Re: General question on understanding Solr log output

[Loren Siebert <lo...@siebert.org>]

Mikhail-
That makes sense about the commits. Solr's reporting that it has 2 commits
(the new one and the prior one), and then it deletes the old one and reports
that there is 1 remaining commit in the directory (the one just applied).
Thanks for explaining that.

My next question is on all the logging around the autowarming for the
various caches. For instance, I see three entries for filterCache:

filterCache{lookups=891,hits=794,hitratio=0.89,inserts=97,evictions=0,size=1125,warmupTime=2760,cumulative_lookups=33112,cumulative_hits=30112,cumulative_hitratio=0.90,cumulative_inserts=3000,cumulative_evictions=0}
filterCache{lookups=0,hits=0,hitratio=0.00,inserts=0,evictions=0,size=1024,warmupTime=2602,cumulative_lookups=33137,cumulative_hits=30132,cumulative_hitratio=0.90,cumulative_inserts=3005,cumulative_evictions=0}
filterCache{lookups=916,hits=814,hitratio=0.88,inserts=102,evictions=0,size=1130,warmupTime=2760,cumulative_lookups=33137,cumulative_hits=30132,cumulative_hitratio=0.90,cumulative_inserts=3005,cumulative_evictions=0}

I'm going to take a guess here at answering my own question. The first two,
like with the commits, are a report of what filterCaches are currently
instantiated. The first one is the one that's been around since the last
commit. It's had 891 lookups and took 2760ms to warm up back when it got
launched. The second one is the new one that just got warmed but hasn't been
made active yet. It took 2602ms to warm up. In that time, the original
filterCache got an additional (916-891=25) lookups before getting
decommissioned. That's the third filterCache line.

Am I reading that correctly?

--
View this message in context: http://lucene.472066.n3.nabble.com/General-question-on-understanding-Solr-log-output-tp3779907p3798351.html
Sent from the Solr - User mailing list archive at Nabble.com.