https://bugzilla.novell.com/show_bug.cgi?id=727771
https://bugzilla.novell.com/show_bug.cgi?id=727771#c35
--- Comment #35 from Marius Tomaschewski
Hi
I have the same problem, but the fix has been revoked, So I'm unable to test.
Which fix has been revoked?
For me it works if I add: #/etc/init.d/network start -o debug
This works because there is no systemd involved / no redirection done by the /etc/rc.status script because of this guard: if test $PPID -ne 1 -a $# -eq 1 ; then [when not started by systemd and only one arg given -> redirect to systemd] but it causes another problem: systemd does not know anything about your start and continue to consider it dead/failed/not yet started, even a "rcnetwork status -o" will show it started. The problem (part of) is IMO, that systemd a) tries to resolve dependencies [generally a good thing], b) does not use status reported by the LSB scripts (does not call /etc/init.d/network status), but own meaning instead. Further, there is IMO no way to tell let it refresh the systemd state, e.g. via "systemctl reset-failed". It is seems to not work for LSB scripts, e.g. by executing the LSB status action... Additional problems may be caused by hooks installed by samba-client or ntp, that cause a systemd deadlock by executing init scripts that depend from the network script via hooks inside of the network scripts (if-up.d, ...), see bug 732930 alias bug 732395 alias bug 725503 alias ... Workarounds are (until release of fixes for samba-client, ...): - move the /etc/sysconfig/network/if-*.d/21* links away - remove ntp-runtime from NETCONFIG_MODULES_ORDER in /etc/sysconfig/network/config. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.