Andrew Deason <[email protected]> writes:
> Russ Allbery <[email protected]> wrote:

>> Is there any reason not to do this?  If not, I can just make this
>> change in the Debian package.  I don't recall why start-stop-daemon was
>> used there in the first place.

> Hmm, I can't think of much. Does start-stop-daemon do anything like
> clean the environment, etc? Has afsd always exited after mount?

I suspect that at some point in the very distant past, afsd didn't exit
after mount.  Checking the Git history, Debian has been using
start-stop-daemon to start afsd since 2000-11-05, which would correspond
to 1.0.snap20001106-2.  I have no idea how afsd worked then.  :)

However, start-stop-daemon itself doesn't background anything unless you
tell it to explicitly, and we aren't telling it to.  So I think there's
some other problem here; it looks from the original bug report like afsd
itself is exiting before the cache is set up and AFS is mounted.

> Does it matter if some afsd children only fork() without doing anything
> like daemon()izing?

Yes, probably.  At least you probably don't want them attached to a tty or
any of the other bad things that can happen if they don't daemonize.

-- 
Russ Allbery ([email protected])             <http://www.eyrie.org/~eagle/>
_______________________________________________
OpenAFS-info mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-info

Reply via email to