Date: Tue, 8 Mar 2016 11:50:01 +0800 From: Fred Liu <fred.fliu@gmail.com> To: Julian Elischer <julian@freebsd.org> Cc: illumos-zfs <zfs@lists.illumos.org>, Discussion list for OpenIndiana <openindiana-discuss@openindiana.org>, omnios-discuss <omnios-discuss@lists.omniti.com>, developer <developer@open-zfs.org>, "zfs-devel@freebsd.org" <zfs-devel@freebsd.org>, illumos-developer <developer@lists.illumos.org>, "freebsd-fs@FreeBSD.org" <freebsd-fs@freebsd.org>, "smartos-discuss@lists.smartos.org" <smartos-discuss@lists.smartos.org>, "zfs-discuss@list.zfsonlinux.org" <zfs-discuss@list.zfsonlinux.org> Subject: Re: [zfs] an interesting survey -- the zpool with most disks you have ever built Message-ID: <CALi05Xz8R2nnwM%2BD_3wuQLeg3Hc9g6ROfRn2H9KczMeeAz0nDg@mail.gmail.com> In-Reply-To: <56DD17CA.90200@freebsd.org> References: <95563acb-d27b-4d4b-b8f3-afeb87a3d599@me.com> <CACTb9pxJqk__DPN_pDy4xPvd6ETZtbF9y=B8U7RaeGnn0tKAVQ@mail.gmail.com> <CAJjvXiH9Wh%2BYKngTvv0XG1HtikWggBDwjr_MCb8=Rf276DZO-Q@mail.gmail.com> <56D87784.4090103@broken.net> <A5A6EA4AE9DCC44F8E7FCB4D6317B1D203178F1DD392@SH-MAIL.ISSI.COM> <CAOjFWZ5YcaAx-v5ZqsoFnHFB1jnvstpXpGObcfewMx75WU0TeQ@mail.gmail.com> <A5A6EA4AE9DCC44F8E7FCB4D6317B1D203178F1DD39E@SH-MAIL.ISSI.COM> <CAOjFWZ7E-LTvUy60UTe2Yi2Egw6%2BbrKZx3r70UbtJJ9haNL5Hg@mail.gmail.com> <CALi05Xwc3dKTsyuaSLeVQSptMp537XeLxXf6Pj%2B15jRtXKXCfA@mail.gmail.com> <CAOjFWZ6YvtpBf2J9F6OTGLh0UfRuBxiY6iF-gNFNAhv=QCB7YQ@mail.gmail.com> <CALi05Xyy3voKVHTR=bHSG5JszQBW4NC0=XL_C-YTQdwzBPwnag@mail.gmail.com> <56DD17CA.90200@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
2016-03-07 13:55 GMT+08:00 Julian Elischer <julian@freebsd.org>:
> On 6/03/2016 9:30 PM, Fred Liu wrote:
>>
>> 2016-03-05 0:01 GMT+08:00 Freddie Cash <fjwcash@gmail.com>:
>>
>>> On Mar 4, 2016 2:05 AM, "Fred Liu" <fred.fliu@gmail.com> wrote:
>>>>
>>>> 2016-03-04 13:47 GMT+08:00 Freddie Cash <fjwcash@gmail.com>:
>>>>>
>>>>> Currently, I just use a simple coordinate system. Columns are letters,
>>>
>>> rows are numbers.
>>>>>
>>>>> "smartos-discuss@lists.smartos.org" <smartos-discuss@lists.smartos.org
>>>>
>>>> 、
>>>
>>> developer <developer@open-zfs.org>、
>>>
>>> illumos-developer <developer@lists.illumos.org>、
>>>
>>> omnios-discuss <omnios-discuss@lists.omniti.com>、
>>>
>>> Discussion list for OpenIndiana <openindiana-discuss@openindiana.org>、
>>>
>>> illumos-zfs <zfs@lists.illumos.org>、
>>>
>>> "zfs-discuss@list.zfsonlinux.org" <zfs-discuss@list.zfsonlinux.org>、
>>>
>>> "freebsd-fs@FreeBSD.org" <freebsd-fs@freebsd.org>、
>>>
>>> "zfs-devel@freebsd.org" <zfs-devel@freebsd.org>
>>>
>>>>> Each disk is partitioned using GPT with the first (only) partition
>>>
>>> starting at 1 MB and covering the whole disk, and labelled with the
>>> column/row where it is located (disk-a1, disk-g6, disk-p3, etc).
>>>>
>>>> [Fred]: So you manually pull off all the drives one by one to locate
>>>
>>> them?
