Created attachment 357275 [details] network manager Description of problem: Network connection is stopped after service Networkmanager was stopped Version-Release number of selected component (if applicable): NetworManager-0.7.995-1 How reproducible: 100% Steps to Reproduce: 1.Boot into system from f12 LiveCD 2.check network connection and make sure it's active 3.stop service NetworkManager,then network connection is stopped. start NetworkManager,then network connection is available. Actual results: Expected results: https://fedoraproject.org/wiki/QA:Testcase_NetworkManager_assume Additional info:
*** Bug 517267 has been marked as a duplicate of this bug. ***
The problem is resolved with package NetworkManager0.7.995-3.git20090813.fc12.x86_64 - http://koji.fedoraproject.org/koji/buildinfo?buildID=127187
I still have the same problem with 0.7.995-3.git20090813.fc12 on an installed system. [root@ventus ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0 # Broadcom Corporation BCM4401-B0 100Base-TX DEVICE=eth0 BOOTPROTO=dhcp HWADDR=00:19:B9:4E:AA:0A ONBOOT=yes PEERDNS=yes PEERROUTES=yes TYPE=Ethernet UUID=5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03 Aug 13 15:51:09 localhost NetworkManager: <WARN> nm_signal_handler(): Caught signal 15, shutting down normally. Aug 13 15:51:09 localhost NetworkManager: <info> (wlan0): now unmanaged Aug 13 15:51:09 localhost NetworkManager: <info> (wlan0): device state change: 2 -> 1 (reason 36) Aug 13 15:51:09 localhost NetworkManager: <info> (eth0): now unmanaged Aug 13 15:51:09 localhost NetworkManager: <info> (eth0): device state change: 8 -> 1 (reason 36) Aug 13 15:51:09 localhost NetworkManager: <info> (eth0): deactivating device (reason: 36). Aug 13 15:51:09 localhost NetworkManager: <info> (eth0): canceled DHCP transaction, dhcp client pid 4797 Aug 13 15:51:10 localhost NetworkManager: <WARN> check_one_route(): (eth0) error -34 returned from rtnl_route_del(): Sucess#012 Aug 13 15:51:10 localhost NetworkManager: <info> (eth0): cleaning up... Aug 13 15:51:10 localhost NetworkManager: <info> (eth0): taking down device. Aug 13 15:51:10 localhost kernel: b44: eth0: powering down PHY Aug 13 15:51:10 localhost NetworkManager: <info> exiting (success)
Okay, basically seeing this work with now with eth0 (interface stays up). However, it still brings down wlan0, even though I've configured it as a "system" interface. Perhaps this cannot be avoided, though there isn't much explanation as to why wlan0 goes down. /var/log/messages shows: Sep 22 16:31:13 localhost NetworkManager: <WARN> nm_signal_handler(): Caught signal 15, shutting down normally. Sep 22 16:31:13 localhost avahi-daemon[969]: Interface wlan0.IPv4 no longer relevant for mDNS. Sep 22 16:31:13 localhost avahi-daemon[969]: Leaving mDNS multicast group on interface wlan0.IPv4 with address 192.168.0.34. Sep 22 16:31:13 localhost dhclient: receive_packet failed on wlan0: Network is down Sep 22 16:31:13 localhost avahi-daemon[969]: Withdrawing address record for fe80::219:d2ff:fe48:ad08 on wlan0. Sep 22 16:31:13 localhost avahi-daemon[969]: Withdrawing address record for 192.168.0.34 on wlan0. Sep 22 16:31:14 localhost NetworkManager: <info> exiting (success) NetworkManager-0.7.996-2.git20090921.fc12.x86_64
Hmm, that /var/log/messages shows something sending SIGTERM to NetworkManager, which causes NM to quit. Is that somebody running 'service NetworkManager stop' somewhere?
Gah, ignore that, wrong bug. Only wired non-802.1x connections are preserved. Wifi connections have entirely too much state to be kept up.
In any case, fixed in rawhide. thanks for the testing.