After upgrade to August Patches - kernel trap 12
-
After Upgrade to August patches my system crash so I revert it back to RC3 release June
kernel trap 12 with interrupts disabled
Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address = 0x37
fault code = supervisor write, page not present
instruction pointer = 0x20:0xc0a78c45
stack pointer = 0x28:0xc47bcb10
frame pointer = 0x28:0xc47bcb14
code segment = base 0x0, limit 0xfffff, type 0x1b
= DPL 0, pres 1, def32 1, gran 1
processor eflags = resume, IOPL = 0
current process = 12 (swi4: clock)
-
Without the backtrace this is just not useful report.
-
I tried almost all full update of August but no luck
-
At the db> prompt, type "bt" (no quotes) and we need the full output from that to be able to tell anything.
Or try "call doadump" and then see if the GUI shows you that a crash report was saved when it boots back up.
-
after executing call doadump on
Version 2.0-RC3 (i386)
built on Wed Aug 24 10:02:03 EDT 201Diagnostics: Crash reporter
Processing…
Uploading...
minfree only report received. Skipping. This was fixed in a recent snapshot.
Continue and delete crash report files from local disk.I will download the latest snapshot today and see if the error goes away.
-
Again after executing "call doadump"
Version 2.0-RC3 (i386)
built on Thu Aug 25 11:15:21 EDT 2011
You are on the latest version.Diagnostics: Crash reporter
Processing…
Uploading...
minfree only report received. Skipping. This was fixed in a recent snapshot.
Continue and delete crash report files from local disk.Still the latest update doest resolve the problem.
-
Then it isn't actually making a crash dump. Get a picture or capture the text somehow of the panic again, and the output of "bt" at that prompt.
-
Please see attach picture of the bt result
I hope this can help you trace the problem
Thank you
-
What are you doing to trigger this?
-
This is the result of "bt" command
I just updated to august patches
Rc3 June released do not have this error. -
No i mean, what is happening in your box to trigger this?
I can assume a master failover from the backtrace but can you confirm that its happening automatically?
What kind of interface are you running carp over? -
I think i found why.
Can you test with tomorrows snapshots. -
Yes, I think you found it.
September 3 patched work.Many thanks.
