Re: Cannot obtain lock file, aborting.

Am 03.06.2016 um 07:30 schrieb thufir:
> It seems like fetchnews isn't running, but neither can I start it.
> 
> thufir@mordor:~$ 
> thufir@mordor:~$ telnet localhost 119
> Trying 127.0.0.1...
> Connected to localhost.
> Escape character is '^]'.
> 200 Leafnode NNTP Daemon, version 1.11.10 running at localhost (my fqdn: 
> mordor.bounceme.net)
> quit
> 205 Always happy to serve!
> Connection closed by foreign host.
> thufir@mordor:~$ 
> thufir@mordor:~$ telnet mordor.bounceme.net 119
> Trying 127.0.1.1...
> Connected to mordor.bounceme.net.
> Escape character is '^]'.
> 200 Leafnode NNTP Daemon, version 1.11.10 running at mordor.bounceme.net 
> (my fqdn: mordor.bounceme.net)
> quit
> 205 Always happy to serve!
> Connection closed by foreign host.
> thufir@mordor:~$ 
> thufir@mordor:~$ ps -ae | grep fetch
> thufir@mordor:~$ 
> thufir@mordor:~$ sudo fetchnews
> Cannot obtain lock file, aborting.
> thufir@mordor:~$ 
> 
> 
> /var/log/news/news.* just shows the same error about obtaining the lock 
> file.
> 
> Should I find this lock file and delete, or delete and then recreate with 
> touch, perhaps?

This usually happens if there's a stale lock file that contains the PID
and hostname and the hostname matched, but there is another process
running that reused the PID.

If you are sure that it's _not_ a leafnode-related process such as
fetchnews or texpire that set the lock file, you can delete the latter.

(Leafnode programs only check if there is a process with the same PID,
not if it's themselves.)

Another possibility is file system corruption.


The reboot you made later would probably have reshuffled your process
IDs so that fetchnews could then figure the lock file was stale, meaning
there was no process with the PID in the lock file, and then deleted the
lock file.

There's some randomness in PIDs, either deliberate by the OS, or through
the inherent randomness in timing.

HTH
Matthias