Closed
Bug 1436122
Opened 8 years ago
Closed 8 years ago
Get planet.bugzilla.org working again
Categories
(Infrastructure & Operations Graveyard :: WebOps: Other, task)
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: justdave, Assigned: danielh)
References
Details
(Whiteboard: [kanban:https://webops.kanbanize.com/ctrl_board/2/6160])
Attachments
(1 file)
+++ This bug was initially created as a clone of Bug #1433835 +++
(Daniel Hartnell [:danielh] - bug 1433835 comment 9)
> As you mentioned, planet.bugzilla.org currently resides on the generic
> cluster. There is a shell script which is periodically run via a cron job.
> This shell script runs a git pull on the planet-content directory which has
> the following remote configured:
>
> https://github.com/mozilla/planet-content.git
>
> I'm surprised to see that it has retained the bugzilla folder in the
> branches directory - despite being removed from the upstream repo (which
> Mike mentioned above).
In otherwords, the site is currently configured to pull from a repository that no longer exists, and has apparently been running an a config file that's been sitting in stasis for a year and a half.
It apparently didn't come along for the ride with Planet Mozilla when it migrated to Git.
It's pulling from blogs of people who are no longer involved in Bugzilla, and is missing feeds from a few people who *have* been blogging about Bugzilla recently and really should be included there.
Not sure of the best way to deal with this - we could restore the bugzilla directory in the planet-content repo, or we could fork the whole thing to the @bugzilla github account and do the bugzilla config there or something. Up for suggestions on that one.
Comment 1•8 years ago
|
||
Apparently I'm the person with their hands on this machine, so I'll see if I can reinstate the old Bugzilla directory in the planet-content codebase and see if I can make the appropriate changes.
Assignee: server-ops-webops → mhoye
Comment 2•8 years ago
|
||
OK, pushed up https://github.com/mozilla/planet-content/commit/62c1ab9ef194166aab19276ef02d6c01276922bf
Now we wait.
| Reporter | ||
Comment 3•8 years ago
|
||
I don't see the stuff you pushed on the site yet - I suspect since it had a bugzilla directory in its checkout that wasn't in the upstream that the pull from upstream re-adding it is going to conflict, you might have the nuke it and manually update once.
Once you've confirmed it's pulling, I added a pull request to get our intended recent changes back into it. https://github.com/mozilla/planet-content/pull/2
| Reporter | ||
Comment 4•8 years ago
|
||
| Assignee | ||
Comment 5•8 years ago
|
||
Dave and Mike,
Thank you. Our cron job kicked off shortly after Mike pushed the restored Bugzilla folder back to the source repository and we picked up those changes. To Dave's point above - in my experience, Planet can cache some content rather aggressively. I no longer see Byron's feed in the config.ini that Mike put in place:
https://github.com/mozilla/planet-content/blob/62c1ab9ef194166aab19276ef02d6c01276922bf/branches/bugzilla/config.ini
I see the same file on the server. We still have Byron's feed in the page source though. I believe that we can remove the planet.bugzilla.org cache (or some subset of it), allow the site to rebuild and get the updated content. I'll look at the cache now and see what we can do.
| Assignee | ||
Comment 6•8 years ago
|
||
I actually submitted a PR to fix a path issue:
https://github.com/mozilla/planet-content/pull/3
Once that is merged, I can review the Planet build process and see if we're running into any other errors.
See Also: → https://github.com/mozilla/planet-content/pull/3
| Assignee | ||
Comment 7•8 years ago
|
||
Hi Mike,
Would you be able to review and merge that pull request? I do not have permission to do so.
https://github.com/mozilla/planet-content/pull/3
Flags: needinfo?(mhoye)
| Assignee | ||
Comment 8•8 years ago
|
||
I think we can consider this resolved. Each of our PRs have been merged (Thanks, Mike). I see the new Twitter addition from Dave, the subscriptions appear to be correct (including the removal of glob) and my PR allowed builds to successfully write content to the correct location on the generic cluster.
If there are any additional issues I can help with, let me know.
Status: NEW → RESOLVED
Closed: 8 years ago
Flags: needinfo?(mhoye)
Resolution: --- → FIXED
Updated•7 years ago
|
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•