opendev/system-config
Code Issues Proposed changes

add a logs directory help file

Browse Source 
this adds more details on debugging at a log level, it only currently
has tips for nova and cinder on the hope that others will add in
tips for their services as well.

Change-Id: I056979344ba8903420b8b096298cd76f6a5c75ce
Reviewed-on: https://review.openstack.org/33691
Approved: Jeremy Stanley <fungi@yuggoth.org>
Reviewed-by: Jeremy Stanley <fungi@yuggoth.org>
Tested-by: Jenkins
This commit is contained in:
Sean Dague 2013-06-19 15:55:22 -04:00 committed by Jenkins
parent 39019818f6
commit eb955c541e
2 changed files with 53 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
modules/openstack_project/files/logs/help/tempest-logs.html Normal file
View File

@ -0,0 +1,50 @@
<h1>Guide to Devstack Gate Logs</h1>
<p>
Above is a collection of log files from
the <a href="../console.html">current tempest run</a>. Within them
should be everything you need to get to the bottom of a test
failure. The screen-* logs will be your most valuable tools in this
process. Use the timestamp of a the failed test in
the <a href="../console.html">current tempest run</a>.
</p>
<h2>Nova Compute Fails</h2>
<p>
If there is a compute test failure, especially a server not getting
created correctly, or being in an unexpected state, the following is
typically the most fruitful order to look at things:
<ul>
<li><a href="screen-n-api.txt.gz">screen-n-api.txt.gz</a> - the nova
api log, which will show top level failures. Make sure the request
that was being sent in actually suceeded.
<li><a href="screen-n-cpu.txt.gz">screen-n-cpu.txt.gz</a> - the nova
compute log. If a libvirt or qemu issue happened during guest
creation it will be here.
<li><a href="screen-n-sch.txt.gz">screen-n-sch.txt.gz</a> - the nova
scheduler. Some times there are races in allocating resources, and
the scheduler will throw a WARNING if it couldn't allocate the
requested resources.
<li>all other nova logs
</ul>
<h2>Cinder Volume Fails</h2>
<p>
If there is a volume failure in the test, the following is typically
the most fruitful order to look at things:
<ul>
<li><a href="screen-c-api.txt.gz">screen-c-api.txt.gz</a> - the
cinder api log, which will show top level failures. Make sure
the request that was being sent in actually suceeded.
<li><a href="screen-c-vol.txt.gz">screen-c-vol.txt.gz</a> - the
cinder agent log. If there was a local allocation error it will be
here.
<li><a href="screen-c-sch.txt.gz">screen-c-sch.txt.gz</a> - the
cinder scheduler. Some times there are races in allocating
resources, and the scheduler will throw a WARNING if it couldn't
allocate the requested resources.
</ul>
</p>
<h2>About this Help</h2>
<p>
This help file is part of the openstack-infra/config project, and
can be updated by pushing a review request to
<a href="https://github.com/openstack-infra/config/tree/master/modules/openstack_project/files/logs/help">https://github.com/openstack-infra/config/tree/master/modules/openstack_project/files/logs/help</a>
</p>

3
modules/openstack_project/templates/logs.vhost.erb
View File

@ -30,6 +30,9 @@ NameVirtualHost <%= vhost_name %>:<%= port %>
<Directory <%= docroot %>/*/*/*/gate-tempest-devstack*/*> <Directory <%= docroot %>/*/*/*/gate-tempest-devstack*/*>
ReadmeName /help/tempest-overview.html ReadmeName /help/tempest-overview.html
</Directory> </Directory>
<Directory <%= docroot %>/*/*/*/gate-tempest-devstack*/*/logs/>
ReadmeName /help/tempest-logs.html
</Directory>
ErrorLog /var/log/apache2/<%= name %>_error.log ErrorLog /var/log/apache2/<%= name %>_error.log
LogLevel warn LogLevel warn