2

Possible Duplicate:
Feature Request: Fast, comprehensive, powerful search

We are using FogBugz Version 7.3.3 (DB 745, Build 2047). We have separate boxes for FogBugz and for the SQL Server database. I don't know the specs of the computers offhand, but it's something reasonable. They aren't old Commodore 64s or anything.

If I do a search that returns no hits, it takes 3 seconds. If I do a search that returns some hits, it takes 6-12 seconds. Not unusably slow, but Google lets me search the entire internet and it takes less than a second. I'm just searching 50k FogBugz cases, it doesn't seem like it should take 6 seconds.

How do we determine why the searches are so slow so we can do something to speed it up?

flag
This not, obviously, an exact duplicate of the feature request, but please vote that up to express support. – Rich Armstrong Feb 8 2011 at 20:51

closed as exact duplicate by Rich Armstrong♦♦ Feb 8 2011 at 20:51

1 Answer

1

Hmmm. While FogBugz search is by no means fast, that level of latency for a small database is not reasonable. Can you try putting a copy of the database locally, then switch to that on off-hours to see if searching gets faster.

As for matching Google, they have a few more resources than we do. :)

link|flag
In general, do you guys recommend FB and the SQL Server databases be on separate boxes, Rich? – Ken Morse Aug 30 2010 at 14:22
It's usually not a problem, but it's the first place we look when we see greater-than-expected latency. – Rich Armstrong Aug 30 2010 at 14:54
Thanks for your response. I was hoping for some kind of logging we could turn on to verify what part of the search was taking what time. Moving the database on the same box isn't trivial, and without logs to prove that it would help I doubt that will happen. – Mhenry1384 Aug 30 2010 at 15:33
Hmmm. I'm not sure logging has this, but I'm pretty sure it should. Can you contact us directly to hash this out? customer-service@fogcreek.com – Rich Armstrong Aug 31 2010 at 14:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.