[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] how to start VMs in a particular order

  • To: xen-users@xxxxxxxxxxxxx
  • From: "J. Roeleveld" <joost@xxxxxxxxxxxx>
  • Date: Sun, 29 Jun 2014 10:39:49 +0200
  • Delivery-date: Sun, 29 Jun 2014 08:41:22 +0000
  • List-id: Xen user discussion <xen-users.lists.xen.org>

On Sunday, June 29, 2014 10:09:32 AM lee wrote:
> "J. Roeleveld" <joost@xxxxxxxxxxxx> writes:
> >> On Saturday, June 28, 2014 01:11:05 PM lee wrote:
> >> 
> >> Apparently it's using arcconf to query the status of the
> >> volumes/controller.  Last crash, I got a message on the console saying
> >> that arcconf appeared to be hanging for 120 seconds, and the regularity
> >> of the crashes was suspicious.  The software is ancient, so it's
> >> possible that it causes problems with somewhat recent kernels :(
> > 
> > Quite possibly. Am I correct in assuming you are using old hardware with
> > closed-source software?
> It's an IBM x3650 7979 L2G with a ServeRaid 8k.  Arcconf seems to be
> closed source --- I don't really need arcconf, though.
> Unfortunately, disabling the status checking hasn't solved the problem.
> The server goes down with messages about the SCSI bus hanging and trying
> to reset it.  I suspect that the controller doesn't like the --- rather
> unsuited --- WD20EARS I plugged in.  They have been working fine with a
> HP smart array P800, though.  I might have to take them out to see if
> the problem persists.

SCSI bus hanging, sounds like an I/O issue.
Try to read the SMART-values of the disk. Also, try a different disk...

The WD20EARS is a "green" desktop disk. I had numerous issues when using a 
couple of those in my old server when using software raid (mdadm).
Some hardware raid cards do not like disks that do not properly return error-
states. And especially the green disks that have a tendency to go into 
powersave mode when not used for a short period of time.
The raid-card can easily end up trying to throw that disk out of the raid-
array. If that is the only disk, that will mean the disk suddenly disappeared, 
causing kernel panics.

I currently use WD Red drives with hardware raid cards.

> >> >> On Debian, /etc/init.d/xendomains seems to be what brings up the
> >> >> domains
> >> >> automatically.  It would be a good place to add a timer or to add
> >> >> checks
> >> >> for services.
> >> > 
> >> > That script is part of Xen, not just Debian. I have the same file in
> >> > Gentoo.
> >> 
> >> Hm, is it still there when systemd is used?
> > 
> > When using systemd, it is not used.
> > However, I see more issues with using systemd instead of openrc when
> > compared to the benefits.
> I'm not a fan of systemd --- it's more complicated, and where are
> advantages?

To some people, there are. To others, there are more disadvantages.
But, this is the wrong place to go into the differences.

> > I use Xen on servers where stability is more important then a fast
> > boottime. (especially as the BIOS takes longer then booting the OS)
> Well, I wish the server was running stable!

See my comment about your disk above. Replace it or connect it directly to the 
mainboard, bypassing the raid controller.

> > All the domUs have their console inside a screen-session. These also log
> > the output to:
> > /var/log/xen-consoles/<domu-name>.log
> > 
> > By checking if these have the login prompt, you can also ensure the domU
> > has started correctly. At least the scripts I get with Gentoo cycle when
> > the screen-session is created.
> Hmmm ... That is really going to lengths.

Many roads lead to Rome :)
Likewise, many ways exist to achieve what you (and I) want. I do not know of 
an existing tool that does this simply. On a different list, people talk about 
using puppet or adding additional scripts as dependencies.


Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.