db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Oyvind.Bakk...@Sun.COM
Subject Re: DERBY-31: setQueryTimeout semantics
Date Fri, 22 Apr 2005 14:27:06 GMT
Daniel John Debrunner skrev:
> Oyvind.Bakksjo@Sun.COM wrote:
> 
>>We're then left with three choices:
>>1) setQueryTimeout() only affects Statement.execute()
>>2) setQueryTimeout() affects Statement.execute() and ResultSet.next(),
>>starting from zero for each invocation
>>3) setQueryTimeout() affects Statement.execute() and ResultSet.next(),
>>accumulating time spent in each invocation
>>
>>I think option 3 is the most sensible one, but I'm open to other
>>opinions. Also, it would be interesting to consider what other vendors
>>do. Does anybody know?
> 
> 
> I'd assumed from the wording of the setQueryTimeout that it was option 2
> , more of response time issue than resource usage issue. But it would be
> good to see what the other driver vendors do.

I've spoken to several other people today who also think option 2 is a 
better choice, and I don't have any strong objections. It's also 
slightly easier to implement than option 3.

I'll write a test program that will time the execute and fetch 
operations on certain long-running queries, run with and without a 
specific query timeout set. Based on these timings, it should be 
possible to analyze the cancellation strategy (if any) used by a certain 
vendor.

But for the time being, I think we should go with option 2.

-- 
Oyvind Bakksjo
oyvind.bakksjo@sun.com
Office: +47 73842119
Mobile: +47 92283522

Mime
View raw message