[DRBD-user] Crash during synchronisation when RATE >= 1024 ???

Lars Ellenberg Lars.Ellenberg at linbit.com
Thu Jun 10 11:40:04 CEST 2004

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


/ 2004-06-10 09:29:42 +0200
\ Benoit.Ropartz at alcatel.fr:
> Hi,
> 
> My Platform is Linux SML_B 2.4.21-rc5-20031126 #1 Tue Feb 17 18:41:17 MET
> 2004 ppc unknown (RS6000).
> 
> When the synchronisation use a rate < 1024 : there is no problem !!!
> 
> But when the rate is >= 1024 bytes there is a DRBD crash.
> 
> Syslog file :
> 
> drbd0: size = 847040 KB
> drbd0: 16384 KB marked out-of-sync by on disk bit-map.
> drbd0: Found 4 transactions (4 active extents) in activity log.
> drbd0: Marked additional 0 KB as out-of-sync based on AL.

> drbd0: Connection established.
> drbd0: Resync started as source (need to sync 16384 KB).
> Oops: kernel access of bad area, sig: 11
> NIP: D104CE50 XER: 00000000 LR: D104CE38 SP: CF65BF10 REGS: cf65be60 TRAP: 0300    Not tainted
> MSR: 00001032 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 11
> DAR: 00000000, DSISR: 42000000
> TASK = cf65a000[233] 'drbd0_receiver' Last syscall: 120
> last math cf38a000 last altivec 00000000
> GPR00: 00001032 CF65BF10 CF65A000 00000000 00001032 000004BA CE3DD154 CE3DD040
> GPR08: BA2E8BA3 00000000 00000000 CF65BF40 84000042 00000000 00000000 00000000
> GPR16: 00000000 00000000 00000000 00000000 00009032 0E5C7F40 00000000 C0005ED4
> GPR24: C0005BC0 10003F3C 00001000 00000100 C01F0000 CF65BF38 CFCFA420 CFCFA000
> Call backtrace:
> D104CB50 D104D034 D104EFBC D10500A8 D105077C D1045C6C C00088D8
> note: drbd0_receiver[233] exited with preempt_count 2

would you please pipe that through ksymoops?
otherwise it is just garbage.
--> see man ksymoops.

and you could:
 * try with latest cvs
   [but I doubt that there have been changes relevant to this problem]
 * try with preemt disabled

	lge



More information about the drbd-user mailing list