Oh, that's right! I do remember the discussion about changing to CNAMEs.
I wonder if Quarry also keeps audit logs of the queries it ran and how long
they took. It would be useful to display them in form of a trend line, for
instance.
On Sat, Dec 30, 2017 at 12:28 PM, Bryan Davis <bd808(a)wikimedia.org> wrote:
On Sat, Dec 30, 2017 at 9:51 AM, Huji Lee
<huji.huji(a)gmail.com> wrote:
I feel like Quarry is slower than before (before
being last week or last
month). Queries almost always get queued, and once executed, simple
queries
take longer to result.
I have no idea how to investigate this though. Any thoughts?
Quarry uses the *.analytics.db.svc.eqiad.labsdb servers. Before
2017-12-13 these servers were only used by tools which had explicitly
changed their configuration to point to the new servers. After that
date the legacy *.labsdb service names became CNAMEs (DNS aliases) to
the analytics.db service names. This greatly increased the load on the
backing servers which would logically make them slower to return
results.
Bryan
--
Bryan Davis Wikimedia Foundation <bd808(a)wikimedia.org>
[[m:User:BDavis_(WMF)]] Manager, Cloud Services Boise, ID USA
irc: bd808 v:415.839.6885 x6855
_______________________________________________
Wikimedia Cloud Services mailing list
Cloud(a)lists.wikimedia.org (formerly labs-l(a)lists.wikimedia.org)
https://lists.wikimedia.org/mailman/listinfo/cloud