9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* [9fans] bringing down vacfs in p9p
@ 2006-03-09 10:08 Eoghan Sherry
  2006-03-09 12:57 ` erik quanstrom
  2006-03-09 14:01 ` Russ Cox
  0 siblings, 2 replies; 3+ messages in thread
From: Eoghan Sherry @ 2006-03-09 10:08 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

I can't figure out how to gracefully stop vacfs (or any other
file server for that matter) in p9p.

In plan 9 I can do an rm /srv/vacfs.test which immediately
removes the file causing vacfs to exit.

It'd be nice if rm `{namespace}^/vacfs.test would do the
same in p9p but it doesn't since unix won't remove the socket file
until vacfs closes it. I've resorted to 9 kill vacfs | rc
which doesn't seem right.

Is there something I'm missing? Is kill not that bad?

eoghan


^ permalink raw reply	[flat|nested] 3+ messages in thread

* [9fans] bringing down vacfs in p9p
  2006-03-09 10:08 [9fans] bringing down vacfs in p9p Eoghan Sherry
@ 2006-03-09 12:57 ` erik quanstrom
  2006-03-09 14:01 ` Russ Cox
  1 sibling, 0 replies; 3+ messages in thread
From: erik quanstrom @ 2006-03-09 12:57 UTC (permalink / raw)
  To: 9fans, Eoghan Sherry

the unix domain sockets in `{namespace} don't need to be removed;
9pserve -- not vacfs -- creats the socket file. it is not removed,
regardless of how things exit.

since vacfs doesn't do rfork(RFNOTEG), hitting delete in the 9term
window from which vacfs was started should kill vacfs off.

i don't see how that could be dangerous since vacfs is only a cache.

- erik

"Eoghan Sherry" <ejsherry@gmail.com> writes

|
| I can't figure out how to gracefully stop vacfs (or any other
| file server for that matter) in p9p.
|
| In plan 9 I can do an rm /srv/vacfs.test which immediately
| removes the file causing vacfs to exit.
|
| It'd be nice if rm `{namespace}^/vacfs.test would do the
| same in p9p but it doesn't since unix won't remove the socket file
| until vacfs closes it. I've resorted to 9 kill vacfs | rc
| which doesn't seem right.
|
| Is there something I'm missing? Is kill not that bad?
|
| eoghan


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [9fans] bringing down vacfs in p9p
  2006-03-09 10:08 [9fans] bringing down vacfs in p9p Eoghan Sherry
  2006-03-09 12:57 ` erik quanstrom
@ 2006-03-09 14:01 ` Russ Cox
  1 sibling, 0 replies; 3+ messages in thread
From: Russ Cox @ 2006-03-09 14:01 UTC (permalink / raw)
  To: 9fans

> In plan 9 I can do an rm /srv/vacfs.test which immediately
> removes the file causing vacfs to exit.

This is only true if vacfs is not mounted anywhere.
If someone's using it, removing the srv file won't
make vacfs exit.  Vacfs exits when reading from its
9P pipe returns eof.  That happens on Plan 9 when
all references to the other end of the pipe have
gone away.

> It'd be nice if rm `{namespace}^/vacfs.test would do the
> same in p9p but it doesn't since unix won't remove the socket file
> until vacfs closes it. I've resorted to 9 kill vacfs | rc
> which doesn't seem right.
>
> Is there something I'm missing? Is kill not that bad?

Kill is not that bad.  If you wanted to simulate Plan 9
and force an eof on the 9P pipe, you could kill the
appropriate 9pserve.

I can't think of any file servers that need to be shut
down gracefully on p9p except acme (which has an
Exit command).

Russ



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2006-03-09 14:01 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-03-09 10:08 [9fans] bringing down vacfs in p9p Eoghan Sherry
2006-03-09 12:57 ` erik quanstrom
2006-03-09 14:01 ` Russ Cox

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).