[arch-dev-public] [signoff] syslog-ng 3.0.4-2

[Paul Mattal]

On 02/21/2010 09:53 PM, Eric Bélanger wrote:
> On Sun, Feb 21, 2010 at 9:45 PM, Paul Mattal<paul at mattal.com>  wrote:
>> Please signoff, 1 for each arch.
>>
>> The only difference is that I've added /var/log/crond.log to the logrotate
>> script so that its rotation is now handled when syslog-ng rotates all its
>> logs. The benefit of this is that it's coordinated properly with the HUPping
>> of syslog-ng, which is the actual creator of this log now that dcron logs to
>> syslog.
>>
>> There is an intended overlap here. In order to ensure smooth migration of
>> log rotation, the log will be "double covered" for a while (but logrotate
>> detects this and ignores one of them) until I (shortly) push the dcron 4.4-2
>> which removes its own logrotate script for /var/log/crond.log.
>>
>> - P
>>
>
> Other log files handled by syslog-ng should also be added to
> syslog-ng.logrotate :
>
> /var/log/lpr.log
> /var/log/uucp.log
> /var/log/news.log
> /var/log/ppp.log
> /var/log/debug.log
> /var/log/acpid.log

Does anyone disagree? If not, I'll add all of these and push a 3.0.4-3.

One thing we should think about in general is whether or not we should 
hijack signoff threads to pile on additional requests that should really 
be their own flyspray tasks. It discourages folks from doing small good 
deeds to core packages if they end up having to do all the maintenance 
on the package in order to do one good deed.

This will unfortunately delay the carefully-synchronized crond operation 
I was hoping to get through today; and the way my schedule stacks up 
this week, it will probably be next weekend before I can wrap all this 
cron operation up. Perhaps that's an acceptable delay, for the good that 
gets done, but we should consider whether this is really the intent of 
the signoff process.

- P