Closed Bug 576667 Opened 14 years ago Closed 14 years ago

Need to set up Sphinx Server and push Fx Input staging to live on 7/5

Categories

(mozilla.org Graveyard :: Server Operations, task)

task
Not set
critical

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: aakashd, Assigned: aravind)

References

Details

We've learned that our current search engine doesn't scale well (thanks to users grabbing our Fx4 Beta 1 candidate builds and sending us a ton of feedback). We need a push before Fx4 Beta 1 as there's going to be a lot of people using this App at the Summit and elsewhere. Dave Dash is currently implementing Sphinx onto our Input App and we want to get a push out on Monday night (or after he's completed the code changes).

Stuff to do from my knowledge:

- set up the sphinx indexing on the sphinx cluster
- get the sphinx server running on the sphinx cluster
- point the input app code at it
- ``git pull origin master`` from input.stage.mozilla.com  
- Input needs an apache reload after it goes live 

I think that's enough, but I'll defer to Frederic and Dave here. Will someone be avilable to get this done?
Depends on: 576535
Dropping the priority to normal for now, so that it doesn't page on-call. Please bump it back up to critical once we have the code and then the on-call person will get paged about it and can take care of.
Severity: critical → normal
Raising this back to critical. The code using Sphinx has landed on trunk, and presumably input has auto-pulled it in the mean time.

We are still ironing out some things, so no push to production yet. Still, on staging we need the following:
- a running sphinx server, whether you hook us up to an existing one or it's a new local instance
- if the SPHINX_* settings in settings.py are not correct, copy them to settings_local.py and change them.
- make sure that ``./manage.py update_index -r`` works by running it manually. Then update the cron job (it did not include "-r" before). If Sphinx is not local, consult with Dave Dash what cron job to set up instead.
Severity: normal → critical
Assignee: server-ops → jdow
Assignee: jdow → aravind
What website is this for?  It appears that we are re-using the existing sphinx cluster to index multiple sites production.  I would like to reuse the same setup if this is for a new site.
Where do I get the sphinx configs from, I tried looking under http://svn.mozilla.org/projects/, but I didn't see an input directory there.  Is this checked into a git repo somewhere else?
(In reply to comment #5)
> Where do I get the sphinx configs from, I tried looking under
> http://svn.mozilla.org/projects/, but I didn't see an input directory there. 
> Is this checked into a git repo somewhere else?

Nevermind, I see the staging repo is checked out from git://github.com/fwenzel/reporter.git

Will look for the configs there.
Yes, here: http://github.com/fwenzel/reporter/tree/master/configs/sphinx/

Sorry, if I had worked with sphinx before I'd probably have given you those straight away ;)
Okay, staging should be set, can we get some eyes on it and test it out?
On it!
Alright, everything looks good for a push to production.
Prod should be good to go now.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
verified FIXED
Status: RESOLVED → VERIFIED
Product: mozilla.org → mozilla.org Graveyard
You need to log in before you can comment on or make changes to this bug.