About "panic: bufwrite: buffer is not busy???"

[Andrew Boyer]

Moving this to -current and -stable and following up...

Something is broken with coredumps on stable/8 amd64.  I tried a vanilla 8.2-RC3 and yesterday's csup of stable/8; neither can dump a core with 'sysctl debug.kdb.panic=1'.

For the 8.2-RC3 / amd64 / GENERIC install, I used the memstick image, installed on ad7 (a 250GB SATA drive), used the default partition map, and set dumpdev to AUTO.

I added enough tracing to show that the second panic is due to the syncer process flushing buffers to the other filesystems in parallel with the dump.  I've seen this panic and a similar one 'buffer not locked' coming from ffs_write().  One time out of about 30 the core ran to completion, but slowly (~1MB/sec).  Other times the dump just locks up completely with no other output.

Does anyone know what might have changed to expose this problem?

I don't ever see it under 7.1.

Thanks,
 Andrew

On Feb 3, 2011, at 12:11 AM, Eugene Grosbein wrote:

> On 02.02.2011 00:50, Gleb Smirnoff wrote:
> 
>> E> Uptime: 8h3m51s
>> E> Dumping 4087 MB (3 chunks)
>> E>   chunk 0: 1MB (150 pages) ... ok
>> E>   chunk 1: 3575MB (915088 pages) 3559 3543panic: bufwrite: buffer is not busy???
>> E> cpuid = 3
>> E> Uptime: 8h3m52s
>> E> Automatic reboot in 15 seconds - press a key on the console to abort
>> Can you add KDB_TRACE option to kernel? Your boxes for some reason can't
>> dump core, but with this option we will have at least trace.
> 
> I see Mike Tancsa's box has "bufwrite: buffer is not busy???" problem too.
> Has anyone a thought how to fix generation of crashdumps?
> 
> Eugene Grosbein
> 
> 
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"

--------------------------------------------------
Andrew Boyer	aboyer at averesystems.com