475815 - provide a migration plan to move old data into the right partitions.

Status: RESOLVED WONTFIX

(Socorro :: General, task)

Description

[K Lars Lohn [:lars] [:klohn]]

We're all wound up and ready to proceed to production.  As we've discussed, we're scheduled to begin on Thursday, January 28 at 7PM PST.

Steps (should be about the same as when we pushed to staging)

1 - stop the Web App and replace with the "we're temporarily down" message - aking has details

2 - stop processors and monitor

3 - get latest version of Socorro from trunk.  This includes the migration scripts and new versions of monitor and processor

4 - set up config/setupdatabase.py.dist.  This is exactly the same as it was in
Bug 468465

5 - run .../scripts/migrate.py – lars would really like to have a way to watch this process run live either through sharing with 'screen' or 'vnc' or something.  It's also important to make sure that logging writes to a permanent transcript.

6 - restart processors

7 - restart monitor

8 - restore the Web App

Comment 1

[Austin King [:ozten]]

site down message, we can 
Skip Step 1

After Step 3 - get latest Socorro will get you a new outage.php file

cp webapp-phpdocs/outage.php webapp-php/index.php

Step 8 - restore the Web App

svn revert webapp-php/index.php

Details:
https://intranet.mozilla.org/SocorroOutagePage

Comment 2

[matthew zeier [:mrz]]

Holding on to this so I don't forget to include in downtime notice.

Comment 3

[Michael Morgan [:morgamic]]

Data's there, we're not sure if we'll need to reimport it but wontfix for now.  Just don't delete it please.