We're currently working with re-indexing one of our indexes on one of our 
test servers, and we appear to be having issues. The operation should be 
fairly obvious, and there's not a lot of data -- around 320.000 total 
documents in the index, at around 1.15GB of data. We scan from the old 
index, and bulk index 1000 documents at a time (we've also tried 500 
documents at a time to no avail). Since our production indexes are more 
around the 30GB size, we're a bit discouraged by hitting this wall.

Once in a while, we get a random "resource not available" error, but they 
usually resolves themselves and go away -- however, after indexing around 
260.000 documents, the Python script starts behaving weirdly, throwing *a 
lot* of resource not available errors. A full gist is 
here: https://gist.github.com/HenrikOssipoff/ac63060d4c18182719dd (note 
that this is Python 3.4, so chained exceptions and everything -- the top 
exception is not actually the important one).

Now, thing is. During this time, Elasticsearch itself seems fine, and the 
system is generally available for querying. The JVM heap is fine, and the 
Elasticsearch log itself has absolutely no warnings or errors (log level 
INFO enabled).

Our next course of action will be try to doing this by just cURLs outside 
of Python, or even enabling debug logging in Elasticsearch, but we wanted 
to see if anyone has some ideas for what could be the course.

It seems logical, that since it only happens every one in a while, and gets 
worse the more documents we index, it has something to do with the amount 
of _bulk indexes we do, but we can't seem to find anything in logs or stats 
that support this.

Any help appreciated, or any pointers in the right direction.

Thank you!

Regards,
Henrik Ossipoff Hansen

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/2afa6d99-b9d4-42a2-9f4e-b717a3a41251%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to