Version: syncthing v0.14.46 “Dysprosium Dragonfly” (go1.10.1 linux-arm) firstname.lastname@example.org - on a Raspi
I have now seen this for the second or third time:
Syncthing suddenly just not running anymore.
This time i checked:
sudo service syncthing@sync status
With this output:
● email@example.com - Syncthing - Open Source Continuous File Synchronization for sync Loaded: loaded (/lib/systemd/system/syncthing@.service; enabled) Active: inactive (dead) since Mo 2018-04-09 00:14:26 UTC; 10h ago Docs: man:syncthing(1) Process: 4304 ExecStart=/usr/bin/syncthing -no-browser -no-restart -logflags=0 (code=killed, signal=PIPE) Main PID: 4304 (code=killed, signal=PIPE)
In the journal there was no error message at the end. What could be causing this?
I have one suspect:
I have one folder which is VERY big (more than a million files, >100GB).
I have seen this before, that while scanning this folder syncthing stopped working. My guess would be that there is some kind of is-alive-check of the system to verify that the service is still running, but probably when the scan takes too long it is not responding or not generating a heartbeat or whatever triggers the is-alive-check. Thus the service is killed.
When i paused this particular folder the service survived.
When i first paused it until al other folders were scanned (after restart) and then activated it, syncthing would also survive.
Anyway … it seems connected. (But of course this could be coincidence as well …)