Re: tmpfs

hi.

did you try to check which deleted logfiles are still opened by running 
processes? try to run 'lsof /var/log' to see which files are currently 
opened; deleted ones are marked as (deleted) - they aren't displayed, 
but are still allocated on FS (until file is closed).

On 11/20/24 23:03, Boris via leaf-user wrote:
> Hej Erich,
>
> Am 20.11.24 um 01:09 schrieb Erich Titl:
>> Hi Boris
>>
>> Am 19.11.2024 um 22:47 schrieb Boris:
>>> Hej Erich,
>>>
>> ...
>>
>>> aproximately 25 MB while being rotated (kept) 4 days.
>>>
>>> So, if there are 36 MB logged (and compressed) in 4 days and the box 
>>> ia up for 10 days, we don't even reach the 121 MB from df .
>>>
>>> Funny, isn't it?
>>>
>>> With the uncompressed logs, df -h shows
>>> tmpfs                   800.0M    149.0M    651.0M  19% /var/log
>>>
>>> For me it smells like tmpfs does not release former occupied storage 
>>> after logrotating.
>>
>> I doubt this, in my case to clear the space a correct restart of the 
>> service was needed. 
>
> At the risk of revealing my complete cluelessness: Which service 
> (daemon) are you restarting correctly??
>
> [snip]
>
>>>
>>> My habit ist to reboot the box when finding it on 100%.
>>
>> Well right, this clears the memory storage.
>>
>> what you can do is play with the LOGSIZE parameter. It determines if 
>> a log gets rotated and if I am not mistaken it is in KB. Then to 
>> watch what logrotate does you can always start it like
>>
>> sh -x /usr/bin/logrotate
>>
>> You will see that savelog is used to compress and rename the file
>>
>
> Thank you for explaining that to me.
>
> Indeed, executing logratate without changing any parameter does eefect 
> _something_:
>
> # df -h
> tmpfs                   800.0M    133.0M    667.0M  17% /var/log
> # sh -x /usr/bin/logrotate
> Hangup
> # df -h
> tmpfs                   800.0M    130.1M    669.9M  16% /var/log
>
> I'll follow your hint and experiment with LOGSIZE, which now is
> # grep LOGSIZE *
> hostapd:LOGSIZE=1024
> lighttpd:LOGSIZE=1024
> ntp:LOGSIZE=4096
> pppd:LOGSIZE=1024
> shorewall:LOGSIZE=1024
> shorewall6:LOGSIZE=1024
> syslog:# NOTE: depends if LOGSIZE has been reached
> syslog:# LOGSIZE - size in kb when logrotate move and compress log files
> syslog:LOGSIZE=1024
> ulogd:LOGSIZE=1024
>
> Thanks,
>
>
> Boris
>
>
> ------------------------------------------------------------------------
> leaf-user mailing list: [email protected]
> https://lists.sourceforge.net/lists/listinfo/leaf-user
> Support Request -- http://leaf-project.org/




------------------------------------------------------------------------
leaf-user mailing list: [email protected]
https://lists.sourceforge.net/lists/listinfo/leaf-user
Support Request -- http://leaf-project.org/