Tony Byrne wrote:
Folks,
We have a 4.10-STABLE production server which has an Intel SRCU42X
RAID controller installed:
amr0: <LSILogic MegaRAID> mem 0xfe580000-0xfe5fffff,0xfbef0000-0xfbefffff irq
22 at device 0.0 on pci4
amr0: <LSILogic Intel(R) RAID Controller SRCU42X> Firmware 411M, BIOS H404,
128MB RAM
The server crashed yesterday in the small hours of the morning and
when we arrived on site to reboot it, there was a "bad slot" kernel
message on the console, which places the RAID controller in the frame.
The amr driver man page says that this message is indicative of a firmware or
hardware problem with the controller, but we are not convinced. We
experienced the same message and lockups daily during stress testing of the
box under FreeBSD 5.3 and this ultimately forced us to 'downgrade'
to 4.10 for production. The box had been rock solid under 4.10 for a
number of weeks before yesterday's crash.
Could this indicate a bug in the driver, or at least in its support
for our re-badged RAID controller? Has anyone else had problems with
the amr driver with the same card?
Many thanks,
Regards,
Tony.
I've been seeing this problem recently too. I believe that there is
some sort of timing bug/race in the driver, but I haven't been able to
figure it out yet. It also seems to be related to panic from the block
layer that point to commands being completed twice. To be clear with
your observations, are you saying that 4.10-RELEASE is behaving the same
or differently than 4.10-STABLE?
Scott
_______________________________________________
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
