How To Stop Yum When Networkmanager Has Broken Resolv.conf?
So, I’m getting an error where the network service and NetworkManager apparently don’t agree on how to bring up vlans on bonded nics. Things come up if you ‘ifup ..’ manually. I thought I’d check if there were any updates, forgetting to fix what NetworkManger had done to /etc/resolv.conf and:
http://CentOS.arvixe.com/7.0.1406/os/x86_64/repodata/repomd.xml:
[Errno 12] Timeout on http://CentOS.arvixe.com/7.0.1406/os/x86_64/repodata/repomd.xml: (28,
‘Resolving timed out after 30381 milliseconds’)
Trying other mirror.
^C^C^C
^C^C^C
Current download cancelled, interrupt (ctrl-c) again within two seconds to exit.
Yeah – I did hit it about 6 times there. What do you have to do to make it actually stop in some reasonable amount of time?
5 thoughts on - How To Stop Yum When Networkmanager Has Broken Resolv.conf?
try CTRL-Z to put that running process in your shell to sleep, then:
bg 1
to put the process in the background. Then you can run:
killall -TERM yum
to kill all the yum processes. If that doesn’t kill it after running a couple of times, then use:
killall -9 yum
and that should do it.
Chris
Oh, and if you don’t have killall installed, you can use:
ps -ef | grep yum
to get the pid and then use:
kill -TERM pid
or easier even is:
kill -TERM `pid of yum`
Chris
oops – it’s Friday – that should be:
kill -TERM `pidof yum`
no space between pid and of.
Chris
You’re missing a % there
Or just “control-Z” and then ‘kill -9 %1’
You don’t need to background the job and then kill every process matching a name.
I probably would have just ssh’d in from a different window, except that SSH logins take forever with DNS broken. Thanks – and I think I
found the source of the network issue. I had a typo in the NM_CONTROLLED=no line in the ifcfg-bond0 file and NetworkManger was failing with:
Error: no device found for connection ‘System bond0’ when it shouldn’t even have been involved.