<div dir="ltr">Matt,<div><br></div><div>Yes, Sean already opened a critical bug - <a href="https://bugs.launchpad.net/oslo.reports/+bug/1510740">https://bugs.launchpad.net/oslo.reports/+bug/1510740</a></div><div><br></div><div>Please note that this change was added to oslo.reports *after* the liberty release in support of Mitaka. So i am not sure why we need to add it to <span style="font-size:12.8000001907349px">liberty release notes. Also this is a consequence of NOT having version caps which was a decision a while ago as well.</span></div><div><br></div><div>-- DIms</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Oct 30, 2015 at 4:47 AM, Matt Riedemann <span dir="ltr"><<a href="mailto:mriedem@linux.vnet.ibm.com" target="_blank">mriedem@linux.vnet.ibm.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5"><br>
<br>
On 10/21/2015 7:43 AM, Kashyap Chamarthy wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Background<br>
----------<br>
<br>
Oslo Guru Meditation (error) Reports (GMR)[*] are a useful debugging<br>
mechanism that allows one to capture the current state of a Nova<br>
process/executable (e.g. `nova-compute`, `nova-api`, etc).<br>
<br>
The way to generate the error report is to supply the 'User-defined<br>
signal', SIGUSR1, when killing a Nova process. E.g.<br>
<br>
$ kill -USR1 `pgrep nova-compute`<br>
<br>
which results in GMR being printed to your standard error ('stderr')<br>
stream, wherever it ends up being redirected to (e.g. to a corresponding<br>
Nova process-specific log file, otherwise, on systemd-enabled systems,<br>
to its journal).<br>
<br>
<br>
Change in Mitaka (and above)<br>
----------------------------<br>
<br>
From the upcoming Mitaka release onwards, the default expected UNIX<br>
signal to generate GMR has been changed[1] from USR1 to USR2 (another<br>
User-defined singal), because the USR1 is reserved by Apache 'mod_wsgi'<br>
for its own purpose.<br>
<br>
So, to generate GMR, from Mitaka release:<br>
<br>
$ kill -USR2 `pgrep nova-compute`<br>
<br>
A corresponding Nova documentation change[2] has been submitted to<br>
reflect this new reality.<br>
<br>
<br>
[1] <a href="https://review.openstack.org/#/c/223133/" rel="noreferrer" target="_blank">https://review.openstack.org/#/c/223133/</a> -- guru_meditation_report:<br>
Use SIGUSR2 instead of SIGUSR1<br>
[2] <a href="https://review.openstack.org/#/c/227779/" rel="noreferrer" target="_blank">https://review.openstack.org/#/c/227779/</a> -- doc: gmr: Update<br>
instructions to generate GMR error reports<br>
<br>
<br>
[*] References<br>
--------------<br>
<br>
Related reading:<br>
<br>
- <a href="http://docs.openstack.org/developer/nova/gmr.html" rel="noreferrer" target="_blank">http://docs.openstack.org/developer/nova/gmr.html</a><br>
- <a href="http://docs.openstack.org/developer/oslo.reports/usage.html" rel="noreferrer" target="_blank">http://docs.openstack.org/developer/oslo.reports/usage.html</a><br>
- <a href="https://wiki.openstack.org/wiki/GuruMeditationReport" rel="noreferrer" target="_blank">https://wiki.openstack.org/wiki/GuruMeditationReport</a><br>
- <a href="https://www.berrange.com/posts/2015/02/19/nova-and-its-use-of-olso-incubator-guru-meditation-reports/" rel="noreferrer" target="_blank">https://www.berrange.com/posts/2015/02/19/nova-and-its-use-of-olso-incubator-guru-meditation-reports/</a><br>
<br>
</blockquote>
<br></div></div>
Looks like this broke some tooling in the gate job runs where gmr's are created at the end of the service logs when the services exit. Here is a mitaka change with grenade on the liberty side with the gmr at the end:<br>
<br>
<a href="http://logs.openstack.org/97/227897/13/check/gate-grenade-dsvm/27723a9/logs/old/screen-n-cpu.txt.gz" rel="noreferrer" target="_blank">http://logs.openstack.org/97/227897/13/check/gate-grenade-dsvm/27723a9/logs/old/screen-n-cpu.txt.gz</a><br>
<br>
And on the new side it's gone:<br>
<br>
<a href="http://logs.openstack.org/97/227897/13/check/gate-grenade-dsvm/27723a9/logs/new/screen-n-cpu.txt.gz" rel="noreferrer" target="_blank">http://logs.openstack.org/97/227897/13/check/gate-grenade-dsvm/27723a9/logs/new/screen-n-cpu.txt.gz</a><br>
<br>
So obviously an upgrade impact, I'm hoping we get this into the liberty release notes as something to change when people move up to oslo.reports 1.6.0.<br>
<br>
We should also get the gate tooling fixed around this, I'm not sure where that was configured/triggered though, sdague probably knows.<span class="HOEnZb"><font color="#888888"><br>
<br>
-- <br>
<br>
Thanks,<br>
<br>
Matt Riedemann</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Davanum Srinivas :: <a href="https://twitter.com/dims" target="_blank">https://twitter.com/dims</a></div>
</div>