Showing status of service via systemctl is slow (>10s) if disk journal is used #2460
Description
With big (4GB, few months of logs) on-disk journal, systemctl status service becomes very slow
(!) [13:37:30]:/var/log/journal☠ time systemctl status nginx
* nginx.service - A high performance web server and a reverse proxy server
Loaded: loaded (/lib/systemd/system/nginx.service; enabled; vendor preset: enabled)
Active: active (running) since Thu 2016-01-28 08:25:28 CET; 5h 12min ago
Main PID: 3414 (nginx)
CGroup: /system.slice/nginx.service
|-3414 nginx: master process /usr/sbin/nginx -g daemon on; master_process on;
|-3415 nginx: worker process
|-3416 nginx: worker process
|-3417 nginx: worker process
`-3418 nginx: worker process
Jan 28 08:25:12 ghroth systemd[1]: Starting A high performance web server and a reverse proxy server...
Jan 28 08:25:28 ghroth systemd[1]: Started A high performance web server and a reverse proxy server.
real 0m12.505s
user 0m0.016s
sys 0m0.056s
(!) [13:35:14]:/var/log/journal☠ du -h --max=1
4,2G ./46feef66e59512fcd99c7ddc00000108
4,2G .
(!) [13:40:53]:/var/log/journal☠ strace systemctl status nginx 2>&1 |grep open |grep /var/log |wc -l
88
it is of course faster after it gets to cache... for that service, querying other one is still slow.
Dunno what would be right way to do it.. but opening ~80 log files to just display service status seems a bit excessive