IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

[Q] Installing SRS5

Martin Hilbers mph at dl.ac.uk
Wed May 13 10:07:13 EST 1998


Jean-Loup Risler wrote:
> 
> Dear Gurus,
> 
> I am trying to install SRS5.0.1 on my Dec AlphaServer 2100 running
> DU4.0B.
> 
> I have done what I should (I think) in the srsdb.i, embl.i, etc ...
> files.
> 
> Running srscheck and then srsupdate, I have consistently some
> problems with the "new" databanks: emblnew, swissprotnew, sptremblnew.
> 
> The building of the indices and/or the links for these banks fail
> with a "segmentation fault" and a core dump. I can see no problem
> with the "main" banks (embl, swissprot, sptrembl).
> 
> Curiously, the "new" databanks are processed OK if I re-run srscheck
> and srsupdate after having indexed the main banks.
> 

srs still has a problem with databases sharing the same format - if you
try updating a number of such databases (eq embl and emnew) in one
go, it will only do the srsbuild for the first one, for the second
it will try the "srsbuild -c" and "srsbuild -m", and try to build
the links, but without the results of "srsbuild" these obviously fail.
A workaround for this is to define separate formats for emblnew 
(EMNEW_FORMAT and EmNewFeatures_Format - or something like that)
These formats are identical to the original ones.

(At least this is what I did on our server)


> Could it come from a bad parameter tuning of my machine?
> 
> Finally, after having repeated some times the "srscheck-srsupdate"
> cycle, everything seems to have been indexed OK.
> 
> But now Netscape refuses to start a new SRS session with the
> message "file could not be opened: /srs/tmp/1TZMo193nkD".
> 

Apart from a trivial permission thing - I once spend half a day
trying to find why I got this type of error and it turned out to
be caused by a srs core-dump which had filled the disk....
(srs core dumps tend to be very large - or our disks very small..-)

Cheers,

martin
-- 
-------------------------------------------------------------------
| Martin Hilbers http://www.dci.clrc.ac.uk/Person.asp?m.p.hilbers |
| SEQNET                |     E-mail: m.p.hilbers at dl.ac.uk        |
| Daresbury Laboratory  |     Tel:    +44-1925-603492             |
| Daresbury, Warrington |     Fax:    +44-1925-603100             |
| Cheshire WA4 4AD      | SEQNET is the UK national EMBNet node   |
| United Kingdom        |     http://www.seqnet.dl.ac.uk/         |
-------------------------------------------------------------------




More information about the Bio-srs mailing list

Send comments to us at biosci-help [At] net.bio.net