Re: [Java DB - testing] Error nightly 10.8 (rev 1502329) (Could not reserve enough space for object heap)

[Katherine Marsden <km...@sbcglobal.net>]

On 7/11/2013 5:04 PM, ingemar.aberg@oracle.com wrote:
> Java DB testing and reporting infrastructure.
>
> Error nightly 10.8 (rev 1502329)
>
> There were execution errors and/or timeouts.
> There were at least 1625 failures.
>
> Test report:  http://download.java.net/javadesktop/derby/javadb-5579293-report/
>
For a couple of nights the 10.8 compatibility tests have had lots of 
failures due to:
2     Embedded_TrunkVMj5lib_vs_ClientVM-j7lib_clientTrunk     N/A

Error occurred during initialization of VM
Could not reserve enough space for object heap

There have been multiple code  checkins to the 10.8 branch but I don't 
think they could cause this.  Has there been a change to the vm options 
or something else in the test infrastructure  that might prevent the VM 
from initializing for the compatibility tests on 10.8?

Thanks

Kathey

Re: [Java DB - testing] Error nightly 10.8 (rev 1502329) (Could not reserve enough space for object heap)

[Knut Anders Hatlen <kn...@oracle.com>]

Katherine Marsden <km...@sbcglobal.net> writes:

> On 7/11/2013 5:04 PM, ingemar.aberg@oracle.com wrote:
>> Java DB testing and reporting infrastructure.
>>
>> Error nightly 10.8 (rev 1502329)
>>
>> There were execution errors and/or timeouts.
>> There were at least 1625 failures.
>>
>> Test report:  http://download.java.net/javadesktop/derby/javadb-5579293-report/
>>
> For a couple of nights the 10.8 compatibility tests have had lots of
> failures due to:
> 2     Embedded_TrunkVMj5lib_vs_ClientVM-j7lib_clientTrunk     N/A
>
> Error occurred during initialization of VM
> Could not reserve enough space for object heap
>
> There have been multiple code  checkins to the 10.8 branch but I don't
> think they could cause this.  Has there been a change to the vm
> options or something else in the test infrastructure  that might
> prevent the VM from initializing for the compatibility tests on 10.8?

I asked around, but couldn't find anything that had changed. It has also
failed on multiple hosts, so it doesn't appear to be a problem local to
a single machine.

I noticed that the tests on the 10.8 branch used an older version of
Java 7 than the other branches, where the compatibility test runs have
been running cleanly. I've asked for an update of Java 7 on the 10.8
tests as well. Not sure if that will make any difference, but I'm
keeping my fingers crossed.

-- 
Knut Anders