Proxmox 1.6, 2.6.32 kernel and mptsas driver

[sebastien.k]

Hi,
I've upgraded to proxmox 1.6 and 2.6.32 kernel. All works fine at beginning but after 10 minutes raid array starting degraded and after some hours I have :

Personalities : [raid1] [raid0] [raid6] [raid5] [raid4] 
md2 : active raid6 sda5[0] sdl5[11] sdk5[10] sdj5[9] sdi5[8] sdh5[12](F) sdg5[13](F) sdf5[5] sde5[4] sdd5[14](F) sdc5[2] sdb5[1]
      19319823360 blocks level 6, 64k chunk, algorithm 2 [12/9] [UUU_UU__UUUU]
      
md1 : active raid1 sda1[12](F) sdl1[11] sdk1[13](F) sdj1[14](F) sdg1[15](F) sdi1[16](F) sdh1[17](F) sdf1[18](F) sde1[19](F) sdc1[20](F) sdb1[21](F)
      20480896 blocks [12/1] [___________U]

/var/log/messages show me some problems with mptsas driver (v. 3.04.12) and kernel > 2.6.28.

Sep 10 18:15:56 ns60875 kernel: sd 0:0:3:0: [sdd] CDB: ATA command pass  through(16): 85 08 0e 00 d5 00 01 00 09 00 4f 00 c2 00 b0 00 
Sep 10 18:15:57 ns60875 kernel: mptscsih: ioc0: target reset: SUCCESS  (sc=ffff8803590fc700) 
Sep 10 18:16:00 ns60875 kernel: mptbase: ioc0: LogInfo(0x31111000):  Originator={PL}, Code={Reset}, SubCode(0x1000) 
Sep 10 18:16:00 ns60875 kernel: mptbase: ioc0: LogInfo(0x31112000):  Originator={PL}, Code={Reset}, SubCode(0x2000) 
Sep 10 18:16:07 ns60875 kernel: mptscsih: ioc0: attempting task abort!  (sc=ffff8803590fc700) 
Sep 10 18:16:07 ns60875 kernel: sd 0:0:3:0: [sdd] CDB: Test Unit Ready:  00 00 00 00 00 00 
Sep 10 18:16:07 ns60875 kernel: mptscsih: ioc0: task abort: SUCCESS  (sc=ffff8803590fc700) 
Sep 10 18:16:07 ns60875 kernel: mptscsih: ioc0: attempting bus reset!  (sc=ffff8803590fc700) 
Sep 10 18:16:07 ns60875 kernel: sd 0:0:3:0: [sdd] CDB: ATA command pass  through(16): 85 08 0e 00 d5 00 01 00 09 00 4f 00 c2 00 b0 00 
Sep 10 18:16:07 ns60875 kernel: mptbase: ioc0: LogInfo(0x31112000):  Originator={PL}, Code={Reset}, SubCode(0x2000) 
Sep 10 18:16:09 ns60875 kernel: mptscsih: ioc0: bus reset: SUCCESS  (sc=ffff8803590fc700) 
Sep 10 18:16:12 ns60875 kernel: mptbase: ioc0: LogInfo(0x31111000):  Originator={PL}, Code={Reset}, SubCode(0x1000) 
Sep 10 18:16:12 ns60875 kernel: mptbase: ioc0: LogInfo(0x31112000):  Originator={PL}, Code={Reset}, SubCode(0x2000) 
Sep 10 18:16:19 ns60875 kernel: sd 0:0:3:0: [sdd] Device not ready 
...
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: WARNING - IOC is in FAULT  state (7820h)!!! 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: WARNING - Issuing  HardReset from mpt_fault_reset_work!! 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: Initiating recovery 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: WARNING - IOC is in FAULT  state!!! 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: WARNING - FAULT code = 7820h 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: Recovered from IOC FAULT 
Sep 10 18:17:22 ns60875 kernel: mptbase: ioc0: WARNING -  mpt_fault_reset_work: HardReset: success

This bug is also knowed and there is a patch (http://lkml.org/lkml/2010/6/3/398).

This would be included in next version on pve kernel ?

Thanks
Sébastien

 

[dietmar]

Please can you test with:

ftp://download.proxmox.com/debian/d...4/pve-kernel-2.6.32-3-pve_2.6.32-15_amd64.deb

 

[kro]

I confirm it works perfectly on an identical server as the above mentioned.

Thanks!