3:08 a.m.
Hello all,
I read at https://www.ovirt.org/develop/projects/node/troubleshooting/
that "Changes made from the command line are done at your own risk.
Making changes has the potential to leave your system in an unusable
state." It seems clear that RPMs should not be installed.
Is this accurate for https://www.ovirt.org/node/ ?
We have added smartctl and hpacucli in order to do disk and RAID
monitoring. So far our node servers have retained changes across
reboots, which is the primary reason I'm wondering if perhaps this
applies to an older version of oVirt Node.
If what we have been doing is not supported, what is the suggested
method to do do hardware monitoring (in particular disks)?
Thanks
Matt
1:06 p.m.
On Fri, Dec 15, 2017 at 4:45 AM, Donny Davis <donny(a)fortnebula.com> wrote:
That document mainly refers to vintage node.
In Next Generation Node now we have rpm persistence; please check
https://www.ovirt.org/develop/release-management/features/node/node-next-...
12:58 a.m.
This is a multi-part message in MIME format.
--------------331C82DB35A9B6B6239FAB57
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
On 12/15/2017 03:06 AM, Simone Tiraboschi wrote:
I'm sure glad we tested!
On one Node image we had images locally stored in /exports and shared
out via NFS. After an upgrade & reboot, images are gone.
If we "Convert to local storage" will the data persist? I am planning
to test, but want to be sure how this is designed.
I assume during a Gluster installation something is also updated in
oVirt Node to allow for the Gluster partition to persist?
At this point I'm thinking I should manually install via CentOS7 to
ensure folders and partitions are persistent. Is there any downside to
installing over CentOS7?
Thanks
Matt
--------------331C82DB35A9B6B6239FAB57
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: 8bit
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;
charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
On 12/15/2017 03:06 AM, Simone Tiraboschi wrote:<br>
<blockquote type="cite"
cite="mid:CAN8-ONoC1n9G6U7UPEqytzto0ToFbwbGhrk-W42EMHGWABpVbA@mail.gmail.com">
<div dir="ltr">On Fri, Dec 15, 2017 at 4:45 AM, Donny Davis
<span
dir="ltr"><<a
href="mailto:donny@fortnebula.com"
target="_blank"
moz-do-not-send="true">donny(a)fortnebula.com</a>&gt;</span>
wrote:<br>
<div class="gmail_extra">
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left:1px solid
rgb(204,204,204);padding-left:1ex">
<div dir="ltr">have you gotten an image update
yet?</div>
<div class="gmail-HOEnZb">
<div class="gmail-h5">
<div class="gmail_extra"><br>
<div class="gmail_quote">On Thu, Dec 14, 2017 at
8:08 PM, Matt Simonsen <span
dir="ltr"><<a
href="mailto:matt@khoza.com"
target="_blank"
moz-do-not-send="true">matt(a)khoza.com</a>&gt;</span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0px
0px 0px 0.8ex;border-left:1px solid
rgb(204,204,204);padding-left:1ex">Hello all,<br>
<br>
I read at <a
href="https://www.ovirt.org/develop/projects/node/troubleshooting/&q...
rel="noreferrer" target="_blank"
moz-do-not-send="true">https://www.ovirt.org/develop/<wbr...
that "Changes made from the command line are
done at your own risk. Making changes has the
potential to leave your system in an unusable
state." It seems clear that RPMs should not be
installed.<br>
</blockquote>
</div>
</div>
</div>
</div>
</blockquote>
<div><br>
</div>
<div>That document mainly refers to vintage node.</div>
<div>In Next Generation Node now we have rpm persistence;
please check</div>
<div><a
href="https://www.ovirt.org/develop/release-management/features/node...
moz-do-not-send="true">https://www.ovirt.org/develop/release...
</div>
<br>
</div>
</div>
</div>
</blockquote>
<br>
<br>
<br>
I'm sure glad we tested! <br>
<br>
On one Node image we had images locally stored in /exports and
shared out via NFS. After an upgrade & reboot, images are gone.<br>
<br>
If we "Convert to local storage" will the data persist? I am
planning to test, but want to be sure how this is designed.<br>
<br>
I assume during a Gluster installation something is also updated in
oVirt Node to allow for the Gluster partition to persist?<br>
<br>
At this point I'm thinking I should manually install via CentOS7 to
ensure folders and partitions are persistent. Is there any downside
to installing over CentOS7?<br>
<br>
Thanks<br>
Matt<br>
</body>
</html>
--------------331C82DB35A9B6B6239FAB57--
4:26 p.m.
Hello Matt,
All the partitions will be persisted when gluster is installed on
the ovirt node since gluster recommends user not to create bricks in root
directory. If the gluster bricks are created in root partition then once
the update of the node is done, you will not be able to see any of the
bricks.
Hope this helps !!!
Thanks
kasturi.
On Tue, Dec 19, 2017 at 4:28 AM, Matt Simonsen <matt(a)khoza.com> wrote:
3:37 p.m.
Next-gen node (since 4.0) doesn't have any strong notion of "persistence",
unlike previous versions of oVirt Node.
My guess would be that /exports was not moved across to the new system.
The update process for Node is essentially:
* create a new lv
* unpack the squashfs in the new RPM and put it on the LV
* Go through /var, /etc, and /root to see if any files were modified/added,
and copy those to the new layer
* Add a new bootloader entry
* Re-install any persistent RPMs
I would guess that /exports as not on its own partition, so it didn't carry
across. Mounting the LV for the old image somewhere (/tmp/...) will make
them visible again,
On Tue, Dec 19, 2017 at 9:26 AM, Kasturi Narra <knarra(a)redhat.com> wrote:
--
RYAN BARRY
SENIOR SOFTWARE ENGINEER - TEAM LEAD - RHEV HYPERVISOR
Red Hat NA <https://www.redhat.com/>
rbarry(a)redhat.com M: +1-651-815-9306 IM: rbarry
<https://red.ht/sig>
2647
days inactive
2674
days old
5 comments
5 participants
participants (5)
-
Donny Davis -
Kasturi Narra -
Matt Simonsen -
Ryan Barry -
Simone Tiraboschi