Postfix upgrade queries about replacing /etc/postfix/makedefs.out, when it really should always do so
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
postfix (Debian) |
Fix Released
|
Unknown
|
|||
postfix (Ubuntu) |
Fix Released
|
Undecided
|
Unassigned |
Bug Description
When upgrading Postfix, I was surprised to be asked
Configuration file '/etc/postfix/
==> Modified (by you or by a script) since installation.
==> Package distributor has shipped an updated version.
What would you like to do about it ? Your options are:
Y or I : install the package maintainer's version
N or O : keep your currently-installed version
D : show the differences between the versions
Z : start a shell to examine the situation
The default action is to keep your current version.
*** makedefs.out (Y/I/N/O/D/Z) [default=N] ?
This would appear to have been added with version 3.1.4-1, specifically:
* Install /etc/postfix/
which is a laudable and understandable goal. However, it seems quite unequivocal that the package-provided version of this file should always be used when installing the package, since the file itself opens with the following line:
# Do not edit -- this file documents how Postfix was built for your machine.
There does not seem to be a valid reason for a user to have a version of that file that differs at all from the package-provided version, then; any divergence, in fact, would be misleading.
Presumably then this file should be included in a way that does not require user intervention to update the makedefs.out file (ex. the /etc/ copy could just be a symlink to a copy in /usr/share/
Ran into this on a server running Ubuntu 18.04, when running upgrades which pulled down Postfix version 3.3.0-1ubuntu0.1.
description: | updated |
description: | updated |
Changed in postfix (Debian): | |
status: | Unknown → New |
Changed in postfix (Debian): | |
status: | New → Fix Released |
Note that I attempted to file a bug upstream with Debian too (since that's where this packaging choice originated), but I think I might have mucked up the reporting format as I see no indication my bug ever reached upstream.