>>>
>>> When putting the system together for the first time, I insert each disk
>>> one at a time, wait for it to be detected, partition it, then label it
>>> based on physical location. Then do the next one. It's just part of the
>>> normal server build process, whether it has 2 drives, 20 drives, or 200
>>> drives.
>>>
>>> We build all our own servers from off-the-shelf parts; we don't buy
>>> anything pre-built from any of the large OEMs.
>>>
>> [Fred]: Gotcha!
>>
>>
>>>>> The pool is created using the GPT labels, so the label shows in "zpool
>>>
>>> list" output.
>>>>
>>>> [Fred]: What will the output look like?
>>>
>>> From our smaller backups server, with just 24 drive bays:
>>>
>>> $ zpool status storage
>>>
>>> pool: storage
>>>
>>> state: ONLINE
>>>
>>> status: Some supported features are not enabled on the pool. The pool can
>>>
>>> still be used, but some features are unavailable.
>>>
>>> action: Enable all features using 'zpool upgrade'. Once this is done,
>>>
>>> the pool may no longer be accessible by software that does not support
>>>
>>> the features. See zpool-features(7) for details.
>>>
>>> scan: scrub canceled on Wed Feb 17 12:02:20 2016
>>>
>>> config:
>>>
>>>
>>> NAME STATE READ WRITE CKSUM
>>>
>>> storage ONLINE 0 0 0
>>>
>>> raidz2-0 ONLINE 0 0 0
>>>
>>> gpt/disk-a1 ONLINE 0 0 0
>>>
>>> gpt/disk-a2 ONLINE 0 0 0
>>>
>>> gpt/disk-a3 ONLINE 0 0 0
>>>
>>> gpt/disk-a4 ONLINE 0 0 0
>>>
>>> gpt/disk-a5 ONLINE 0 0 0
>>>
>>> gpt/disk-a6 ONLINE 0 0 0
>>>
>>> raidz2-1 ONLINE 0 0 0
>>>
>>> gpt/disk-b1 ONLINE 0 0 0
>>>
>>> gpt/disk-b2 ONLINE 0 0 0
>>>
>>> gpt/disk-b3 ONLINE 0 0 0
>>>
>>> gpt/disk-b4 ONLINE 0 0 0
>>>
>>> gpt/disk-b5 ONLINE 0 0 0
>>>
>>> gpt/disk-b6 ONLINE 0 0 0
>>>
>>> raidz2-2 ONLINE 0 0 0
>>>
>>> gpt/disk-c1 ONLINE 0 0 0
>>>
>>> gpt/disk-c2 ONLINE 0 0 0
>>>
>>> gpt/disk-c3 ONLINE 0 0 0
>>>
>>> gpt/disk-c4 ONLINE 0 0 0
>>>
>>> gpt/disk-c5 ONLINE 0 0 0
>>>
>>> gpt/disk-c6 ONLINE 0 0 0
>>>
>>> raidz2-3 ONLINE 0 0 0
>>>
>>> gpt/disk-d1 ONLINE 0 0 0
>>>
>>> gpt/disk-d2 ONLINE 0 0 0
>>>
>>> gpt/disk-d3 ONLINE 0 0 0
>>>
>>> gpt/disk-d4 ONLINE 0 0 0
>>>
>>> gpt/disk-d5 ONLINE 0 0 0
>>>
>>> gpt/disk-d6 ONLINE 0 0 0
>>>
>>> cache
>>>
>>> gpt/cache0 ONLINE 0 0 0
>>>
>>> gpt/cache1 ONLINE 0 0 0
>>>
>>>
>>> errors: No known data errors
>>>
>>> The 90-bay systems look the same, just that the letters go all the way to
>>> p (so disk-p1 through disk-p6). And there's one vdev that uses 3 drives
>>> from each chassis (7x 6-disk vdev only uses 42 drives of the 45-bay
>>> chassis, so there's lots of spares if using a single chassis; using two
>>> chassis, there's enough drives to add an extra 6-disk vdev).
>>>
>> [Fred]: It looks like the gpt label shown in "zpool status" only works in
>> FreeBSD/FreeNAS. Are you using FreeBSD/FreeNAS? I can't find the similar
>> possibilities in Illumos/Linux.
>
>
> Ah that's a trick.. FreeBSD exports an actual /dev/gpt/{you-label-goes-here}
> for each labeled partition it finds.
> So it's not ZFS doing anything special.. it's what FreeBSD is calling the
> partition.
>
Super cool!
Fred
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CALi05Xz8R2nnwM%2BD_3wuQLeg3Hc9g6ROfRn2H9KczMeeAz0nDg>