Added an empty wrapper for a GridEngine job runner which is only activated when SGE_ROOT is detected.
Refactored a bit more common code into CommandLineJobRunner / JobRunner / FunctionEdge.
Status for analyisNames now includes the number of functions in state.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5399 348d0f76-0448-11de-a6fe-93d51630548a
- LSF recycles memory for C arrays, but sets a separate variable setting the size of the array to zero.
- JNA only sees the non-NULL pointer and starts to auto-access it, sometimes causing a SIGSEGV.
- In the short term neutered the jobInfoEnt structure so that this bad array is not autoRead().
QGraph updates:
- Job status is now checked in bulk every 30 seconds instead of one job at a time, even in the middle of dispatching jobs.
- If there is a hiccup (unexpected but not fatal error) during status check then the the error is ignored and status is checked again 30 seconds later.
- Jobs prefer to dispatch depth vs. breadth first.
More refactoring of SG framework separating the reusable code from the implementations.
The DistributedScatterFunction is still a work in progress and is not enabled yet. Still need to think through how Queue handles when a job dies.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5387 348d0f76-0448-11de-a6fe-93d51630548a
Bug smashes for the MCFP:
Synchronized access to LSF library and modifications to the QGraph.
If values are missing from the graph with -run make sure to exit with a non-zero.
Refactored QGraph to pre-generate a unique Int for each QNode speeding up getHashCode/equals inside the graph.
Added jobPriority and removed jobLimitSeconds from QFunction.
All scatter gather is by default in a single sub directory queueScatterGather.
Moved some FCPTest into BaseTest/PipelineTest for use by MFCPTest.
Rev'ed the 1000G bams used for validation from v1 to v2 and added code to look for the bams before running other tests.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5247 348d0f76-0448-11de-a6fe-93d51630548a
Moved VariantEval validation from FCPTest to PipelineTest.
Cleaned up some duplicate code for writing temp files during tests.
Moved FCPTest to playground namespace to match move for FCP.q.
Added a basic HelloWorldPipelineTest for the HelloWorld QScript.
Moved duplicated error handling from JobRunners into the FunctionEdge.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5068 348d0f76-0448-11de-a6fe-93d51630548a
- bsub command line is no longer fully printed out.
- extraBsubArgs hack is now a callback function updateJobRun.
Updated FullCallingPipelineTest to reflect latest changes to fullCallingPipeline.q.
Added a pipeline that tests the UGv2 runtimes at different bam counts and memory limits.
Updated VE packages that live in oneoffs to compile to oneoffs.
Added a hack to replace the deprecated symbol environ in Mac OS X 10.5+ which is needed by LSF7 on Mac.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4816 348d0f76-0448-11de-a6fe-93d51630548a
Added a brute force -retry <count> option to Queue for transient errors.
Waiting up to 2 minutes for the LSF logs to appear before trying to display the errors from the logs.
Updates to the local job runner error logging when a job fails.
Refactored QGraph's settings as duplicate code was getting out of control.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4563 348d0f76-0448-11de-a6fe-93d51630548a
Queue GATK generated .intervals is now a List(File) again removing special case handling in the generator.
Instead of using @Scatter annotation, using ScatterFunction instance to determine if a job can be scattered.
Implemented special VcfGatherFunction which only uses the header from the first file, even if the other files differ in their headers.
Added a -deleteIntermediates to Queue to delete the outputs from intermediate commands after a successful run.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4536 348d0f76-0448-11de-a6fe-93d51630548a
Re-logging the failed jobs and the path to their log files at the end of a run.
Added a parameter -bigMemQueue for the fullCallingPipeline.q instead of hardcoding gsa (gsa was backed up and it was actually faster to run on week).
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4520 348d0f76-0448-11de-a6fe-93d51630548a
On -run cleaning up .done and .fail files for jobs that will be run.
Added detection to Firehose YAML generator shell script for (g)awk versions that ignore "\n" in patterns.
Removed obsolete mergeText and splitIntervals shell scripts.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4452 348d0f76-0448-11de-a6fe-93d51630548a
Queue now submits new LSF jobs only after previous functions have completed successfully.
When the Queue process is shutdown (ex: via Control-C) sends a bkill command for any running jobs.
Ported commands like creating directories and scatter/gather interval list to scala functions.
Updates to LSF status tracking by porting the python to internally generated bash scripts.
Temporarily disabled job name submission to LSF. Plus side is that the full command is now available in "bjobs -w". TODO: Put back jobName passing to LSF based on an option?
Changed BaseTest to allow scala to access paths to references.
Changed the extension generator to default the analysis name to the walker "name".
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4442 348d0f76-0448-11de-a6fe-93d51630548a
kshakir