My analysis of the problem is the following:

postrm (of the old package) removes the loaders.cache file unconditionally and
postinst (of the new package) creates it again.
That means, during upgrades there is a timeframe where loaders.cache does not 
exist.

When gtk-update-icon-cache is triggered via a file system trigger, it is
possible that it is run while loaders.cache is not present.

I talked to Joss on IRC.
He suggested to on remove the loaders.cache file in postrm only on remove|purge.
Otherwise gtk-update-icon-caches could be non-functional (this needs
verification though)

As we can't fix the postrm of the old package though, we should consider doing a
stable update via a point release.

We also need to make sure, to move/remove the loaders.cache for the multiarch
transition explicitly, as this will be no longer done as side-effect of postrm.

Michael
-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to