Hi,
I increased the heap to 256MB and the delete statement completed
after about 2 hours.
During this time derby was using around 190MB pretty steadily.
Thanks again,
Tim
On 8 Feb 2007, at 12:47, John Embretsen wrote:
Tim Troup wrote:
May I ask which JVM (vendor and version) you are using? I find it
strange that you run out of heap space in this case, but this of
course depends on what heap settings your JVM is using... (and
what you do before executing that delete statement).
I am running Mac OS X 10.4.8 with the JVM provided by apple:
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_06-112)
Java HotSpot(TM) Client VM (build 1.5.0_06-64, mixed mode, sharing)
I haven't made any modifications so I assume derby sues some
default heap size.
Given the version information above, I assume your default max heap
size is 64 MB. But I have no experience running Java on Mac, so I
cannot say for sure.
I don't do anything proir to issuing the delete statement beyond
starting up the server.
OK, that's interesting. Do you know roughly how many tuples are in
your tables? I've never seen a single (delete) statement using as
much as up to ~60 MB heap space before. Can anyone explain this?
If the problem persists, it would probably help to provide a small,
reproducible test case.
--
John
