Package: ctdb Version: 2:4.4.5+dfsg-3 Severity: important Tags: newcomer
-- System Information: Debian Release: stretch/sid APT prefers testing APT policy: (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 4.5.0-2-amd64 (SMP w/2 CPU cores) Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) Versions of packages ctdb depends on: ii init-system-helpers 1.42 ii iproute2 4.3.0-1+b1 ii libbsd0 0.8.3-1 ii libc6 2.23-5 ii libpopt0 1.16-10 ii libtalloc2 2.1.7-1 ii libtdb1 1.3.9-1 ii libtevent0 0.9.28-1 ii lsb-base 9.20160629 ii psmisc 22.21-2.1+b1 ii samba-libs 2:4.4.5+dfsg-3 ii sudo 1.8.17p1-2 ii tdb-tools 1.3.9-1 ii time 1.7-25.1 Versions of packages ctdb recommends: ii ethtool 1:4.6-1 Versions of packages ctdb suggests: ii logrotate 3.8.7-2 ii lsof 4.89+dfsg-0.1 -- Configuration Files: /etc/default/ctdb changed: CTDB_RECOVERY_LOCK=/mnt/gv_main/samba/ctdb.lock CTDB_NODES=/etc/ctdb/nodes CTDB_PUBLIC_ADDRESSES=/etc/ctdb/public_addresses CTDB_MANAGES_SAMBA=yes CTDB_MANAGES_WINBIND=yes -- no debconf information as of ctdb package version 4.4.5+dfsg-3 there seems to be a bug in the ctdb.service script. With a configuration tested with the previous version "service ctdb start" fails after service start timeout. While starting and waiting for the timeout, ctdb service is usable but the startup scripts kill the processes if no pid file can be found. journalctl message: systemd[1]: ctdb.service: PID file /var/run/sambe/ctdb/ctdbd.pid not readable (yet?) after start: No such file or directory replacing the pidfile inside /lib/systemd/system/ctdb.service fixed the problem for me: sed -i 's/^PIDFile=\/var\/run\/samba\/ctdb\/ctdbd.pid$/PIDFile=\/run\/ctdb\/ctdbd.pid/' /lib/systemd/system/ctdb.service systemctl daemon-reload PS: offtopic (probably known issue): the following statement inside /etc/default/ctdb CTDB_MANAGES_SAMBA=yes can be made working with: cd /lib/systemd/system/ rm samba.service ; ln -s smbd.service samba.service
