Re: init scripts problems ?

"Also sprach denis bonnenfant:"
> I'm exporting two enbd device on the server, and it seems that enbd init 
> script is doing something wrong when restarting enbd link  :

Oh - I've just realized; you must be talking about the 2.6 kernel.

YES, I recall noticing a self-blocking problem as well in 2.6 - it's not
me!  The kernel never tells the driver anything about it. 

It looks to me (when I see it) like the second client can't actually
open the device until the first client has gone away (but the problem
condition never arrives at the driver in order to be seen!).  When that
happens, I do an echo 0 >/proc/nbdinfo and that temporarily downs and
ups the device, which seems to release whatever is the kernel semaphore
involved, and lets negotiation proceed.

This problem needs investigating!  When you have the two clients stuck
like that please please please get the WCHAN field from ps!  As far as I
recall, the stuck client is in a semaphore that I don't own!  So what
can I do?

I could double-check the code AGAIN to make sure that I'm not
accidentally providing single access only, but how can it be me if the
stuck semaphore is elsewhere in the kernel?



Peter