636620 - Identify all SQL items that need to be performed on Stage for 1.7.7

Status: VERIFIED FIXED

(Socorro :: General, task)

Description

[Ryan Snyder [:ryansnyder] [:rsnyder] [:rysny]]

I'm not sure what has already been run on stage and what has not.  Per the Kohana error log, I know that the reports_duplicates table has not been added to stage.  What else in this list needs to be performed?

https://code.google.com/p/socorro/wiki/SocorroUpgrade#Postgres_Database

Comment 1

[Stephen Donner [:stephend] Not actively reading bugmail]

This is blocking testing, as we've had to disable automated tests, as well as haven't really started functional testing; appreciate quick movement on this if we're to ship the 8th; thanks!

Comment 2

[Laura Thomson :laura]

Over to Jabba but Josh may have a better idea.  Guys?

Comment 3

[Ryan Snyder [:ryansnyder] [:rsnyder] [:rysny]]

If nothing else we need schema updates to be run on stage.  

I understand that trying to run all of the update scripts to update the data in the db will take days to run on that box.  But we still need data for testing.  Can we run the update scripts on a small subset of data so that we can at least test the features we've built?

Comment 4

[Justin Dow [:jabba]]

Can't run on the current stage box, for various reasons, but we can run it on the new stage box that just got setup, however, it is best to do so after all of staging has been moved over to the new system. This is blocked by the embargo right now.

Comment 5

[[:jberkus] Josh Berkus]

Yes, I can limit the backfilling so that we can apply all 1.7.7 changes in around 24 hours.  It will take that  long, though.

Comment 6

[Laura Thomson :laura]

Nothing has been run on staging yet.  All instructions on the upgrade page need to be run.

Comment 7

[Justin Dow [:jabba]]

OK, so nothing has been run on stage yet. We clarified this at the meeting yesterday and jberkus was given the green light to run everything against the new stage server.

Comment 8

[[:jberkus] Josh Berkus]

The new stage should be fully updated.

However, the new dev is not updated.  Can we clone dev from stage?

Comment 9

[Ryan Snyder [:ryansnyder] [:rsnyder] [:rysny]]

I'm still getting errors regarding the stage database.  Per the Kohana logs, the reports_duplicates is not available in the stage db:

2011-03-03 16:41:56 -08:00 --- error: [5xx Error] File: application/libraries/drivers/Database/Pgsql.php; Line: 74; Message: pg_query() [<a href='function.pg-query'>function.pg-query</a>]: Query failed: ERROR:  relation &quot;reports_duplicates&quot; does not exist

Comment 10

[[:jberkus] Josh Berkus]

Are you connecting to socorro1.stage.db.sjc1.mozilla.com ?

Comment 11

[Ryan Snyder [:ryansnyder] [:rsnyder] [:rysny]]

Per the db setting in application/config/database.php, https://crash-stats.stage.mozilla.com is connecting to tm-stage01-master01.

Comment 12

[[:jberkus] Josh Berkus]

That's the old stage, and indeed it is not updated.  It will be going away soon.

You need to connect to the new stage.

Comment 13

[Ryan Snyder [:ryansnyder] [:rsnyder] [:rysny]]

@Jabba - Can you make that change please?  I do not have access to write to that file.

Comment 14

[Robert Helmer [:rhelmer]]

(In reply to comment #13)
> @Jabba - Can you make that change please?  I do not have access to write to
> that file.

I can do this right now, puppet is disabled so I'll make this change manually.

Comment 15

[Robert Helmer [:rhelmer]]

(In reply to comment #14)
> (In reply to comment #13)
> > @Jabba - Can you make that change please?  I do not have access to write to
> > that file.
> 
> I can do this right now, puppet is disabled so I'll make this change manually.

This is done, restarted httpd and memcached for good measure.

I see connections to 'socorro1.stage.db.:postgres' in netstat, no connections to tm-stage01-master01.

Comment 16

[Matt Brandt [:mbrandt]]

Verified per Robs comment 15. Test automation is also now back to normal.