Mailinglist Archive: opensuse-bugs (4250 mails)

< Previous Next >
[Bug 1012047] start on boot is not ticked in yast-dns-server
  • From: bugzilla_noreply@xxxxxxxxxx
  • Date: Thu, 05 Jan 2017 10:27:00 +0000
  • Message-id: <bug-1012047-21960-RWaEvqbyzi@http.bugzilla.suse.com/>
http://bugzilla.suse.com/show_bug.cgi?id=1012047
http://bugzilla.suse.com/show_bug.cgi?id=1012047#c4

Franck Bui <fbui@xxxxxxxx> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |fbui@xxxxxxxx
Flags| |needinfo?(igonzalezsosa@sus
| |e.com)

--- Comment #4 from Franck Bui <fbui@xxxxxxxx> ---
(In reply to Imobach Gonzalez Sosa from comment #2)

Pay attention to this line: "loaded (/etc/init.d/named; bad; vendor preset:
disabled)". So instead of "enabled" or "disabled", we get a "bad".


That's indeed a shortcoming that has been fixed by upstream commit
a3c4eb07106b29, which is included in v230+.

This means that the pending systemd update (v232) for factory will "fix" the
description of the unit state ("bad"). But the change won't fix this particular
bug, see below.

@Imobach, it seems that yast is parsing the output and of "systemctl status"
and expect to either find "enabled" or "disabled". If it's the case, this
doesn't sound correct as:

1. there's no guaranty that the output of "systemctl status" will be stable.
For example with v232 (which has the "fix") you'll see "generated"
(instead of "bad").

2. systemd provides a real (D-BUS) API to query the state of a unit file.

3. if 2. is not an option you still might rely on "systemctl is-enabled",
which
seems more appropriate and has at least a well defined behavior.

So IMHO, a better fix would be to change yast to do the right thing.

WDYT ?

--
You are receiving this mail because:
You are on the CC list for the bug.
< Previous Next >