Bug#281372: Crashed again despite firmware upgrade (#281372)
Damn, I really must learn how to use the BTS properly... resending to the BTS.
As mentioned in the earlier email, I flashed the RAID card to the latest Dell
firmware. The server bailed last night. From kern.log on a remote logging host:
Jan 25 00:54:47 tardis kernel: aacraid: Host adapter reset request. SCSI hang ?
Jan 25 00:54:47 tardis kernel: scsi: device set offline - command error
recover failed: host 0 channel 0 id 0 lun 0
Jan 25 00:54:47 tardis kernel: I/O error: dev 08:06, sector 1506680
Jan 25 00:54:47 tardis kernel: SCSI disk error : host 0 channel 0 id 0 lun 0
return code = 6000000
Jan 25 00:54:47 tardis kernel: I/O error: dev 08:06, sector 1953513
Jan 25 00:54:47 tardis kernel: I/O error: dev 08:06, sector 1953520
Jan 25 00:54:47 tardis kernel: I/O error in filesystem ("sd(8,6)") meta-data
dev sd(8,6) block 0x1dcee9 ("xlog_iodone") error 5 buf count 3584
Jan 25 00:54:47 tardis kernel: I/O error: dev 08:06, sector 1506680
Jan 25 00:54:47 tardis kernel: xfs_force_shutdown(sd(8,6),0x2) called from
line 959 of file xfs_log.c. Return address = 0xf89274aa
$ lspci -vvv
00:04.0 RAID bus controller: Digital Equipment Corporation DECchip 21554 (rev 01)
Subsystem: Adaptec Dell PowerEdge RAID Controller 2
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr+
Stepping- SERR+ FastB2B-
Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 128, cache line size 08
Interrupt: pin A routed to IRQ 61
Region 0: Memory at f4000000 (32-bit, non-prefetchable) [size=8K]
Region 1: I/O ports at 1000 [size=256]
Expansion ROM at <unassigned> [disabled] [size=128K]
Capabilities: <available only to root>
$ cat /proc/scsi/aacraid/0
Adaptec Raid Controller 1.1-3 Nov 14 2004 11:01:31, scsi hba number 0
kernel: 2.8-4[6089]
monitor: 2.8-4[6089]
bios: 2.8-0[6089]
serial: 895e87fafaf001
$ uname -a
Linux tardis 2.4.27-ainet-p3-smp #1 SMP Sun Nov 14 10:54:19 GMT 2004 i686
GNU/Linux
Built from kernel-source-2.4.27 (2.4.27-5).
I'll build a kernel from the latest debian 2.4.27 (2.4.27-8) and reboot this
evening, though bear in mind that the server has taken up to ~40 days to crash
in the past.
Yell if you need more info.
Ronny
--
Technical Director
Amazing Internet Ltd, London
t: +44 20 8607 9535
f: +44 20 8607 9536
w: www.amazinginternet.com
Reply to: