[OpenIndiana-discuss] Simple zfs vs zpool space question.

Michael Hase michael at edition-software.de
Fri Apr 11 16:06:24 UTC 2014


John,

On Fri, 11 Apr 2014, John McEntee wrote:

> Michael,
>
> You had me worried then an doubt my own sanity. I just remove the 2 drives and re-added them.

Well, no intend to worry people. But in this case I think it's better to 
look a bit closer. Again: are you sure the 2 drives where really removed 
when you told them to? What did you do to remove them?

To my knowledge the block pointer rewrite project was never addressed, 
even Matt Ahrens said it's too difficult. So you can't remove a data 
device from a zpool.

If your zpool iostat output really resembles the actual situation, I think 
your only option is to attach additional disks to the stale devices.

Cheers,
Michael

>
> #  zpool add tank log  c1t500A075103053202d0p2
> #  zpool add tank log c1t500A07510306F9A7d0p2
> # zpool iostat -v
>                              capacity     operations    bandwidth
> pool                       alloc   free   read  write   read  write
> -------------------------  -----  -----  -----  -----  -----  -----
> rpool                      13.2G  26.5G      0     19  3.01K  89.2K
>  mirror                   13.2G  26.5G      0     19  3.01K  89.2K
>    c1t5000CCA216DA22DEd0p1      -      -      0      9  6.33K  89.6K
>    c1t500A075103053202d0s0      -      -      0     10  5.60K  89.6K
>    c1t500A07510306F9A7d0s0      -      -      0     10  4.59K  89.6K
> -------------------------  -----  -----  -----  -----  -----  -----
> tank                       13.9T  5.10T    106  1.36K  5.23M  9.47M
>  mirror                   1.99T   746G     15    199   765K  1.35M
>    c1t5000CCA225C5244Ed0      -      -      3     86   350K  1.35M
>    c1t5000CCA225C54DDDd0      -      -      3     86   353K  1.35M
>    c1t5000CCA225C505B8d0      -      -      3     86   350K  1.35M
>  mirror                   1.99T   746G     15    199   766K  1.35M
>    c1t5000CCA225C50784d0      -      -      3     86   349K  1.35M
>    c1t5000CCA225C5502Ed0      -      -      3     86   352K  1.35M
>    c1t5000CCA225C49869d0      -      -      3     86   352K  1.35M
>  mirror                   1.99T   746G     15    199   766K  1.35M
>    c1t5000CCA225C54ED8d0      -      -      3     86   351K  1.35M
>    c1t5000CCA225C56814d0      -      -      3     86   351K  1.35M
>    c1t5000CCA225C4E775d0      -      -      3     86   350K  1.35M
>  mirror                   1.99T   746G     15    199   765K  1.35M
>    c1t5000CCA225C2ADDAd0      -      -      3     85   351K  1.35M
>    c1t5000CCA225C04039d0      -      -      3     85   352K  1.35M
>    c1t5000CCA225C53428d0      -      -      3     85   352K  1.35M
>  mirror                   1.99T   746G     15    199   766K  1.35M
>    c1t5000CCA225C50517d0      -      -      3     85   352K  1.36M
>    c1t5000CCA225C55025d0      -      -      3     85   352K  1.36M
>    c1t5000CCA225C5660Dd0      -      -      3     85   351K  1.36M
>  mirror                   1.99T   745G     15    199   764K  1.35M
>    c1t5000CCA225C5502Dd0      -      -      3     85   350K  1.35M
>    c1t5000CCA225C484A3d0      -      -      3     85   351K  1.35M
>    c1t5000CCA225C4824Dd0      -      -      3     85   352K  1.35M
>  mirror                   1.99T   746G     15    199   766K  1.35M
>    c1t5000CCA225C4E366d0      -      -      3     86   351K  1.35M
>    c1t5000CCA225C54DDCd0      -      -      3     85   352K  1.35M
>    c1t5000CCA225C56751d0      -      -      3     85   351K  1.35M
>  c1t500A075103053202d0p2  3.35M  7.93G      0     98  4.46K  1.32M
>  c1t500A07510306F9A7d0p2  3.37M  7.93G      0     22  12.3K   464K
> cache                          -      -      -      -      -      -
>  c1t500A075103053202d0p3   143G  7.88M     21      4   464K   487K
>  c1t500A07510306F9A7d0p3   143G  7.88M     21      4   471K   488K
> -------------------------  -----  -----  -----  -----  -----  -----
>
> #
>
> And it is still not labelling them as logs on the output, I assume this is due to running oi_148 unless I have done something wrong.
>
> I know the logs are not a mirror, but I am going for performance over the risk of an ssd failure at the time of a reboot, at which I believe I can just throw again the ZIL and continue? According to http://docs.oracle.com/cd/E19253-01/819-5461/ghbxs/
>
> Thanks
>
> John
>
>
>
> -----Original Message-----
> From: Michael Hase [mailto:michael at edition-software.de]
> Sent: 11 April 2014 15:59
> To: Discussion list for OpenIndiana
> Subject: Re: [OpenIndiana-discuss] Simple zfs vs zpool space question.
>
> On Fri, 11 Apr 2014, John McEntee wrote:
>
>> I have a zpool  (3way mirror and striped across 21 disks), tank, that the zfs and zpool commands differ slightly.
>
> Are you sure it's a 3way mirror pool? Seems like you're running without any redundancy, look at devices c1t500A075103053202d0p2 and c1t500A07510306F9A7d0p2. If one of these breaks I think your pool is toast.
>
> Maybe you wanted to add zil devices and forgot to specify as such? It should look something like
>
> zpool iostat -v p1
>                 capacity     operations    bandwidth
> pool         alloc   free   read  write   read  write
> -----------  -----  -----  -----  -----  -----  -----
> p1            876G   980G      3     51   106K   295K
>   mirror      529G   399G      1     18  47.8K  73.7K
>     c4t1d0       -      -      0      8  42.0K  74.5K
>     c4t2d0       -      -      0      8  42.1K  74.5K
>   mirror      347G   581G      1     30  58.7K   141K
>     c4t0d0       -      -      0     12  42.6K   142K
>     c4t3d0       -      -      0     12  42.7K   142K
> logs             -      -      -      -      -      -
>   c3t12d0s3  1.20M  7.94G      0      1      0  80.6K
> cache            -      -      -      -      -      -
>   c3t12d0s0  48.0G     8M      1      0  13.9K  20.7K
> -----------  -----  -----  -----  -----  -----  -----
>
>>
>> # zpool list tank
>> NAME   SIZE  ALLOC   FREE    CAP  DEDUP  HEALTH  ALTROOT
>> tank  19.0T  14.5T  4.51T    76%  1.00x  ONLINE  -
>>
>> # zfs list tank
>> NAME   USED  AVAIL  REFER  MOUNTPOINT
>> tank  14.5T  4.21T    38K  /tank
>>
>>
>> There are no quotas. Please could some suggest why zpool states more space is available over zfs?
>
> Metadata?
>
> Cheers,
> Michael
>
>>
>> My average IOPS to each single SATA disk is 90, 1390 across the pool as a whole. Due to pushing each disk fairly hard , if the zpool gets more than 80% full the way free space is allocated changes and we have a severe performance hit. This makes (generally one) VMWare host virtually unusable until I free up some space. Does anyone know if this behaviour changes due to the result of zpool or zfs. i.e. which one should I carefully monitor?
>>
>> Thanks
>>
>> John
>>
>> P.S.
>>
>> # zpool iostat -v
>>                              capacity     operations    bandwidth
>> pool                       alloc   free   read  write   read  write
>> -------------------------  -----  -----  -----  -----  -----  -----
>> rpool                      13.2G  26.5G      0     19  3.01K  89.1K
>>  mirror                   13.2G  26.5G      0     19  3.01K  89.1K
>>    c1t5000CCA216DA22DEd0p1      -      -      0      9  6.30K  89.4K
>>    c1t500A075103053202d0s0      -      -      0     10  5.57K  89.4K
>>    c1t500A07510306F9A7d0s0      -      -      0     10  4.56K  89.4K
>> -------------------------  -----  -----  -----  -----  -----  -----
>> tank                       14.5T  4.51T    106  1.39K  5.23M  10.1M
>>  mirror                   2.07T   659G     15    199   765K  1.35M
>>    c1t5000CCA225C5244Ed0      -      -      3     86   350K  1.35M
>>    c1t5000CCA225C54DDDd0      -      -      3     86   353K  1.35M
>>    c1t5000CCA225C505B8d0      -      -      3     86   350K  1.35M
>>  mirror                   2.07T   659G     15    199   765K  1.35M
>>    c1t5000CCA225C50784d0      -      -      3     86   349K  1.35M
>>    c1t5000CCA225C5502Ed0      -      -      3     86   352K  1.35M
>>    c1t5000CCA225C49869d0      -      -      3     86   352K  1.35M
>>  mirror                   2.07T   659G     15    199   766K  1.35M
>>    c1t5000CCA225C54ED8d0      -      -      3     86   351K  1.35M
>>    c1t5000CCA225C56814d0      -      -      3     86   351K  1.35M
>>    c1t5000CCA225C4E775d0      -      -      3     86   350K  1.35M
>>  mirror                   2.08T   659G     15    199   765K  1.35M
>>    c1t5000CCA225C2ADDAd0      -      -      3     85   351K  1.35M
>>    c1t5000CCA225C04039d0      -      -      3     85   352K  1.35M
>>    c1t5000CCA225C53428d0      -      -      3     85   352K  1.35M
>>  mirror                   2.07T   659G     15    199   766K  1.35M
>>    c1t5000CCA225C50517d0      -      -      3     85   352K  1.36M
>>    c1t5000CCA225C55025d0      -      -      3     85   352K  1.36M
>>    c1t5000CCA225C5660Dd0      -      -      3     85   351K  1.36M
>>  mirror                   2.08T   659G     15    199   764K  1.35M
>>    c1t5000CCA225C5502Dd0      -      -      3     85   350K  1.35M
>>    c1t5000CCA225C484A3d0      -      -      3     85   351K  1.35M
>>    c1t5000CCA225C4824Dd0      -      -      3     85   352K  1.35M
>>  mirror                   2.07T   659G     15    199   766K  1.35M
>>    c1t5000CCA225C4E366d0      -      -      3     86   351K  1.35M
>>    c1t5000CCA225C54DDCd0      -      -      3     85   352K  1.35M
>>    c1t5000CCA225C56751d0      -      -      3     85   351K  1.35M
>>  c1t500A075103053202d0p2  6.89M  7.93G      0     15      1   298K
>>  c1t500A07510306F9A7d0p2  6.73M  7.93G      0     15      1   298K
>> cache                          -      -      -      -      -      -
>>  c1t500A075103053202d0p3   143G  7.86M     20      4   463K   488K
>>  c1t500A07510306F9A7d0p3   143G  7.75M     21      4   470K   488K
>> -------------------------  -----  -----  -----  -----  -----  -----
>>
>> ______________________________________________________________________
>> _
>>
>> The contents of this e-mail and any attachment(s) are strictly
>> confidential and are solely for the person(s) at the e-mail
>> address(es) above. If you are not an addressee, you may not disclose,
>> distribute, copy or use this e-mail, and we request that you send an
>> e-mail to admin at stirling-dynamics.com and delete this e-mail.
>> Stirling Dynamics Ltd. accepts no legal liability for the contents of
>> this e-mail including any errors, interception or interference, as
>> internet communications are not secure.  Any views or opinions
>> presented are solely those of the author and do not necessarily
>> represent those of Stirling Dynamics Ltd. Registered In England No.
>> 2092114 Registered Office: 26 Regent Street, Clifton, Bristol. BS8 4HG
>> VAT no. GB 464 6551 29
>> ______________________________________________________________________
>> _
>>
>> This e-mail has been scanned for all viruses MessageLabs.
>> _______________________________________________
>> 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
>
> _______________________________________________________________________
>
> The contents of this e-mail and any attachment(s) are strictly confidential and are solely for the person(s) at the e-mail address(es) above. If you are not an addressee, you may not disclose, distribute, copy or use this e-mail, and we request that you send an e-mail to admin at stirling-dynamics.com and delete this e-mail.  Stirling Dynamics Ltd. accepts no legal liability for the contents of this e-mail including any errors, interception or interference, as internet communications are not secure.  Any views or opinions presented are solely those of the author and do not necessarily represent those of Stirling Dynamics Ltd. Registered In England No. 2092114 Registered Office: 26 Regent Street, Clifton, Bristol. BS8 4HG
> VAT no. GB 464 6551 29
> _______________________________________________________________________
>
> This e-mail has been scanned for all viruses MessageLabs.
>
> _______________________________________________
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss
>



More information about the OpenIndiana-discuss mailing list