Container Build Info URL: https://github.com/fkluknav/fedora-init-container Description: Image to run daemons as systemd units easily Fedora Account System Username: fkluknav
For the record: the slowness of systemd happens only in Fedora 25, not Fedora 24.
LGTM
Systemd guys please review.
Specifically look at https://github.com/fkluknav/fedora-init-container/blob/master/Dockerfile
On a freshly installed and updated centos virtual machine: # docker run --rm -it --name kontis fedora-init # docker exec -it kontis systemd-analyze critical-chain 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 @20.135s └─systemd-user-sessions.service @18.127s +1.004s └─basic.target @16.118s └─sockets.target @13.107s └─dbus.socket @12.104s └─sysinit.target @11.100s └─systemd-update-utmp.service @9.086s +1.009s └─systemd-tmpfiles-setup.service @7.076s +1.004s └─systemd-journal-flush.service @47ms +6.024s └─systemd-journald.service @23ms +21ms └─systemd-journald-dev-log.socket @22ms └─-.mount @18ms └─system.slice @20ms └─-.slice @18ms On RHEL7 (updated, not freshly installed): # docker exec condescending_poincare systemd-analyze critical-chain 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 @19.142s └─systemd-user-sessions.service @16.129s +2.009s └─basic.target @15.125s └─sockets.target @14.121s └─dbus.socket @13.118s └─sysinit.target @10.107s └─systemd-update-utmp.service @8.098s +1.005s └─systemd-tmpfiles-setup.service @6.089s +1.003s └─systemd-journal-flush.service @53ms +5.031s └─systemd-journald.service @28ms +23ms └─system.slice @27ms └─-.slice @24ms It does not run at all on a freshly installed and updated fedora 25 virtual machine.
Could you please retry with the current rawhide version or fedora 26?