Closed Bug 721223 Opened 12 years ago Closed 8 years ago

fully scripted postgresql dataload

Categories

(Socorro :: Infra, task)

task
Not set
normal

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: rhelmer, Unassigned)

References

Details

Socorro needs a schema, roles and some pre-populated data loaded in order to "bootstrap" a fully functional environment.

Right now this is the job of the "import.sh" script in:

https://github.com/mozilla/socorro/tree/master/tools/dataload

This process is documented in http://socorro.readthedocs.org/en/latest/populatepostgres.html

However, it is not 100% complete, it should implement all of the tables described in:
http://socorro.readthedocs.org/en/latest/databasetablesbysource.html

Additionally, I believe that the import script should not only load the data from CSV for these sections in the doc:

manually-populated-tables
tables-receiving-external-data
automatically-populated-reference-tables

It should also run the appropriate backfill or other jobs for matviews, which I believe should be fully covered by calling the "backfill_matviews()" function for the proper date range ("proper" meaning "whatever was used in the CSV files". There are some hidden requirements here, like the buildID must be no more than 30 days old or it'll be ignored etc).

application-management-tables - ideally these should not need bootstrapping (beyond simple DDL load), I would like it to be up to applications to manage setup and cleanup of these tables if at all possible.
Resolving as invalid, as this was opened baaaaack in the day.
Status: ASSIGNED → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.