Red Hat Bugzilla – Bug 147866
haldaemon dies leaving no indication why
Last modified: 2013-03-05 22:42:41 EST
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5) Gecko/20041111 Firefox/1.0 Description of problem: haldaemon starts and immediately dies. Version-Release number of selected component (if applicable): hal-0.4.7-1.FC3 How reproducible: Always Steps to Reproduce: 1.service haldaemon start 2. 3. Additional info: [root@tux ~]# hald --daemon=no --verbose=yes 16:56:08.850 [I] hald.c:394: hal 0.4.7 16:56:08.851 [I] hald.c:398: Will not daemonize 16:56:08.851 [E] hald_dbus.c:1934: dbus_bus_get(): Failed to connect to socket /var/run/dbus/system_bus_socket: Connection refused [root@tux ~]# cd /var/run/dbus [root@tux dbus]# ls system_bus_socket [root@tux dbus]# ls -l system* srwxrwxrwx 1 root root 0 Feb 11 15:38 system_bus_socket
hal can't start because dbus-daemon didn't stay up. Found a stale /var/run/messagebus.pid file that prevented dbus from starting. Removed the file, started messagebus, and then started haldaemon. Both seem fine now. Will test during reboot. Seems that /etc/rc.d/init.d/messagebus could stand to look for and remove stale pidfiles (being careful to differentiate between the system-wide and user daemons) like more well-behaved members of the family... Side note: Why have RH taken to renaming the system services to something other than the binaries. Has the desktop application renaming craze struck the system processes team as well? Seems counterproductive to me.
David, I too have this condition, about 30 minutes to a hour after boot HAL just dies, it's getting frustrating, is it possible these symptoms are the result of the memory leak that you recently found and posted a patch for? Any thoughts? Steve
The symtoms described in comment 1 seems to be a D-BUS bug, not a hal bug. Steve, about comment 2, I'm not sure why the hal daemon would die after 30 minutes. If you can reproduce this please open a new bug. Thanks.