This is likely a pidfile mixup.. (it's quite common).

CC: (none) => mageia

Yup that was it.. A simple missing "d" in the initscript.

New version hitting updates_testing shortly  c-icap-0.1.7-3.1.mga2

Tested on i586, MGA2.

I can reproduce the problem here, and the update candidate fixes it.

before:
[root@localhost wassi]# /etc/init.d/icapd start
Starting icapd (via systemctl):  Job failed. See system journal and 'systemctl status' for details.

[root@localhost wassi]# systemctl status icapd.service
icapd.service - LSB: Starts the c-icap daemon
          Loaded: loaded (/etc/rc.d/init.d/icapd)
          Active: failed (Result: exit-code) since Fri, 07 Dec 2012 19:27:58 +0100; 44s ago
         Process: 8740 ExecStart=/etc/rc.d/init.d/icapd start (code=exited, status=255)
          CGroup: name=systemd:/system/icapd.service
                  â 7752 icapd -f /etc/icapd/c-icap.conf
                  â 7754 icapd -f /etc/icapd/c-icap.conf
                  â 7755 icapd -f /etc/icapd/c-icap.conf
                  â 7756 icapd -f /etc/icapd/c-icap.conf


after:
[root@localhost wassi]# /etc/init.d/icapd start
Starting icapd (via systemctl):  Job failed. See system journal and 'systemctl status' for details.

-> Reboot

[root@localhost wassi]# /etc/init.d/icapd start
Starting icapd (via systemctl):                          [  OK  ]
[root@localhost wassi]# /etc/init.d/icapd stop
Stopping icapd (via systemctl):                          [  OK  ]

However, I am not sure how to proceed now. Remco, as you seem to use this package: could you see if the version in updates_testing works for you and maybe provide a testing procedure? If so, please also tell us which architecture you are on, so we can take your test results into account.

CC: (none) => wassi

@Colin: I guess you did, but as you didn't mention it above I'll ask you just in case: please don't forget to push this change also to Cauldron. If you did, feel free to ignore this remark. :)

@Remco: Never mind, I overlooked parts of your last comment on bug 6313.

Note to other QA testers: As drakguard uses dansguardian which in turn uses the icap server, testing icap in combination with drakguard should be sufficient. If everything works, both update candidates (6313 & 8199) can be validated.

Thanks user7. For what it's worth, I briefly tested (only confirmed icapd and dansguardian started at boot time) on x86_64 and that worked with the updated packages. I was hoping that the original reporter of the dansguardian bug could also confirm dansguardian working as expected, but haven't heard from him yet.

No other changes have been made to either of the packages though other than working on the systemd units / files to ensure proper start up at boot.

@user7: Yes this is in my list to push to cauldron - wanted to convert it to a proper systemd unit tho' first before pushing there...

I consider this update candidate and bug 6313 ready for validation, taking into account Remco's and my testing. As it is late already, I'll validate tomorrow if noone else is quicker. I'll add the keywords for now.

@Colin: ok, thanks.

Whiteboard: (none) => MGA2-32-OK MGA2-64-OK

Update validated.
Thanks.

Proposed Advisory:
This update fixes a pidfile mixup, preventing icapd from starting.

SRPM: c-icap-0.1.7-3.1.mga2.src.rpm

Could sysadmin please push from core/updates_testing to core/updates.

Thank you!

Keywords: (none) => validated_update
CC: (none) => sysadmin-bugs

No linking needed according to depcheck.

Update pushed:
https://wiki.mageia.org/en/Support/Advisories/MGAA-2012-0243

Status: NEW => RESOLVED
CC: (none) => tmb
Resolution: (none) => FIXED