Tpac - record details retrieval can be very slow
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Evergreen |
Fix Released
|
Undecided
|
Unassigned |
Bug Description
Evergreen version: ~2.2-beta2
With Tpac, we've noticed that bib record retrieval can be extremely slow when retrieving a bib associated with a results set larger than 2k hits. The initial search results can come up reasonably fast for us regardless of how many hits we get, but clicking on a given record ends up being extremely slow.
Examples from our production Tpac of more than 10k hits:
A search for "dogs" : http://
A search for "cats" : http://
If you click on a record from that set, it takes ages to load, but when it finally does, things appear cached and "faster". The initial load time is ridiculously bad though.
I've discussed this briefly in IRC with senator, gmcharlt, and others, and the current hypothesis is that the differing method of results retrieval vs. record retrieval is the problem. And record retrieval being slower given the number of hits that come up.
Changed in evergreen: | |
status: | Fix Committed → Fix Released |
For reference, the work committed from bug #988146 should be helpful in starting to pinpoint the exact problem. Which we're pretty sure at this point is either the way we're calling unapi.bre or something to do with missing the search that's just been cached from the results page when looking for it from the record detail page.