6.2-R on Dell Poweredge 2950 with Dell PERC 5/i [mfi(4)]
Scott Long
scottl at samsco.org
Fri May 11 19:56:14 UTC 2007
David Wolfskill wrote:
> On Thu, May 10, 2007 at 05:14:01PM -0600, Scott Long wrote:
>> ...
>> Not sure that this impression is entirely accurate. The biggest problem
>> with MFI machines is online RAID management. The storage driver itself
>> matured very quickly and has been very reliable.
>
> Ah; good to know: thank you.
>
>>> Well, now a colleague is trying to run 6.2-R on one of these 2950s; dmesg
>>> says the controller is:
>>>
>>> mfi0: <Dell PERC 5/i> mem 0xd80f0000-0xd80fffff,0xfc4e0000-0xfc4fffff irq
>>> 78 at device 14.0 on pci2
>> ...
>>> and the disks looks like:
>>>
>>> mfid0: <MFI Logical Disk> on mfi0
>>> mfid0: 418176MB (856424448 sectors) RAID volume '' is optimal
>>>
>> Looks A OK to me.
>
> Even better. :-)
>
>>> The intended production workload involves creation and deletion of
>>> a large number of files rather rapidly.
>> ...
>> sysctl vfs.ffs.doasyncfree=0 might help. Running the syncer more
>> frequently might also help, but I don't recall the sysctl node for
>> that.
>
> OK; I've relayed your suggestion to my colleague, but haven't heard back
> from her yet.
>
>> ...
>> Very strange. No chance that it was due to files that were deleted but
>> still referenced by open apps?
>
> I don't think so. She's deployed 13 other boxen over the last few years
> with -- naturally! -- different hardware specs, but all running
> essentailly the same application.
>
> The big question for her is whether or not the Dell 2950, as specified,
> will do the job.
>
>> ...
>> This sounds purely like a filesystem issue, not an MFI driver issue.
>
> Hmmm... I'll admit to knowing little about RAID configurations; is it
> possible that some RAID configurations might exacerbate problems with
> such a workload -- or that others might be more amenable to it?
>
If anything, a fast RAID controller will help reduce the lag that you
get when the syncer does its periodic run. But beyond that, I can't
think of anything that would cause problems.
Scott
More information about the freebsd-stable
mailing list