megaraid (PERC 4e/DC) resets under I/O load

Kilian CAVALOTTI kilian.cavalotti at lip6.fr
Thu May 17 13:22:12 CDT 2007


Hi all, 
 
I have two PE 1950 sharing a PV220S setup in cluster mode, using PERC 4/eDC 
cards. PERC 4e/DC firmware is 522A, and kernel modules are from 
megaraid-v2.20.4.4-2dkms, which I believe are the latest. The virtual disk  
is a 3 disks Raid5 volume.
 
Under I/O load, I got the following errors: 
megaraid: aborting-59364 cmd=2a <c=2 t=0 l=0> 
megaraid abort: 59364:6[255:128], fw owner 
megaraid: reseting the host... 
megaraid: 2 outstanding commands. Max wait 180 sec 
megaraid mbox: Wait for 2 commands to complete:180 
megaraid mbox: reset sequence completed sucessfully 
megaraid: fast sync command timed out 
megaraid: reservation reset failed 
megaraid: reseting the host... 
megaraid mbox: reset sequence completed sucessfully 
megaraid: fast sync command timed out 
megaraid: reservation reset failed 
megaraid: reseting the host... 
megaraid mbox: reset sequence completed sucessfully 
megaraid: fast sync command timed out 
megaraid: reservation reset failed 
SCSI error : <1 2 0 0> return code = 0x6000000 
end_request: I/O error, dev sdb, sector 293683517 
scsi1 (0:0): rejecting I/O to offline device 
scsi1 (0:0): rejecting I/O to offline device 
Buffer I/O error on device sdb2, logical block 88269 
lost page write due to I/O error on sdb2 
Aborting journal on device sdb2. 
journal commit I/O error 
journal commit I/O error 
 
I don't know if that's related, but the controller logs show the following: 
05/16/07 19:02:24: EVT#03560-05/16/07 19:02:24: 113=Unexpected sense: PD 08
(e1/s255), CDB: 12 01 80 00 3a 00, Sense: 70 00 05 00 00 00 00 0a 00 00 00 
00 24 00 00 00 00 00 
 
I thought this issue has been solved by previous firmware releases, but it 
doesn't seem to be the case.  
 
I'd appreciate any hint regarding this issue. 
Thanks a lot, 
-- 
Kilian



More information about the Linux-PowerEdge mailing list