Volume autogrow and message "Write failed due to lack of space"

christoph_reeg · ‎2013-01-15

Hello,

I've tested a LUN in a volume with auto grow option enabled.

When writing to this LUN normally everything is fine and as expected, the volume gets full and then the size increases:

Jan 4 11:19:00 [monitor.volume.full:debug]: /vol/VOLUME is full (using or reserving 100% of space and 0% of inodes, using 100% of reserve).

Jan 4 11:19:00 [monitor.globalStatus.nonCritical:warning]: /vol/VOLUME is full (using or reserving 100% of space and 0% of inodes, using 100% of reserve).

Jan 4 11:19:01 [wafl.vol.autoSize.done:info]: Automatic increase size of volume 'VOLUME' by 524280 kbytes done.

Jan 4 11:20:05 [wafl.vol.autoSize.done:info]: Automatic increase size of volume 'VOLUME' by 524280 kbytes done.

Jan 4 11:20:19 [wafl.vol.autoSize.done:info]: Automatic increase size of volume 'VOLUME' by 524280 kbytes done.

But sometime I get error messages:

Jan 4 11:18:46 [wafl.vol.full:notice]: Insufficient space on volume VOLUME to perform operation. 32.0KB was requested but only 20.0KB was available.

Jan 4 11:18:47 [scsitarget.write.failureNoSpace:error]: Write to LUN /vol/VOLUME/lun/lun101 failed due to lack of space.

Jan 4 11:18:47 [wafl.vol.autoSize.done:info]: Automatic increase size of volume 'VOLUME' by 524280 kbytes done.

On the client I didn't see an error, but I'm always getting a little bit nervous when the storage produces error messages...

Does anyone know this message? Is there a problem or is everything fine?

Thanks,

Christoph

martin_fisher · ‎2013-01-15

Hi - The syslog/console messages are showing that ONTAP is trying to write the data to the LUN as normal, but cant strip the data down due to insufficient space. ONTAP/WAFL then increases the size of the volume with the autogrown parameters set so it could continue service. If these were not set and the LUN ran out of space, the LUN would be auto dismounted to protect and preserve the data existing in the LUN. One way to test this, if you are unsure is turn off all the autogrow settings and start writing data to the LUN.

martin

aborzenkov · ‎2013-01-15

What is not clear here, was write request retried or data is lost.

Отправлено с iPhone

15.01.2013, в 20:43, "Martin Fisher" <xdl-communities@communities.netapp.com<mailto:xdl-communities@communities.netapp.com>> написал(а):

<https://communities.netapp.com/index.jspa>

Re: Volume autogrow and message "Write failed due to lack of space"

created by Martin Fisher<https://communities.netapp.com/people/martin.fisher> in Data ONTAP - View the full discussion<https://communities.netapp.com/message/98288#98288>

Hi - The syslog/console messages are showing that ONTAP is trying to write the data to the LUN as normal, but cant strip the data down due to insufficient space. ONTAP/WAFL then increases the size of the volume with the autogrown parameters set so it could continue service. If these were not set and the LUN ran out of space, the LUN would be auto dismounted to protect and preserve the data existing in the LUN. One way to test this, if you are unsure is turn off all the autogrow settings and start writing data to the LUN.

martin

Reply to this message by replying to this email -or- go to the message on NetApp Community<https://communities.netapp.com/message/98288#98288>

Start a new discussion in Data ONTAP by email<mailto:discussions-community-products_and_solutions-data_ontap@communities.netapp.com> or at NetApp Community<https://communities.netapp.com/choose-container.jspa?contentType=1&containerType=14&container=2877>

christoph_reeg · ‎2013-01-16

On the client I don't see an error. It seems that the write request is retried although the message on NetApp sounds for me that the write request is lost ("Write to LUN /vol/VOLUME/lun/lun101 failed due to lack of space. ").

martin_fisher · ‎2013-01-16

Only engineering would be able to confirm for sure on this one.. open a support case >?

christoph_reeg · ‎2013-01-16

Thanks for your answer.

I already opened a case and got the answer "No problem, when the LUN gets full, you get this message and after that the volume grows".

But IMHO they didn't understand my question why the "write failed" message occurs only sometimes and not every time.

I hoped that someone of you has seen this message in this context and knows more.

martin_fisher · ‎2013-01-18

Ask for the case to be escalated to enquire whether this fact is indeed true, as NetApp Support have confirmed. A different support engineer sometimes gives a different response.