[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

[MoM] Regarding status of fis-gtm-initial package



Hi,

I had another look into fis-gtm-initial package and noticed that it
creates a deb package containing two (=both available) binary tarballs
contained in the source ball.  IMHO that's pretty useless and thus I
changed debian/rules to only install the tar which matches the
architecture that matches the build system.  Thus I've got a binary
deb which basically contains:

  /usr/lib/fis-gtm/distribution/gtm_V54002B_linux_x8664_pro-amd64.tar.gz

(in my case the amd64 binary) and a postinst file created by Thorsten
that seems to implement some installation magic.  I did not dived into
this but it fails for me:

...
Installation completed. Would you like all the temporary files
removed from this directory? (y or n)  
... automatic configuration finished
##############################################
rmdir: failed to remove `TMPPOSTINST': Directory not empty
dpkg: error processing fis-gtm-initial (--configure):
 subprocess installed post-installation script returned error exit status 1
Errors were encountered while processing:
 fis-gtm-initial


This problem now goes into GT.M expert land.  I have no idea what is
needed to do the GT.M initialisation and I have not read the needed
docs.  So it seems somebody with GT.M knowledge needs to fix the
postinst file.

Remark: My gut feeling tells me that doing such complex operations
should not be done in /usr/lib/fis-gtm but rather in /var/lib/fis-gtm.
I'm also definitely not convinced that this should be done in the
postinst.  I really think whatever needs to be done there should be done
on the machine that builds the package.

SO everything is kind of a gut feeling - but it seems something in the
current packaging is not done the way I regard it as most
straightforeward.  However, it might perfectly be that this is because
of the GT.M nature.

Kind regards

        Andreas.

-- 
http://fam-tille.de


Reply to: