yellow team mailing list archive

[Gary Poster <gary.poster@xxxxxxxxxxxxx>]

Hi Stuart.  We were working today on our parallel testing project and 
ran into a problem that postgresql-common was not installing in Lucid, 
when it had been doing fine just last week.  It turned out that this 
only happened with the new package you uploaded to the Launchpad PPA. 
When we specified the standard Lucid version, it was fine.

It fails like this:

root@lptests:~# apt-get install postgresql-common
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
  postgresql-common
0 upgraded, 1 newly installed, 0 to remove and 1 not upgraded.
Need to get 0B/135kB of archives.
After this operation, 573kB of additional disk space will be used.
WARNING: The following packages cannot be authenticated!
  postgresql-common
Install these packages without verification [y/N]? Y
Preconfiguring packages ...
Selecting previously deselected package postgresql-common.
(Reading database ... 35183 files and directories currently installed.)
Unpacking postgresql-common (from 
.../postgresql-common_128~lucid_all.deb) ...
Adding `diversion of /usr/bin/pg_config to /usr/bin/pg_config.libpq-dev 
by postgresql-common'
dpkg: unrecoverable fatal error, aborting:
 failed to fstat previous diversions file: No such file or directory
E: Sub-process /usr/bin/dpkg returned an error code (2)

That's the one from the PPA:

root@lptests:~# apt-cache madison postgresql-common 

postgresql-common |  128~lucid | 
http://ppa.launchpad.net/launchpad/ppa/ubuntu/ lucid/main Packages
postgresql-common | 106ubuntu1 | http://archive.ubuntu.com/ubuntu/ 
lucid-updates/main Packages
postgresql-common |        106 | http://archive.ubuntu.com/ubuntu/ 
lucid/main Packages

When we installed the one from Lucid, everything was fine:

apt-get install postgresql-common=106ubuntu1

Could you check into what is going on here, and fix it, or remove the 
package, or give us a longer term solution than the workaround we use 
above?  It looks like the problem might be trivial for a packager who 
understands diversions.  We don't right now.

Thank you

Gary