8be6550, lan=0x8be6608,
> > addr_num=0,
> >
> > payload_type=146731104, tmsg=0xb75dd03e "\201???\037",
> > payload_len=14)
> >
> > at ipmi_lan.c:2994
> >
> > #9 0x00d1b2f8 in data_handler (fd=7, cb_data=0x8be9e60,
id=0x8be9f18)
> >
> > at ipmi_lan.c:3308
> >
> > #10 0x0027a09f in fd_handler (fd=7, data=0x8be0148) at
> > posix_os_hnd.c:87
> >
> > #11 0x0027bacc in process_fds (sel=0x8be01f8, send_sig=0x8bef3c0,
> > thread_id=0,
> >
> > cb_data=0x0, timeout=0xb75dd358) at selector.c:638
> >
> > #12 0x0027c069 in sel_select (sel=0x8be01f8, send_sig=0,
thread_id=0,
> > cb_data=0x0,
> >
> > timeout=0x0) at selector.c:740
> >
> > #13 0x0054761a in ipmi_get_event (hnd=0x8bdc6e8) at ipmi.c:538
> >
> > #14 0x0805beb6 in harvest_events_for_handler (h=0x8bdc720) at
> > event.c:111
> >
> > #15 0x0805c0a5 in oh_harvest_events () at event.c:141
> >
> > #16 0x0805d663 in oh_evtget_thread_loop (data=0x0) at threaded.c:130
> >
> > #17 0x00218812 in g_static_private_free ()
> > from /usr/lib/libglib-2.0.so.0
> >
> > #18 0x005fc3ae in start_thread () from /lib/tls/libpthread.so.0
> >
> > #19 0x0047caee in clone () from /lib/tls/libc.so.6
> >
> > (gdb)
> >
> >
> >
> >
> >
> >
> > Here is the extract of openhpi.conf for libipmi....
> >
> >
> >
> >
> >
> > OPENHPI_LOG_ON_SEV = "MINOR"
> >
> > OPENHPI_ON_EP = "{SYSTEM_CHASSIS,3}"
> >
> > OPENHPI_EVT_QUEUE_LIMIT = 10000
> >
> > OPENHPI_DEL_SIZE_LIMIT = 10000
> >
> > OPENHPI_DEL_SAVE = "NO"
> >
> > OPENHPI_DAT_SIZE_LIMIT = 0
> >
> > OPENHPI_DAT_USER_LIMIT = 0
> >
> > OPENHPI_DAT_SAVE = "NO"
> >
> > OPENHPI_PATH = "/usr/local/lib/openhpi:/usr/lib/openhpi"
> >
> > OPENHPI_VARPATH = "/usr/local/var/lib/openhpi"
> >
> >
> >
> >
> >
> > ## Section for ipmi plugin based on OpenIPMI:
> >
> > handler libipmi {
> >
> > entity_root = "{SYSTEM_CHASSIS,3}"
> >
> > name = "lan"
> >
> > addr = "172.25.10.150" #ipaddress
> >
> > port = 623
> >
> > auth_type = "none"
> >
> > auth_level= "admin"
> >
> > username = ""
> >
> > password = ""
> >
> > }
> >
> >
> >
> >
> >
> >
> >
> >
>
------------------------------------------------------------------------
> -
> > This SF.net email is sponsored by: Microsoft
> > Defy all challenges. Microsoft(R) Visual Studio 2005.
> > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> > _______________________________________________ Openhpi-devel
mailing
> list [email protected]
> https://lists.sourceforge.net/lists/listinfo/openhpi-devel
>
>
>
------------------------------------------------------------------------
> -
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Openhpi-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/openhpi-devel
>
>
>
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Openhpi-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/openhpi-devel
--
Bryan Sutula <[EMAIL PROTECTED]>
------------------------------------------------------------------------
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Openhpi-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openhpi-devel
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Openhpi-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openhpi-devel
