PDA

View Full Version : [Mondrian] RE: VirtualCubeTest.testCalculatedMemberAcrossCubesfailing on SMP



Pappyn Bart
01-22-2007, 08:30 AM
John,

Could you tell me what kind of OS is running on your
4-way SMP machine ? And on you laptop ?

Thanks,
Bart
_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian

John V. Sichi
01-22-2007, 02:00 PM
Pappyn Bart wrote:
> John,
>
> Could you tell me what kind of OS is running on your
> 4-way SMP machine ? And on you laptop ?

SMP: Red Hat Enterprise Linux (not sure of version); JVM is HotSpot
1.5.0_10-b03

Laptop: Edgy Eft version of Ubuntu Linux; JVM is HotSpot 1.5.0_04-b05

My guess is that it's likely to be a timing-sensitive thing. The
property trigger stuff looked suspicious, but I tried disabling that out
and it still happens. I also tried disabling
testFormatStringExpressionCubeNoCache (since it runs just before and has
cache disabled on a base cube underlying a virtual cube) but the failure
still occurred.

JVS
_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian

John V. Sichi
01-22-2007, 03:11 PM
Something interesting: I noticed that HotSpot was automatically
selecting Server mode on the SMP machine (whereas on my laptop it
autoselects Client mode). I changed build.xml to force usage of Client
mode, and then "ant test" ran through with no failures.

Haven't tried forcing Server mode on laptop yet; if it's timing-related,
it may just be that it takes the combination of a very fast machine plus
Server mode to hit it.

Also, Bart, if you want to send me code modifications to enable tracing
targeted at debugging the problem, please do, and I'll run it through on
the SMP machine and send you the output.

JVS

John V. Sichi wrote:
> Pappyn Bart wrote:
>> John,
>>
>> Could you tell me what kind of OS is running on your
>> 4-way SMP machine ? And on you laptop ?
>
> SMP: Red Hat Enterprise Linux (not sure of version); JVM is HotSpot
> 1.5.0_10-b03
>
> Laptop: Edgy Eft version of Ubuntu Linux; JVM is HotSpot 1.5.0_04-b05
>
> My guess is that it's likely to be a timing-sensitive thing. The
> property trigger stuff looked suspicious, but I tried disabling that out
> and it still happens. I also tried disabling
> testFormatStringExpressionCubeNoCache (since it runs just before and has
> cache disabled on a base cube underlying a virtual cube) but the failure
> still occurred.
>
> JVS
>

_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian