Daniel, thank you for pointing this out.
However, 64M was only used to trigger OutOfMemorys sooner in the first
place. A more realistic load szenario should probably have much bigger
heap, so I increased Xmx to 500M. After 12hrs still no shortage in memory...
-- Bernd
Daniel John Debrunner wrote:
Bernd Fondermann wrote:
I set up a local test environment and eventually saw
(a) OutOfMemorys (but only after a longer period of time, with -Xmx64M)
It may be that the Derby page cache is too big for this memory size. By
default the page cache is 1000 pages, and the default page size is 4k.
However I would guess with James that the default page size is 32k, as
it increases for tables with BLOBs/CLOBs or long rows. Thus the page
cache is much bigger when using BLOBs etc.
You can try bumping the page cache size down,
derby.storage.pageCacheSize=100
http://db.apache.org/derby/docs/10.1/tuning/rtunproper81359.html
Either as a system property, or in the derby.properties file.
This is a know issue in Derby, I think there are a couple of possible
solutions:
1) define memory usage in Mb, not pages
2) better dynamic configuration when OutOfMemory exceptions are seen,
e.g. dynamically reduce cache.
Dan.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
