FS#11345 - Network Manager initscript does not wait for daemon to stop, resulting in failed restarts
Attached to Project:
Arch Linux
Opened by Ryan (DarwinAwardWinner) - Saturday, 30 August 2008, 01:47 GMT
Last edited by Allan McRae (Allan) - Saturday, 30 August 2008, 02:17 GMT
Opened by Ryan (DarwinAwardWinner) - Saturday, 30 August 2008, 01:47 GMT
Last edited by Allan McRae (Allan) - Saturday, 30 August 2008, 02:17 GMT
|
Details
Description: When killing the Network Manager daemon, the
init script simply sends the kill signal and assumes the
daemon is dead. In fact, the daemon takes several seconds to
die after it receives the kill signal. When issuing a
restart, the init script does not wait long enough for the
daemon to die, so the start fails.
Steps to reproduce: # /etc/rc.d/networkmanager stop :: Stopping NetworkManager [DONE] # /etc/rc.d/networkmanager start :: Starting NetworkManager [FAIL] # sleep 10; /etc/rc.d/networkmanager start :: Starting NetworkManager [DONE] Solution: Make these changes to /etc/rc.d/networkmanager to ensure that the stop command waits until the daemon is dead: 37a38,42 > while [ ! -z "$PID" ]; do > PID=`pidof -o %PPID $NETWORKMANAGER_BIN` > sleep 1 > done > 47d51 < sleep 1 |
This task depends upon
Closed by Allan McRae (Allan)
Saturday, 30 August 2008, 02:17 GMT
Reason for closing: Duplicate
Additional comments about closing: FS#8259
Saturday, 30 August 2008, 02:17 GMT
Reason for closing: Duplicate
Additional comments about closing: