[maker-devel] 2.27 and 2.28 incompatible

[Carson Holt]

There is something very odd, because I've never seen those errors before,
and 2.28 should use the same datastore structure as 2.27. I'm going to
write a script that will print out certain configuration information about
your install that might help me see what's going on.  My plane is boarding
now, so I'll send it to you later this evening.

Thanks,
Carson


On 9/19/13 9:37 AM, "Malcolm Hinsley" <mhinsley at ebi.ac.uk> wrote:

>To try to fix SQL lock file errors I installed 2.28 and made the mistake
>of running on a directory made by 2.27 (to run snap and augustus for the
>first time).
>
>Every contig fails due to errors like:
>
>
>------------- EXCEPTION: Bio::Root::Exception -------------
>MSG: Can't open file 04723bc0c22478764d90bbaebca96d23
>STACK: Error::throw
>STACK: Bio::Root::Root::throw
>/nfs/panda/ensemblgenomes/external/bioperl/BioPerl-1.6.901//Bio/Root/Root.
>pm:472
>STACK: Bio::DB::Fasta::fh
>/nfs/panda/ensemblgenomes/external/bioperl/BioPerl-1.6.901//Bio/DB/Fasta.p
>m:948
>STACK: Bio::DB::Fasta::subseq
>/nfs/panda/ensemblgenomes/external/bioperl/BioPerl-1.6.901//Bio/DB/Fasta.p
>m:929
>STACK: Bio::PrimarySeq::Fasta::seq
>/nfs/panda/ensemblgenomes/external/bioperl/BioPerl-1.6.901//Bio/DB/Fasta.p
>m:1089
>STACK: FastaSeq::seq
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/FastaSeq.pm:50
>STACK: Process::MpiChunk::_go
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Process/MpiChunk.pm:478
>STACK: Process::MpiChunk::run
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Process/MpiChunk.pm:341
>STACK: Process::MpiChunk::run_all
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Process/MpiChunk.pm:357
>STACK: Process::MpiTiers::run_all
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Process/MpiTiers.pm:286
>STACK: /nfs/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/maker:667
>-----------------------------------------------------------
>--> rank=NA, hostname=ebi3-198.ebi.ac.uk
> at
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Error.pm
>line 38
>        Error::_throw_Error_Simple('HASH(0x388cb78)') called at
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../
>lib/Error.pm line 306
>        Error::subs::run_clauses('HASH(0x388cbf0)', '\x{a}-------------
>EXCEPTION: Bio::Root::Exception -------------\x{a}...', undef,
>'ARRAY(0x38a0d18)') called at
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/Error.pm
>line 426
>        Error::subs::try('CODE(0x38f93f8)', 'HASH(0x388cbf0)') called at
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich
>2/bin/../lib/FastaSeq.pm line 95
>        FastaSeq::seq('FastaSeq=HASH(0x388dda0)') called at
>/nfs/production/panda/ensemblgenomes/external/maker/2.28_mpich2/bin/../lib
>/
>Process/MpiChunk.pm line 478
>        Process::MpiChunk::_go('Process::MpiChunk=HASH(0x38a0e50)',
>'run', 'HASH(0x38a0ec8)', 0, 0) called at /nfs/production/panda/ens
>emblgenomes/external/maker/2.28_mpich2/bin/../lib/Process/MpiChunk.pm
>line 341
>        Process::MpiChunk::run('Process::MpiChunk=HASH(0x38a0e50)', 0)
>called at /nfs/production/panda/ensemblgenomes/external/maker/2.
>28_mpich2/bin/../lib/Process/MpiChunk.pm line 357
>        Process::MpiChunk::run_all('Process::MpiChunk=HASH(0x38a0e50)',
>0) called at /nfs/production/panda/ensemblgenomes/external/make
>r/2.28_mpich2/bin/../lib/Process/MpiTiers.pm line 286
>        Process::MpiTiers::run_all('Process::MpiTiers=HASH(0x3867960)',
>0) called at /nfs/panda/ensemblgenomes/external/maker/2.28_mpic
>h2/bin/maker line 667
>
>Is there an easy to reset the datastore/ file names so that i can switch
>over to 2.28 without starting over? (eg maker -dsindex)
>I killed the job and ran 2.27 instead which seems to be jim dandy.
>
>-- 
>malcolm hinsley | EnsEMBL Genomes | +44 (0)1223 49 4669
>European Bioinformatics Institute (EMBL-EBI)
>European Molecular Biology Laboratory
>Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD
>United Kingdom 
>
>
>_______________________________________________
>maker-devel mailing list
>maker-devel at box290.bluehost.com
>http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org