Don't use the LCIO branch -- you should use trunk instead. I sent an email about this issue last week to the software list with all the details. At least, I'm assuming that this is a problem with new SLIC and using an LCIO that doesn't support one specific change to the MCParticle class. The LCIO C++ trunk should handle this fine, and lcsim should be able to read new and old files just fine.
We basically just need to build a version of hps-dst using LCIO trunk, and then I think this problem should go away. I'm testing this locally now.
Do you want me to build it at JLAB? (Or Omar can?)
-----Original Message-----
From: [log in to unmask] [mailto:[log in to unmask]] On Behalf Of Bradley T Yale
Sent: Monday, August 15, 2016 1:23 PM
To: Omar Moreno
Cc: Graham, Mathew Thomas; hps-software; Uemura, Sho
Subject: Re: [Hps-analysis] New WABs
It's probably related to this:
https://jira.slac.stanford.edu/browse/HPSJAVA-674
I notice the dst_maker is still being built using /home/hps/hps_soft/lcio/v02-06
instead of the newer 'collection_type_fix' version (as Maurik mentioned).
However, when trying to build the 'collection_type_fix' version at JLab, I get an error about using the correct JDK:
[freehep.ant] BUILD FAILED
[freehep.ant] /u/group/hps/hps_soft/lcio/collection_type_fix/trunk/build.xml:283: Unable to find a javac compiler; [freehep.ant] com.sun.tools.javac.Main is not on the classpath.
[freehep.ant] Perhaps JAVA_HOME does not point to the JDK
Currently, JAVA_HOME is set to
/apps/scicomp/java/jdk1.7
This is probably simple enough to fix though.
________________________________
From: [log in to unmask] <[log in to unmask]> on behalf of Omar Moreno <[log in to unmask]>
Sent: Monday, August 15, 2016 2:16:15 PM
To: Bradley T Yale
Cc: Graham, Mathew Thomas; [log in to unmask]; Uemura, Sho
Subject: Re: [Hps-analysis] New WABs
The DST issue might have to do with the LCIO version that the DST maker is being linked against. After Jeremy's patch of LCIO, the trunk of C++ LCIO is required to the newest recon files. I'll look into it in a bit.
On Mon, Aug 15, 2016 at 10:58 AM, Bradley T Yale <[log in to unmask]> wrote:
When looking at the recon events (.slcio), the file size looks normal, but when printing the collections from the events:
java -jar /group/hps/hps_soft/lcio/jarfiles/lcio-2.4.4-SNAPSHOT-bin.jar print -H /cache/mss/hallb/hps/production/postTriSummitFixes/recon/wab/1pt05/wabv2_10to1_HPS-EngRun2015-Nominal-v5-0-fieldmap_3.10-20160813_pairs1_338.slcio
It prints one event, and then the error:
Exception in thread "main" java.lang.OutOfMemoryError: Requested array size exceeds VM limit
at java.util.Arrays.copyOf(Arrays.java:2245)
at java.util.Arrays.copyOf(Arrays.java:2219)
at java.util.Vector.grow(Vector.java:262)
at java.util.Vector.ensureCapacityHelper(Vector.java:242)
at java.util.Vector.setSize(Vector.java:285)
at hep.io.sio.SIOInputStream.readPTag(SIOInputStream.java:43)
at hep.lcio.implementation.sio.SIOMCParticle.<init>(SIOMCParticle.java:30)
at hep.lcio.implementation.sio.SIOEvent.readData(SIOEvent.java:122)
at hep.lcio.implementation.sio.SIOLCReader.readNextEvent(SIOLCReader.java:120)
at hep.lcio.implementation.sio.SIOLCReader.readNextEvent(SIOLCReader.java:148)
at hep.lcio.util.Printer.print(Printer.java:110)
at hep.lcio.util.PrintCommandHandler.execute(PrintCommandHandler.java:89)
at hep.lcio.util.CommandLineTool.parse(CommandLineTool.java:219)
at hep.lcio.util.CommandLineTool.main(CommandLineTool.java:126)
Does anyone else know about this problem? Being Java-related, I wouldn't think that it would affect the DST maker though.
Also, did anyone work on the JLab DST maker recently?
I see some new stuff in the source code dated July 28, which I think is more recent than the last batch of working DSTs...
________________________________
From: Bradley T Yale
Sent: Monday, August 15, 2016 12:49:06 PM
To: Graham, Mathew Thomas
Cc: Uemura, Sho
Subject: Re: [Hps-analysis] New WABs
The recon logs look ok, but then no events were written in the dst for some reason, with no errors reported.
I'll check it out.
________________________________
From: Graham, Mathew Thomas <[log in to unmask]>
Sent: Monday, August 15, 2016 12:26:35 PM
To: Bradley T Yale
Cc: Uemura, Sho
Subject: Re: [Hps-analysis] New WABs
tritrig DSTs all look empty too...
On Aug 15, 2016, at 9:13 AM, Graham, Mathew Thomas <[log in to unmask]> wrote:
-rw-r--r-- 1 mgraham
________________________________
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1 <https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1>
________________________________
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
|
