[Gluster-infra] Awkwardness with HTTP-proxy on build.gluster.org

Niels de Vos ndevos at redhat.com
Wed Jan 14 19:48:07 UTC 2015


Hi,

on build.gluster.org there is a Jenking job that can reboot the
slave*.cloud.gluster.org machines. This job uses the Rackspace JSON API
over https. For some (to me) unknown reason, this script sometimes
fails. When it fails, it does not reach the Rackspace cloud, but a 404
error and gets the Gluster Community Website returned...

Error case:
    http://build.gluster.org/job/reboot-vm/9/console

Success case with partial (non-fatal) error:
    http://build.gluster.org/job/reboot-vm/10/console

Success case without errors:
    http://build.gluster.org/job/reboot-vm/7/console

The script is that does the work is located on build.gluster.org as
/usr/local/bin/reboot-vm.py and tries to disable proxies and skips SSL
verification (otherwise it always complains about URL is not in the
gluster.org domain).

Some assistance with making this work more reliable is much appreciated.
Other ideas are welcome too.

Thanks,
Niels
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 181 bytes
Desc: not available
URL: <http://www.gluster.org/pipermail/gluster-infra/attachments/20150114/a29f8c2e/attachment.sig>


More information about the Gluster-infra mailing list