On 06/11/2018 20.14, Michael Fischer wrote:
On Tue, Nov 06, Carlos E. R. wrote:
On 06/11/2018 14.12, Michael Fischer wrote:
Try with critical chain when you can, please.
Below.
The time after the unit is active or started is printed after the "@" character. The time the unit takes to start is printed after the "+" character.
multi-user.target @6.446s └─cron.service @6.446s └─postfix.service @6.308s +137ms └─time-sync.target @6.307s └─ntpd.service @6.290s +17ms └─network.target @6.287s └─wicked.service @963ms +5.323s └─wickedd-nanny.service @960ms +2ms └─wickedd.service @955ms +4ms └─wickedd-dhcp4.service @934ms +18ms └─dbus.service @907ms └─basic.target @902ms └─sockets.target @902ms └─dbus.socket @902ms └─sysinit.target @902ms └─systemd-update-utmp.service @897ms +4ms └─systemd-tmpfiles-setup.service @786ms +109ms └─local-fs.target @784ms └─home.mount @598ms +185ms └─systemd-fsck@dev-disk-by\x2duuid-7d15344a\x2dca83\x2d4d3d\x2d8e72\x2d54dc4d2bc2cc.service @432ms +165ms └─dev-disk-by\x2duuid-7d15344a\x2dca83\x2d4d3d\x2d8e72\x2d54dc4d2bc2cc.device @431ms
Well, I do not see the problem. I see something I do not understand, though: multi-user.target starts at 6 seconds (@6.446s). A bit later, wicked.service starts at 0.9 seconds (@963ms) and takes 5 seconds to complete (+5.323s). How can it start at 1 second after boot, thus before than "multi-user.target", and be later in the path? I must be reading it incorrectly. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)