Closed YannickB closed 7 years ago
@@ master #204 diff @@
==========================================
Files 73 73
Lines 5700 5723 +23
Methods 0 0
Messages 0 0
Branches 0 0
==========================================
+ Hits 1800 1804 +4
- Misses 3900 3919 +19
Partials 0 0
Powered by Codecov. Last update 5395471...7b9377c
Actually looks like my Odoo exec is in a restart loop. Honestly I can't remember if I've actually tested the Odoo instance, I think I've only gone as far as Gitlab due to the connection error. Always just kind of figured it was dead 😆
df84b7b8b79c dev-clouder9-all-clouder9-exec-20170116.021804 "/bin/sh -c '/opt/odo" 11 minutes ago Restarting (127) 4 minutes ago
Ok it shall be ok now, waiting for review.
@lasley The odoo exec image in Docker hub will be fixed with this PR, that's why you have the error. During my tests I use the Dockerfile field in clouder.image, the image on Docker hub shall work after the merge. I believe it's critical now that you can use the oneclick, can you open a ticket with the issue you have ? It shall be my top priority this week, along with libcloud DNS. I also take note you use Engine, I'll make some test on this mode to make sure everything is still fine here.
Ok, waiting for travis and then I'll merge.
That's all for now, I'll continue the oneclick debug starting from wednesday
Looks like the template build failed https://hub.docker.com/r/clouder/odoo-exec/builds/bvqak3lcwhz6zhghyiqd72v/
Yes, I receive a mail when a build fail so don't worry I couldn't miss it :). Shall be corrected.
So, maybe it's finally time to make Clouder works again, don't you think?..
This PR correct most of the problems in Odoo template, in images and also database name (no more need of the patch in files, as discussed with @lasley )
Bind and postfix has been rollbacked in a previous PR to use ubuntu. We'll migrate them to alpine later, first let's get back the oneclick. Bind shall works now and my test inside the container looks good but.. I have trouble with Docker Swarm (Docker 1.12.3), it doesn't want to publish the port 53 as UDP, hence I can't fully test the oneclick.
Please feel free to try on your side after the merge, my problem is without doubt a bug and maybe it doesn't happen everywhere. I'll continue my test with libcloud DNS anyway.
Also, a check has been added in swarm mode to wait for the containers to stop/start after we update the service.
require #203