[ovirt-devel] [VDSM] All tests using directio fail on CI

Yaniv Kaul ykaul at redhat.com
Thu Sep 29 08:08:29 UTC 2016


zram does not support direct IO (tested, indeed fails).
What I do is host the VMs there, though - this is working - but I'm using
Lago (and not oVirt). does oVirt need direct IO for the temp disks? I
thought we are doing them on the libvirt level?

This is the command I use:
sudo modprobe zram num_devices=1 && sudo zramctl --find --size 12G && sudo
mkfs.xfs -K /dev/zram0 && sudo mount -o nobarrier /dev/zram0 /home/zram &&
sudo chmod 777 /home/zram

And then I run lago with: ./run_suite.sh -o /home/zram basic_suite_master

Y.


On Thu, Sep 29, 2016 at 10:47 AM, Evgheni Dereveanchin <ederevea at redhat.com>
wrote:

> Hi Yaniv,
>
> this is a physical server with work directories
> created on a zRAM device, here's the patch:
> https://gerrit.ovirt.org/#/c/62249/2/site/ovirt_jenkins_
> slave/templates/prepare-ram-disk.service.erb
>
> I'll still need to read up on this but the only
> slave having this class (ovirt-srv08) is now offline
> and should not cause issues. I tested on VM slaves
> and did not see errors from the dd test command you provided.
>
> Please tell me if you see errors on other nodes and I'll
> check what's going on but it must be something else than RAM disks.
>
> Regards,
> Evgheni Dereveanchin
>
> ----- Original Message -----
> From: "Yaniv Kaul" <ykaul at redhat.com>
> To: "Evgheni Dereveanchin" <ederevea at redhat.com>
> Cc: "infra" <infra at ovirt.org>, "devel" <devel at ovirt.org>, "Eyal Edri" <
> eedri at redhat.com>, "Nir Soffer" <nsoffer at redhat.com>
> Sent: Thursday, 29 September, 2016 9:32:45 AM
> Subject: Re: [ovirt-devel] [VDSM] All tests using directio fail on CI
>
> On Sep 29, 2016 10:28 AM, "Evgheni Dereveanchin" <ederevea at redhat.com>
> wrote:
> >
> > Hi,
> >
> > Indeed the proposed dd test does not work on zRAM slaves.
> > Can we modify the job not to run on nodes with ram_disk label?
>
> Are those zram based or ram based *virtio-blk* disks,  or zram/ram disks
> within the VMs?
> The former should work. The latter -  no idea.
>
> >
> > The node will be offline for now until we agree on what to do.
> > An option is to abandon RAM disks completely as we didn't find
> > any performance benefits from using them so far.
>
> That's very surprising. On my case it doubles the performance,  at least.
> But I assume my storage (single disk) is far slower than yours.
> Y.
>
> >
> > Regards,
> > Evgheni Dereveanchin
> >
> > ----- Original Message -----
> > From: "Eyal Edri" <eedri at redhat.com>
> > To: "Nir Soffer" <nsoffer at redhat.com>, "Evgheni Dereveanchin" <
> ederevea at redhat.com>
> > Cc: "Yaniv Kaul" <ykaul at redhat.com>, "devel" <devel at ovirt.org>, "infra"
> <
> infra at ovirt.org>
> > Sent: Thursday, 29 September, 2016 8:08:45 AM
> > Subject: Re: [ovirt-devel] [VDSM] All tests using directio fail on CI
> >
> > Evgheni,
> > Can you try switching the current RAM drive with zram?
> >
> > On Wed, Sep 28, 2016 at 11:43 PM, Nir Soffer <nsoffer at redhat.com> wrote:
> >
> > > On Wed, Sep 28, 2016 at 11:39 PM, Yaniv Kaul <ykaul at redhat.com> wrote:
> > > > On Sep 28, 2016 11:37 PM, "Nir Soffer" <nsoffer at redhat.com> wrote:
> > > >>
> > > >> On Wed, Sep 28, 2016 at 11:20 PM, Nir Soffer <nsoffer at redhat.com>
> > > wrote:
> > > >> > On Wed, Sep 28, 2016 at 10:31 PM, Barak Korren <
> bkorren at redhat.com>
> > > >> > wrote:
> > > >> >> The CI setup did not change recently.
> > > >> >
> > > >> > Great
> > > >> >
> > > >> >> All standard-CI jobs run inside mock (chroot) which is stored on
> top
> > > >> >> of a regular FS, so they should not be affected by the slave OS
> at
> > > all
> > > >> >> as far as FS settings go.
> > > >> >>
> > > >> >> But perhaps some slave-OS/mock-OS combination is acting
> strangely, so
> > > >> >> could you be more specific and point to particular job runs that
> > > fail?
> > > >> >
> > > >> > This jobs failed, but it was deleted (I get 404 now):
> > > >> > http://jenkins.ovirt.org/job/vdsm_master_check-patch-fc24-
> > > x86_64/2530/
> > > >>
> > > >> Oops, wrong build.
> > > >>
> > > >> This is the failing build:
> > > >>
> > > >>
> > > >> http://jenkins.ovirt.org/job/vdsm_master_check-patch-el7-
> > > x86_64/1054/console
> > > >>
> > > >> And this is probably the reason - using a ram disk:
> > > >>
> > > >> 12:24:53 Building remotely on ovirt-srv08.phx.ovirt.org (phx
> physical
> > > >> integ-tests ram_disk fc23) in workspace
> > > >> /home/jenkins/workspace/vdsm_master_check-patch-el7-x86_64
> > > >>
> > > >> We cannot run the storage tests using a ramdisk. We are creating
> > > >> (tiny) volumes and storage domains and doing copies, this code
> cannot
> > > >> work with ramdisk.
> > > >
> > > > Will it work on zram?
> > > > What if we configure ram based iSCSI targets?
> > >
> > > I don't know, but it is easy to test - it this works the tests will
> work:
> > >
> > > dd if=/dev/zero of=file bs=512 count=1 oflag=direct
> > > _______________________________________________
> > > Infra mailing list
> > > Infra at ovirt.org
> > > http://lists.ovirt.org/mailman/listinfo/infra
> > >
> > >
> > >
> >
> >
> > --
> > Eyal Edri
> > Associate Manager
> > RHV DevOps
> > EMEA ENG Virtualization R&D
> > Red Hat Israel
> >
> > phone: +972-9-7692018
> > irc: eedri (on #tlv #rhev-dev #rhev-integ)
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/infra/attachments/20160929/3587a81a/attachment.html>


More information about the Infra mailing list