[zfs-discuss] Task blocked for more than 120 seconds

Bogdan bogdar at gmail.com
Sun Dec 4 06:31:19 EST 2016


How does this 'task hang' is related to hard drive model?

On Tue, Nov 22, 2016 at 11:39 AM, Hofmann, Timo via zfs-discuss <
zfs-discuss at list.zfsonlinux.org> wrote:

> Just to add a positive note on the WD disks:
>
> I run three ZFS systems with 24-80 RE4 disks or its successors (2 and
> 6TB), the 2TB disks being in use for three years, so far only one 6TB disk
> died all others run fine. All systems are used for Veeam and dirvish
> (rsync) backups. All have quite some load, especially one which gets a lot
> of read/write action, but no load like you'd get from running VMs.
>
> Timo
>
> -----Urspr√ľngliche Nachricht-----
> Von: zfs-discuss [mailto:zfs-discuss-bounces at list.zfsonlinux.org] Im
> Auftrag von Jan Schermer via zfs-discuss
> Gesendet: Samstag, 19. November 2016 01:37
> An: zfs-discuss at list.zfsonlinux.org; Lindsay Mathieson <
> lindsay.mathieson at gmail.com>
> Betreff: Re: [zfs-discuss] Task blocked for more than 120 seconds
>
> That's... strange.
> Reds are meant for NAS - i.e. non-seeky or interactive use I'm pretty sure
> that what I now found as "WD Yellow" used to be Black, and I know I read
> that Blacks are in fact RE4 that are going to be discontinued... yet I see
> "yellow" WD RE as a datacenter drive and it looks like they reshuffled it
> all...
>
> We use 2.5" WD REDs with great success, but all our 3.5" RED deployments
> sucked so bad we switched to Blacks where possible. I must get some Yellows
> to investigate more...
>
> Jan
>
>
> > On 19 Nov 2016, at 01:22, Lindsay Mathieson via zfs-discuss <
> zfs-discuss at list.zfsonlinux.org> wrote:
> >
> > On 19/11/2016 9:11 AM, Jan Schermer via zfs-discuss wrote:
> >>  Fixed by replacing them with Blacks...
> >
> > We use blacks initially for our ceph cluster (VM Hosting via RBD). They
> all died in under a year. They just can't cope with server levels of 24/7
> usage.
> >
> >
> > Not having any problem with the IOPS on Red for our current cluster.
> >
> >
> >
> > --
> > Lindsay Mathieson
> >
> > _______________________________________________
> > zfs-discuss mailing list
> > zfs-discuss at list.zfsonlinux.org
> > http://list.zfsonlinux.org/cgi-bin/mailman/listinfo/zfs-discuss
>
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at list.zfsonlinux.org
> http://list.zfsonlinux.org/cgi-bin/mailman/listinfo/zfs-discuss
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at list.zfsonlinux.org
> http://list.zfsonlinux.org/cgi-bin/mailman/listinfo/zfs-discuss
>



-- 
WBR,  Bogdan B. Rudas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.zfsonlinux.org/pipermail/zfs-discuss/attachments/20161204/a02a5706/attachment.html>


More information about the zfs-discuss mailing list