[OpenIndiana-discuss] MPT SGL mem alloc failed

[Albert Lee]

On Thu, Jul 12, 2012 at 5:41 PM, Timothy Coalson <tsc5yc at mst.edu> wrote:
> After rebooting to oi_151a4, and scrub, clear, scrub, the pool says
> that it is completely healthy, no data errors, no device errors, no
> repaired bytes (on the second scrub that is, on the first scrub there
> were enough checksum errors for it to report all of the devices in one
> vdev as degraded).  Additionally, no mpt_sas errors in
> /var/adm/messages.  Created bug #2994 for it:
> https://www.illumos.org/issues/2994
>
> Not sure if newer firmware on the HBAs would help, but I will wait for
> someone to pick up the bug before trying anything.
>
> Tim
>

The DMA alloc failure appears to be caused by improperly constrained
kernel memory usage. Probably unrelated to the driver in question, but
could be related to recent ZFS changes. Perhaps ::memstat prior to the
import and scrub of the pool in question may show something
significant.

-Albert

> On Mon, Jul 9, 2012 at 5:29 PM, Timothy Coalson <tsc5yc at mst.edu> wrote:
>> Well, I don't have the same symptoms as him, his swap was almost entirely used:
>>
>> tim at myelin2:/var/adm$ swap -lh
>> swapfile             dev    swaplo   blocks     free
>> /dev/zvol/dsk/rpool/swap 96,2        4K      12G      12G
>>
>> Though I am not logged into a desktop (it is sitting at the gdm
>> greeter).  This machine also doesn't have nvidia graphics, so can't
>> compare there, either.
>>
>> Tim
>>
>> On Mon, Jul 9, 2012 at 5:09 PM, Bill Sommerfeld <sommerfeld at alum.mit.edu> wrote:
>>> On 07/09/12 13:17, Timothy Coalson wrote:
>>>> I upgraded a machine to oi_151a5 from oi_151a4 last week, and when its
>>>> weekly scrub rolled around, /var/adm/messages gathered a lot of these,
>>>> in groups of dozens at a time:
>>>>
>>>> Jul  7 01:15:21 myelin2 scsi: [ID 107833 kern.warning] WARNING: /pci at 0,0/pci8086
>>>> ,340a at 3/pci1000,30c0 at 0 (mpt_sas0):
>>>> Jul  7 01:15:21 myelin2         Unable to allocate dma memory for extra SGL.
>>>> Jul  7 01:15:21 myelin2 scsi: [ID 107833 kern.warning] WARNING: /pci at 0,0/pci8086
>>>> ,340a at 3/pci1000,30c0 at 0 (mpt_sas0):
>>>> Jul  7 01:15:21 myelin2         MPT SGL mem alloc failed
>>>
>>> so this may be completely unrelated but over on on oi-dev, someone recently
>>> reported a different driver-related kernel memory allocation issue with the
>>> nvidia driver with oi151a5 which he did not see with oi151a3 / 0.151.1.4
>>>
>>> See http://openindiana.org/pipermail/oi-dev/2012-July/001541.html
>>>
>>> Coincidence?  Another manifestation of the same underlying bug?
>>>
>>>
>>>
>>> _______________________________________________
>>> OpenIndiana-discuss mailing list
>>> OpenIndiana-discuss at openindiana.org
>>> http://openindiana.org/mailman/listinfo/openindiana-discuss
>
> _______________________________________________
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss