From: "Roger Pau Monné" <roger.pau@citrix.com>
To: Ian Campbell <ian.campbell@citrix.com>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
george.dunlap@eu.citrix.com, xen-devel@lists.xenproject.org,
wei.liu2@citrix.com
Subject: Re: Second regression due to libxl: Remove linux udev rules (2ba368d13893402b2f1fb3c283ddcc714659dd9b)
Date: Tue, 4 Aug 2015 11:44:43 +0200 [thread overview]
Message-ID: <55C0898B.5040605@citrix.com> (raw)
In-Reply-To: <1438677173.31129.41.camel@citrix.com>
El 04/08/15 a les 10.32, Ian Campbell ha escrit:
> On Tue, 2015-08-04 at 10:14 +0200, Roger Pau Monné wrote:
>> This was not a problem when using udev, because there's no timeout, but
>> libxl has a hard timeout (10s) regarding hotplug script execution. The
>> only way I see to solve this is to remove the checks done in the block
>> hotplug script, or to increase the timeout (but since the execution
>> time is not bounded this is doomed to fail if enough guests are using
>> the same image).
>
> If we have evidence (as we appear to) that the timeout is not sufficient
> for starting large numbers of guests in // then increasing the timeout
> would seem reasonable.
>
> Perhaps someone could set an insane timeout and measure the maximum time
> they see in practice in this test case. Then we could decide what a
> reasonable timeout should be? Maybe we could even find a linear pattern in
> the number of guests and use that to extrapolate a reasonable timeout?
I've created 40 HVM guest in parallel, all of them using the same file
as CD-ROM image. The highest block hotplug execution time I've measured
is 23s, so IMHO setting the timeout to 40s (or maybe 60s to really be on
the safe side?) seems reasonable. I will send a patch to do that ASAP.
Doing it based on the number of guests is not completely right, ideally
we should do it based on the number of guests using the same image file.
Here is the trace of execution times:
Time to execute: 0.00
Time to execute: 0.00
Time to execute: 0.00
Time to execute: 1.00
Time to execute: 0.00
Time to execute: 0.00
Time to execute: 1.00
Time to execute: 2.00
Time to execute: 1.00
Time to execute: 2.00
Time to execute: 2.00
Time to execute: 3.00
Time to execute: 4.00
Time to execute: 5.00
Time to execute: 7.00
Time to execute: 7.00
Time to execute: 9.00
Time to execute: 11.00
Time to execute: 13.00
Time to execute: 14.00
Time to execute: 16.00
Time to execute: 16.00
Time to execute: 12.00
Time to execute: 12.00
Time to execute: 11.00
Time to execute: 12.00
Time to execute: 12.00
Time to execute: 12.00
Time to execute: 14.00
Time to execute: 15.00
Time to execute: 15.00
Time to execute: 14.00
Time to execute: 19.00
Time to execute: 20.00
Time to execute: 19.00
Time to execute: 19.00
Time to execute: 21.00
Time to execute: 23.00
Time to execute: 21.00
Time to execute: 23.00
Roger.
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2015-08-04 9:44 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-28 19:47 Second regression due to libxl: Remove linux udev rules (2ba368d13893402b2f1fb3c283ddcc714659dd9b) Konrad Rzeszutek Wilk
2015-07-29 9:03 ` Ian Campbell
2015-07-29 10:52 ` Roger Pau Monné
2015-07-29 15:45 ` Konrad Rzeszutek Wilk
2015-07-30 8:17 ` Ian Campbell
2015-07-30 8:43 ` Roger Pau Monné
2015-07-30 8:56 ` Ian Campbell
2015-07-30 8:53 ` Roger Pau Monné
2015-08-04 8:14 ` Roger Pau Monné
2015-08-04 8:32 ` Ian Campbell
2015-08-04 9:44 ` Roger Pau Monné [this message]
2015-08-07 14:54 ` Konrad Rzeszutek Wilk
2015-08-07 14:58 ` Roger Pau Monné
2015-08-12 14:09 ` Konrad Rzeszutek Wilk
2015-08-18 7:49 ` Roger Pau Monné
2015-09-22 14:15 ` Konrad Rzeszutek Wilk
2015-08-11 8:52 ` Ian Campbell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=55C0898B.5040605@citrix.com \
--to=roger.pau@citrix.com \
--cc=george.dunlap@eu.citrix.com \
--cc=ian.campbell@citrix.com \
--cc=konrad.wilk@oracle.com \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.