Ask help on an issue 'XFS (sd*): discard failed for extent, error 5'
Jevon Qiao
qiaojianfeng at unitedstack.com
Tue Sep 1 20:33:45 CDT 2015
On 1/9/15 20:24, Eric Sandeen wrote:
> On 9/1/15 12:49 AM, Jevon Qiao wrote:
>> Hi Eric
>>
>> Thank you very much for the prompt response.
>> On 1/9/15 12:28, Eric Sandeen wrote:
>>> On 8/31/15 10:55 PM, Jevon Qiao wrote:
>>>> Hi All,
>>>>
>>>> Currently, I notice that there are lot's of errors written in the /var/log/messages file.
>>>>
>>>> 2015-08-29T23:58:54.477236+00:00 server-68 kernel: sd 0:0:4:0: [sdc]
>>>> 2015-08-29T23:58:54.477255+00:00 server-68 kernel: Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
>>>> 2015-08-29T23:58:54.477256+00:00 server-68 kernel: sd 0:0:4:0: [sdc]
>>>> 2015-08-29T23:58:54.477257+00:00 server-68 kernel: Sense Key : Illegal Request [current]
>>>> 2015-08-29T23:58:54.477258+00:00 server-68 kernel: sd 0:0:4:0: [sdc]
>>>> 2015-08-29T23:58:54.477259+00:00 server-68 kernel: Add. Sense: Invalid field in parameter list
>>>> 2015-08-29T23:58:54.477260+00:00 server-68 kernel: sd 0:0:4:0: [sdc] CDB:
>>>> 2015-08-29T23:58:54.477261+00:00 server-68 kernel: Unmap/Read sub-channel: 42 00 00 00 00 00 00 00 18 00
>>>> 2015-08-29T23:58:54.477263+00:00 server-68 kernel: end_request: critical target error, dev sdc, sector 325046824
>>>> 2015-08-29T23:58:54.477268+00:00 server-68 kernel: XFS (sdc2): discard failed for extent [0x64815,2021], error 5
>>>>
>>>> Has anyone encountered the same issue before? How can you solve it?
>>> Looks like xfs is simply reacting to errors from the storage; I'd try to get
>>> to the bottom of why you're getting all those scsi errors.
>> I just want to understand the meaning of 'XFS (sdc2): discard failed
>> for extent [0x64815,2021], error 5', does it mean one of the layers
>> beneath xfs does not support DISCARD? Could you please show me more
>> clues?
> "error 5" is EIO. That's as much as I can tell you from the XFS perspective.
>
> I don't know what the block/scsi errors mean, exactly; it does seem as
> if it may not support discard, but it must have advertised that it did, or
> the discard would not have been issued in the first place.
Thank you for the further help on this.
> You haven't said what kernel or storage you're on.
> OTOH, even if you do, it's a scsi/storage issue that the xfs list probably
> can't help much with.
OK, I see.
Thanks,
Jevon
> -Eric
>
>> Thanks,
>> Jevon
>>> But probably not on the xfs list. :)
>>>
>>> -Eric
>>>
>>>> Thanks,
>>>> Jevon
>>>>
>>>>
>>>> _______________________________________________
>>>> xfs mailing list
>>>> xfs at oss.sgi.com
>>>> http://oss.sgi.com/mailman/listinfo/xfs
>>>>
>>> _______________________________________________
>>> xfs mailing list
>>> xfs at oss.sgi.com
>>> http://oss.sgi.com/mailman/listinfo/xfs
>>
>>
More information about the xfs
mailing list