Re: networkmanager spawned dhcpcd and kill it consistantly?



On Tue, Jun 12, 2012 at 9:35 AM, cheng renquan <crquan gmail com> wrote:
>> Any chance you can stop NM, and then run it manually with --no-daemon
>> --log-level=debug for us?  The logs below don't provide enough
>> information to figure out if NM is terminating the dhcpcd process or
>> not, and if it is, for what reason.  But the debug logs would:
>>
>> NetworkManager --no-daemon --log-level=debug
>
> it seems the same, log-level debug didn't give more logs, the dhcpcd
> is still "received SIGTERM, stopping" and "wlan0: removing interface",
> then NetworkManager[2658] may have detected that, then spawned a new
> dhcpcd but a few seconds later that received SIGTERM again;
>
> till I open another terminal, `kill -KILL 2658`; then the dhcpcd could
> run for long time;

last time running and NetworkManager[2714] killed by another terminal
`kill -KILL 2714`:

# /usr/sbin/NetworkManager --no-daemon --log-level=debug
[...]
NetworkManager[2714]:    keyfile: parsing ATT448 ...
NetworkManager[2714]:    keyfile:     read connection 'ATT448'
dhcpcd[2718]: version 5.5.6 starting
dhcpcd[2718]: all: not configured to accept IPv6 RAs
dhcpcd[2718]: wlan0: rebinding lease of 192.168.1.121
dhcpcd[2718]: wlan0: acknowledged 192.168.1.121 from 192.168.1.254
dhcpcd[2718]: wlan0: checking for 192.168.1.121
dhcpcd[2718]: wlan0: leased 192.168.1.121 for 86400 seconds
Killed

# /usr/sbin/NetworkManager --version
0.9.4.0


The weird thing is this is only happening at home with ATT448, that is
an AT&T internet device WIFI with WPA authentication;

when I am working with other WIFI providers, I see NetworkManager
launched only one dhcpcd and running well


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]