Re: [wgs-assembler-users] chunks failed in cns stage : perl interpreter - permission denied

[Rui-Peng W. <rui...@gm...>]

So the CNS stage has finished and the chunks have been processed with
succes, however PBcR does not go to the next stage. Currenly, there is
another pBcR job still on hold queu:




*(smrtanalysis-2.3.0) [rpw@temp]$ qstatjob-ID  prior   name
user         state submit/start at     queue                          slots
ja-task-ID-----------------------------------------------------------------------------------------------------------------
246890
0.00000 pBcR_asm_P rpw          hqw   12/17/2014 04:03:23 *

>From what I know this hold job calls the following script:
runPBcR.sge.out.05.sh
But I do not know what to do next:
1) should I unhold this job with qrls and see what happens with risk that I
have to rerun the whole pipeline again or
2) is there a way to continue manually without unholding this job.

I would prefer option 2 but I do not know how to do it. Also I do not want
rerun from the start again cause there is no time.

After the CNS stage I got the trimmed fastq and fasta files per chunk. I do
not know how far I am with the self correcting step.

Any help or suggestion is very appreciated.



On 18 Dec 2014, at 09:12, Rui-Peng Wang <rui...@gm...> wrote:

Thanks for the suggestion but it did not help..

I restarted runPartition.sh by specifying the failed chunk (i.e #26) .
Although it did not complain about the perl interpreter the final
output.fasta and output.qual returned empty. I do see another stderr from :

26.lay.err

Parsing arguments
Opening stores
Loading library information
Streaming fragments
openLayFile()-- Failed to open '/data/../tempLib/asm.26.olaps' for reading:
No such file or directory
Couldn't open '/data/../tempLib/asm.26.olaps' for read No such file or
directory

Is there way to reproduce these asm*.olaps files again. So I can restart
runPartition.sh on the failed chunks with succes. At the momentI have only
2 failed chunks out of 140 but still 60 chunks are being processed.



On Thu, Dec 18, 2014 at 12:02 AM, Serge Koren <ser...@gm...> wrote:
>
> This is most likely due to a missing or different perl on some of your
> cluster nodes. You can try logging into the nodes on which the chunks are
> failing to see if you can run/access the specified perl path. If a
> consistent set of nodes is failing, you could also restrict the
> runPartition.sh script to only run on a subset of your machines where it
> works. If it only happens for a small subset of chunks, you can re-run the
> failed ones manually using the runPartition.sh script. Just specify the
> missing chunk on the command line:
> cd temp<library name>
> sh runPartiton.sh <failed chunk #>
>
> Sergey
>
> > On Dec 17, 2014, at 6:25 AM, Rui-Peng Wang <rui...@gm...>
> wrote:
> >
> > Hello,
> >
> > I am running PBcR using MHAP. It is currently in the PBcR CNS stage.
> Currently 53 chunks out of 200 are being processed. Three chunks are
> finished, however two of them returned this message in stderr:
> >
> > /opt/sge/default/pool/Node028/job_scripts/246889:
> /data/rp/tools/wgs-8.2/Linux-amd64/bin/convertToPBCNS:
> /opt/smrtanalysis/miscdeps/basesys/usr/bin/perl: bad interpreter:
> Permission denied
> >
> > The strange thing is one chunk took 4 hours and it finished successfully
> with a generated trim.fasta file. The failed chunks did not generate
> trim.fasta but there is a broken symlink..
> >
> > We have tested our SGE system with a dummy perl script (a modified copy
> of convertToPBCNS)  and all our nodes did not complain about the perl
> interpreter.
> >
> > Can someone help me how to deal with this and can I continue with the
> CNS stage process?
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> > Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> > from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> > with Interactivity, Sharing, Native Excel Exports, App Integration & more
> > Get technology previously reserved for billion-dollar corporations, FREE
> >
> http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk_______________________________________________
> > wgs-assembler-users mailing list
> > wgs...@li...
> > https://lists.sourceforge.net/lists/listinfo/wgs-assembler-users
>
>