launchpad-dev team mailing list archive

[Robert Collins <robert.collins@xxxxxxxxxxxxx>]

The referenced bug is causing many timeouts every day at the moment
and the folk who have looked at it so far have no particular ideas on
moving it forward.

I'd like to request that foundations see if they can help here, as an
interrupt (sorry!).

There are two particular angles:
 - the sql query being performed is very slow - 6 seconds. Can we fix
this or do we need a model change to answer the question more
efficiently? The query plan is doing table scans in the inner loop
which sets of alarms for me.

 - we're repeating the same query because of some lazr.restful glue:
this seems wasteful and I don't understand why its necessary. Can we
not 'just' issue the query, and tell the client we don't know the size
of the collection (for expensive API's). Or something.

Thanks,
Rob