<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hi Jay,<br>
<br>
<blockquote type="cite">
<div>We have had similar issues with extending attached volumes
that are iSCSI based. In that case the VM has to be forced to
rescan the scsi bus. </div>
<div><br>
</div>
<div>In this case I am not sure if there needs to be a change to
Libvirt or to rbd or something else. </div>
<div><br>
</div>
<div>I would recommend reaching out to John Bernard for help. </div>
</blockquote>
<br>
In fact, I'm ok with delayed resize (upon power-cycle), and it's not
an issue for me that VM don't detect changes immediately. What I
want to understand is that changes to Cinder (and, thus, underlying
changes to CEPH) are safe for VM while it's in active state.<br>
<br>
Hopefully, Jon will help with this question.<br>
<br>
Thank you!<br>
<br>
<div class="moz-cite-prefix">On 6/23/18 8:41 PM, Jay Bryant wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CA+0S5sRskKYEoee1uMrVAeST4XAa+y5WHzqD0pPa+UZmrVjpAA@mail.gmail.com"><br>
<br>
<div class="gmail_quote">
<div dir="ltr">On Sat, Jun 23, 2018, 9:39 AM Volodymyr Litovka
<<a href="mailto:doka.ua@gmx.com" moz-do-not-send="true">doka.ua@gmx.com</a>>
wrote:<br>
<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">Dear
friends,<br>
<br>
I did some tests with making volume available without stopping
VM. I'm <br>
using CEPH and these steps produce the following results:<br>
<br>
1) openstack volume set --state available [UUID]<br>
- nothing changed inside both VM (volume is still connected)
and CEPH<br>
2) openstack volume set --size [new size] --state in-use
[UUID]<br>
- nothing changed inside VM (volume is still connected and has
an old size)<br>
- size of CEPH volume changed to the new value<br>
3) during these operations I was copying a lot of data from
external <br>
source and all md5 sums are the same on both VM and source<br>
4) changes on VM happens upon any kind of power-cycle (e.g.
reboot <br>
(either soft or hard): openstack server reboot [--hard] [VM
uuid] )<br>
- note: NOT after 'reboot' from inside VM<br>
<br>
It seems, that all these manipilations with cinder just update
internal <br>
parameters of cinder/CEPH subsystems, without immediate effect
for VMs. <br>
Is it safe to use this mechanism in this particular environent
(e.g. <br>
CEPH as backend)?<br>
<br>
From practical point of view, it's useful when somebody, for
example, <br>
update project in batch mode, and will then manually reboot
every VM, <br>
affected by the update, in appropriate time with minimized
downtime <br>
(it's just reboot, not manual stop/update/start).<br>
<br>
Thank you.<br>
<br>
-- <br>
Volodymyr Litovka<br>
"Vision without Execution is Hallucination." -- Thomas
Edison</blockquote>
</div>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
Volodymyr Litovka
"Vision without Execution is Hallucination." -- Thomas Edison</pre>
</body>
</html>