What is the option to quiet the logging? Can’t seem to find it…
Caldera headless ran a PI running on a 16GB microSD out of space in about a month.
LMK, I could look into my own logroll
check your caldera headless settings using this command …
~/caldera-music/caldera-music --prefs
what do you have set for cache.maxSizeBytes ?
you can set it to something other than unlimited with something like this …
~/caldera-music/caldera-music --set cache.maxSizeBytes=4G (or whatever other cache size you want)
log files probably not filling your disk, more likely it’s the media cache 
oh hey thanks for replying.
no it’s def the log in ~/.cache/caldera-music/*.log
I deleted one which was almost 9GB and had to reinstall the headless client from scratch again
housekeeping needed
If I could just turn off INFO level logging it would probably be fine
basically using multi-room you get a zillion sync stats logged as the players go along
interesting … haven’t had that issue here, but yeah i can see how the multi-room logging might do it in some cases, good point
hmm this does not seem to be an issue on another PI node in the set, odd the timing metrics are not there? perhaps because it’s the main or “master” node in a multi-room set?
so far I’m simply guessing how this works by reading the logs.
and looks like there we do have some log rolling in place already so I take it back
oooh yes it seems that when you add a node to a multi-room set it starts logging effusively
look at these file sizes ok then
bruce@pvr:~$ ls -lh ./.config/caldera-music/
total 16G
-rw-rw-r-- 1 bruce bruce 4.1G Jul 12 18:27 ‘Caldera Music.log’
-rw-rw-r-- 1 bruce bruce 191K Jul 8 16:57 ‘Caldera Music.log.1’
-rw-rw-r-- 1 bruce bruce 12G Jul 5 20:12 ‘Caldera Music.log.2’
-rw-rw-r-- 1 bruce bruce 6.7K Jun 24 11:33 ‘Caldera Music.log.3’
-rw-rw-r-- 1 bruce bruce 6.7K Jul 12 18:27 devices.json
-rw-rw-r-- 1 bruce bruce 284 Jul 9 12:57 preferences.json
drwxr-xr-x 2 bruce bruce 4.0K Jul 12 18:16 snapjack
pretty sure that’s not normal … as i understand it, should only be updating log every few seconds with the multi-room stats … if it’s logging much more frequently, are you seeing retries, network buffer underruns etc ?
OK, thanks for pushing for more info.
I started poking thru some of the LARGE log files not finding anything at first but eventually came upon many many many of these
06:47:00.613 [30bf96c0] I (Xita): Counter discontinuity (gap=-4294967296), re-basing stream
06:47:00.613 [30bf96c0] I (Xita): stream resumed (sender restart / silence-watchdog recovery)
not sure the meaning. the gap= value seems to remain consistent
any obvious playback issues that correspond to when the logs show these? gaps, stutters, etc?
not that I’ve noticed. usually run 3-4 nodes in sync all day around the house so it’s not super close listening, but I haven’t noticed any fail mode until that one machine ran out of disc space
13:19:30.803 [1bfff6c0] I (Xita): Counter discontinuity (gap=-4294967296), re-basing stream
13:19:30.804 [1bfff6c0] I (Xita): stream resumed (sender restart / silence-watchdog recovery)
been running ‘tail -f’ on logs in different windows to observe what happens
two of the machines are spewing to the log rapidly with the above message
I think they began doing this at the same time but hard to pin down, the logs grow very rapidly
no problems with playback, everything seems normal otherwise
If you install the beta (check release notes thread), those issues have been addressed.
thanks Elan. beta installed and I’ll keep an eye on it thru the day
I had not noticed until late last night but I did run out of disk due to cache (~25GB) on the one machine which serves as the “main” node in a multiroom set (i.e. the one selected from the “cast” plexamp menu). removed it all and reinstall beta from scratch.
thanks again for carrying this thing forward, very cool
you can set a cache limit as well, using CLI —config and —set.
