PE2850 and MPTFusion error messages

Dale Blount dell-lists at dale.us
Wed Oct 18 12:57:40 CDT 2006


Hello List,

I have a 2850 with 2 rpm 146gb Maxtor drives (in a linux-md raid1) and 4
146gb Seagate drives (in a linux-md raid0).  It's running a fully up to
date copy of Centos 4.4 i386. Whenever I start a proprietary app with
either a partition on the raid1 or one from the raid0 (both mounted in
the same subdir of /var), I end up getting these errors in dmesg:


mptbase: ioc0: IOCStatus(0x0003): Invalid SGL
mptbase: ioc0: LogInfo(0x11070000): F/W: DMA Error
mptbase: ioc0: LogInfo(0x11070000): F/W: DMA Error
mptbase: ioc0: IOCStatus(0x004b): SCSI IOC Terminated
mptbase: ioc0: LogInfo(0x11070000): F/W: DMA Error
mptbase: ioc0: IOCStatus(0x004b): SCSI IOC Terminated
mptbase: ioc0: IOCStatus(0x004b): SCSI IOC Terminated
mptbase: ioc0: IOCStatus(0x0003): Invalid SGL
mptbase: ioc0: LogInfo(0x11070000): F/W: DMA Error


I can run bonnie++ on these partitions and the errors don't show up.
Could this application be causing scsi errors?  It really shouldn't be
doing anything that aggressive either (it's a web cache/filter app). 

The errors don't seem to affect anything, but I did have a problem where
the software raid1 marked 2 partitions as dirty and needed to be
rebuilt.  I can't seem to get that to happen again, though.

Does anyone have any ideas of what this means before I contact the
software vendor or Dell?  If absolutely needed, I guess I could try
installing a copy of RHEL to see if the problem persists there.

Thanks,

Dale



More information about the Linux-PowerEdge mailing list