Open Stack

Joshua Hesketh <joshua.hesketh at gmail.com> writes:

> Hey all,
>
> So this took me a lot longer than I would have liked (the fact that I
> pushed this up in the middle of a conference should be evidence), but I
> believe I have managed to track this down:
> https://review.openstack.org/#/c/275483/
>
> We also have only ever been clearing our internal cache when reloading. We
> should perhaps look at doing that periodically but that is a different
> issue to the newly introduced leak. Although I haven't checked, I suspect
> our cache is a lot larger recently (looking at cacti and other metrics)
> probably due to cross-repo-dependencies so there is more need to keep it
> clear. This may have contributed to the recent larger memory use.

Ah, that explanation makes sense, thanks!

On the subject of clearing the cache more often, I think we may not want
to wipe out the cache more often than we do now -- in fact, I think we
may want to look into ways to keep from doing even that, because
whenever we reload now, Zuul slows down considerably as it has to query
Gerrit again for all of the data previously in its cache.

But we certainly don't want multiple copies of the entire cache.  Thanks
again!

-Jim