panic:Data page fault. what was the cause of the panic?

[stephen peng]

dear all,
i've got my rp5470 panic and reboot. the syslog logged "panic: Data page fault", and there were folling messages in q4's what.out, which i think that it would tell something about the panic:
Interrupt Type 15 (Data page fault):
Instruction Address (pcsq.pcoq) = 0x0.0x148db4
Instruction (iir) = 0x40970020 (load/store)
Target Address (isr.ior) = 0x0.0x0000000000000010
Base Register (gr4) = 0x0000000000000000
Savestate Ptr (ssp) = 0x0.0x00000000015c5eb0
Savestate Return Pointer (ss_rp) = 0x0000000000148d00

+--------------------------------------------+
| Hang probability |
+--------------------------------------------+
No threads ran during the last 1 second before the dump.
No threads ran during the last 5 seconds before the dump.
No threads ran during the last 1 minute before the dump.
115 or more threads ran during the last 1 hour before the dump.
122 or more threads ran during the last 1 day before the dump.
+--------------------------------------------+
| Checking other hang reasons |
+--------------------------------------------+
The statdaemon hadn't run in over 1.06 minutes (1 second is normal).
At least 2 real-time processes ran since the last time the statdaemon
ran.

There were 3 zombie processes that died more than 1 hour before the dump
(but were never reaped). This is not illegal, but it does suggest one
or more processes are either being starved or are not cleaning up after
themselves. The parents of the zombies were:
proc @ 0x4a59a040, pid 1, cmd init
proc @ 0xae6a40, pid 0, cmd swapper

could you please try to figure out the cause of the panic?

thanks a lot
your reply would earn you points!

[Mridul Shrivastava]

Hi Stephen,

Data page fault is caused by software and cause of the panic in these cases mostly is software issue.

Do you have output of crashinfo or could you please post the output of crashinfo.. since the information posted is not sufficient to confirm the cause of the panic.

[Dennis Handly]

>Instruction Address (pcsq.pcoq) = 0x0.0x148db4

If you can disassemble this address, it will give the current function.

>Target Address (isr.ior) = 0x0.0x0000000000000010

This looks like 16 bytes off of a NULL pointer.

So the cause is the kernel messed up. :-(
I'm not sure what is needed to match it up with a patch.

[Steven E. Protter]

Shalom,

Check /etc/rc.config.d/savecrash

If the first variable is set to 1, you hae a crash dump. If not, set it to 1 and wait for it to happen again.

/var/adm/crash

Your crash will be a subdirectory here. You next need to do q4 analysis and send the output to the HP response center. They will tell you which patch you are missing or what other corrective action to take.

It seems like software but take the time to use mstm/cstm or xstm to check the hardware, especially memory.

SEP

[stephen peng]

hi all,
there is no hardware error, and there is not hints in the crash/INDEX, i could post the ana.out and what.out produced by q4 here for every one to take a look.
if i do not have an HP service contract, could i sent the log file to HP to analyze them? how?

thanks

[Mridul Shrivastava · Accepted solution]

I checked the q4 output and found that processor 1 did panic..
There were some igelan related tasks were going on and were not able to finish..That's all I could make out..
I suspect it was caused by igelan.. could you please checkthe sytem for the igelan patches and latest drivers..

there are lots of issues seen with igelan so I would recommend you to install the latest patches and drivers if they are not..

[stephen peng]

Mridul,
hi,
what is igelan?how to examine the present patches for igelan and how tofind the lastest patches for igelan?

thanks

[Mridul Shrivastava]

igelan is the driver used for Gigabit lan card..
and you can search for the igelan driver and its patches via the following command:

#swlist -l product |grep -i igelan

latest patches can be checked at itrc... however driver can be downloaded from software.hp.com..