4 ms·
I like it because I had found a tons of subtle issues over time with some of the scripts that were "simple" and system default of "just put it in ExecStart and
by ilyt 4y ago
I like it because I had found a tons of subtle issues over time with some of the scripts that were "simple" and system default of "just put it in ExecStart and set user" mitigates many of those.
For example:
Java app that gets put to background before it writes PID (which takes some startup time). In result running status right after start will say that app is stopped. Pacemaker didn't liked it and freaked out. Not only Java problem but start exacerbated it
MySQL disto package (Hi Red Hat) that didn't put PID into temporary partition, nor checked whether PID there is one running mysql so after reboot there was an edge case where "old" pid that's still in old file after crash was used by new app, mysql didn't start coz "simple" sysv script thought it's already running
Then there is dependencies that make a lot more cases simple like "start the service to get the encryption keys before you try to mount the encrypted volume" is just few deps, not modifying any code, and auto restart removed our use of monit
I do dislike the vast suite of stuff that worked fine before that comes with it and journald design is a bit shit, especially disk format (still have bug open when running status on app can take tens of seconds on spinning rust, why they just didn't use SQLite and indexes...) but overall, without exaggerating, it allowed us to remove tens of thousands f lines of code of "simple" sysv stuff that needed to be fixed.