Hi Arnd,

The trace logging did not help - on the restarts when the proftpd 
service does not start absolutely nothing is logged to the trace log 
file. Which would seem to indicate to me that the service does not even 
attempt to initialise.

However, I have made some other progress. Having read lots of proftpd 
forum threads I came across one that was discussing profptd not starting 
on Karmic that mentioned a problem with an unknown runlevel. There was 
no resolution on the thread but it did prompt me to investigate.

I have found that:

a. On the restarts when proftpd starts up successfully the output of the 
runlevel command looks normal i.e.

gra...@gt-desktop:~$ runlevel
N 2


b. On the restarts when proftpd does not start up the output of the 
runlevel command is not normal, it gives this:

gra...@gt-desktop:~$ runlevel
unknown

I have also noticed that in scenario b. when proftpd does not start, 
neither does the printing daemon cupsd.

The above two scenarios are absolutely consistent over about 50 restarts 
I have now done.

So this in fact is looking to be not a proftpd bug at all, but something 
going wrong with the init process - with proftpd and cupsd failing to 
start being a symptom of that, along with the failure to export the 
runlevel env.

For clarity, although the thread I came across the runlevel discussion 
on was relating to Karmic, I am running Lucid.

Now I need to try and understand how to debug the init process to 
continue investigating - something that I have never looked at, so I'm 
learning all the way from here, which is a good thing :)

Should this bug be marked as Invalid now because it actually appears not 
to be a proftpd bug?

Thanks for all your help.

Regards,
Graham

-- 
proftpd will not start with kernel 2.6.32-22-generic
https://bugs.launchpad.net/bugs/591865
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to