Re: [gentoo-user] syslogd hibernating

[n952162]

On 1/15/21 4:30 PM, Michael wrote:
> On Friday, 15 January 2021 15:23:34 GMT Peter Humphrey wrote:
>> On Friday, 15 January 2021 15:08:15 GMT n952162 wrote:
>>> On 1/15/21 10:24 AM, Michael wrote:
>>>> On Friday, 15 January 2021 08:47:18 GMT n952162 wrote:
>>>>> Hello,
>>>>>
>>>>> can anyone explain this?
>>>>>
>>>>> I noticed today (15. January) that the
>>>>> /var/log/{messages,kern.log,etc.}
>>>>> files on a box were last touched on 22. November.
>>>>>
>>>>> sysklogd was in the rc-open /started/ state and was running.
>>>>>
>>>>> The configuration file, /etc/syslog.conf matches completely that file
>>>>> on
>>>>> another machine of mine, where the logs are properly updated.
>>>>>
>>>>> /etc/syslog.d/ was empty on both machines.
>>>>>
>>>>> I sent the process a HUP signal, as follows
>>>>>
>>>>> sudo kill -HUP $(cat /var/run/syslogd.pid )
>>>>>
>>>>> and all the log files were immediately updated and reporting.
>>>>>
>>>>> I rebooted my machine, and the log files are untouched, it is again
>>>>> hibernating.
>>>> Do you get something like this on your system?
>>>>
>>>> $ rc-update -s -v | grep syslog
>>>>
>>>>               syslog-ng |      default
>>>>
>>>> and,
>>>>
>>>> $ rc-service -v syslog-ng status
>>>>
>>>>    * Executing: /lib/rc/sh/openrc-run.sh /lib/rc/sh/openrc-run.sh
>>>>    /etc/init.d/
>>>>
>>>> syslog-ng status
>>>>
>>>>    * status: started
>>> $ rc-service -v syslog-ng status
>>>
>>>    * rc-service: service `syslog-ng' does not exist
>>>
>>> $ rc-service -v syslog status
>>>
>>>    * rc-service: service `syslog' does not exist
>>>
>>> $ rc-service -v syslogd status
>>>
>>>    * rc-service: service `syslogd' does not exist
>>>
>>> $ rc-service -v sysklogd status
>>>
>>>    * Executing: /lib/rc/sh/openrc-run.sh /lib/rc/sh/openrc-run.sh
>>>
>>> /etc/init.d/sysklogd status
>>>
>>>    * status: started
>>>
>>> $ rc-update -s -v | grep syslog
>>>
>>> $ rc-update -s -v | grep sysklog
>>>
>>>                sysklogd |      default                    sysinit
>>>
>>> I ran that as a script on a machine where the logs are properly updated:
>>>
>>> 01~>bash -x  /tmp/test
>>> + rc-service -v syslog-ng status
>>>
>>>    * rc-service: service `syslog-ng' does not exist
>>>
>>> + rc-service -v syslog status
>>>
>>>    * rc-service: service `syslog' does not exist
>>>
>>> + rc-service -v syslogd status
>>>
>>>    * rc-service: service `syslogd' does not exist
>>>
>>> + rc-service -v sysklogd status
>>>
>>>    * Executing: /lib/rc/sh/openrc-run.sh /lib/rc/sh/openrc-run.sh
>>>
>>> /etc/init.d/sysklogd status
>>>
>>>    * status: started
>>>
>>> + rc-update -s -v
>>> + grep syslog
>>> + rc-update -s -v
>>> + grep sysklog
>>>
>>>                sysklogd |      default
>> Substitute your syslog program's name for syslog-ng. Michael did say
>> "something like."
> Yes, quite, other log daemons are available.  :-)
>
> It seems your PC on which sysklogd works as expected has its rc script only on
> 'default' runlevel, rather than default +sysinit.  I don't know if sysinit is
> required.  If you remove sysinit and restart sysklogd from a terminal does it
> spew out any errors?


Oh, I missed that sysinit way out there on the left ;-)

It looks like that was it, no errors after removing runlevel sysinit,
and /var/log/messages has like the current time.

Thanks, that was great!