SMRT Portal Job Fails - dyim42/SMRT-Analysis GitHub Wiki

When a SMRT Portal job fails, please do the following before filing a case or issue.

Step 1: Run a Lambda test job

Before looking at the master.log file, run a Lambda test job to determine if the problem is with the software or with the data. Run a RS_Resequencing job using the pre-packaged SMRT Cell in $SEYMOUR_HOME/common/test/primary/lambda and the pre-packaged reference lambda sequence in $SEYMOUR_HOME/common/userdata/reference/lambda.

###Step 2a: Investigate the data if the Lambda job succeeds If the Lambda job succeeds, then the software is working fine and you must investigate the data. Not all SMRT Portal jobs will succeed. The job will fail if your data is not appropriate for the analysis.

Do I have a corrupted SMRT Cell? https://github.com/PacificBiosciences/SMRT-Analysis/wiki/Common-SMRT-Portal-Errors
Do I have too little data to run this job?

e.x. RS_HGAP2/3 jobs will fail if you have less than 20X coverage for the genome size you selected

e.x. RS_Resequencing jobs will fail at the genomic consensus step if you have less than 1x coverage of your genome
Do I have too much data to run this job?

e.x. RS_HGAP2/3 jobs may timeout if you have >200x coverage for the genome size you selected

e.x. SMRT View will not display reads if have more than 1000x coverage

Step 2b: Investigate distributed computing if the Lambda job fails

Misconfigured distributed computing environments are common problems in SMRT Analysis. Turn off distributed computing by editing web.xml and restarting smrtportal-initd, then run another Lambda test job.

$SMRT_ROOT=/opt/smrtanalysis
vi $SMRT_ROOT/redist/apache-tomcat-7.0.23/webapps/smrtportal/WEB-INF/web.xml

Change the jobsAreDistributed parameter to false to turn off distributed computing in SMRT Portal:

   <param-name>jobsAreDistributed</param-name>
   <param-value>false</param-value>

Restart SMRT Portal daemons:

$SMRT_ROOT/admin/bin/smrtportal-inid stop
$SMRT_ROOT/admin/bin/smrtportal-inid start

Step 3a: Investigate the distributed computing configuration if the single-node Lamba job suceeds

Read the detailed section carefully and ask your cluster administrator to assit. https://github.com/PacificBiosciences/SMRT-Analysis/wiki/SMRT-Analysis-Software-Installation-v2.2.0#set-up-distributed-computing

Step 3b: Investigate the master.log if the single-node lamba fails

Look for lines that begin with [ERROR] in the file located at $SEYMOUR_HOME/common/jobs/<job_id_prefix>/<job_id>/log/master.log. If this file does not exist, the job fails immediately, and no job directory is created, look for errors written to $SEYMOUR_HOME/common/log/smrtportal/smrtportal.0.log.

Step 5. Report an issue

Please file a case at customer portal if you have an instrument for expedited response, or an issue at github issues, and provide the following:

Did a distributed RS_Resequencing lambda job succeed?
Did a single-node RS_Resequencing lambda job succeed?
Paste the [ERROR] lines from the master.log file