[9fans] issuing fshalt shuts down rio and gives the following

[9fans] issuing fshalt shuts down rio and gives the following

[ROuNIN]

Hello,

Recently, I noticed a change from what I do not recall to be the
behavoiur I saw some time ago

When I issue an fshalt, it shuts down rio and give the following
outside rio
init: starting /bin/rc
term%

ROuNIN

Re: [9fans] issuing fshalt shuts down rio and gives the following

[David du Colombier]

> Recently, I noticed a change from what I do not recall to be the
> behavoiur I saw some time ago
>
> When I issue an fshalt, it shuts down rio and give the following
> outside rio
> init: starting /bin/rc
> term%

Yes, I got surprised when I first saw this change in may 2011,
but it's just normal behavior.

It kills rio and get back to cga mode to be able to reboot properly.

--
David du Colombier

Re: [9fans] issuing fshalt shuts down rio and gives the following

[Peter A. Cejchan]

[-- Attachment #1: Type: text/plain, Size: 78 bytes --]

is it okay to do ^T^Tr (or, CtlAltDel) then? I guess it is, however..
++pac

[-- Attachment #2: Type: text/html, Size: 84 bytes --]

Re: [9fans] issuing fshalt shuts down rio and gives the following

[David du Colombier]

> is it okay to do ^T^Tr (or, CtlAltDel) then? I guess it is, however..

Yes. You just have to wait for your file system
to synchronize and halt.

--
David du Colombier

Re: [9fans] issuing fshalt shuts down rio and gives the following

[erik quanstrom]

On Thu Jan 19 06:31:30 EST 2012, 0intro@gmail.com wrote:
> > Recently, I noticed a change from what I do not recall to be the
> > behavoiur I saw some time ago
> >
> > When I issue an fshalt, it shuts down rio and give the following
> > outside rio
> > init: starting /bin/rc
> > term%
>
> Yes, I got surprised when I first saw this change in may 2011,
> but it's just normal behavior.
>
> It kills rio and get back to cga mode to be able to reboot properly.

given that output, it is not clear that fshalt is really halting.
(expect, using kernel $fu...)

can you still run normal commands afterwards, or has your fs
shut down?  you may wish to sprinkle some echoes in the
fshalt script and see where things fail.

- erik

Re: [9fans] issuing fshalt shuts down rio and gives the following

[Peter A. Cejchan]

[-- Attachment #1: Type: text/plain, Size: 228 bytes --]

I've just checked it out, just boot/login/logout:
the terminal went to 'black' (== txt) mode with
   prompt:
      fsys all sync
      main sync: wrote  blocks     ## why??
      (then, cursor blinks until <Enter>)
term%

[-- Attachment #2: Type: text/html, Size: 294 bytes --]

Re: [9fans] issuing fshalt shuts down rio and gives the following

[David du Colombier]

> I've just checked it out, just boot/login/logout:
> the terminal went to 'black' (== txt) mode with
>    prompt:
>       fsys all sync
>       main sync: wrote  blocks     ## why??

fshalt runs fossilcons(4) sync to flush the cache,
before halting the file system.

>       (then, cursor blinks until <Enter>)
> term%

It looks fine.

--
David du Colombier

Re: [9fans] issuing fshalt shuts down rio and gives the following

[Peter A. Cejchan]

[-- Attachment #1: Type: text/plain, Size: 170 bytes --]

why did it write anything wnen nothing was changed? just login/fshalt...

> fshalt runs fossilcons(4) sync to flush the cache,

> before halting the file system.
>

[-- Attachment #2: Type: text/html, Size: 349 bytes --]

Re: [9fans] issuing fshalt shuts down rio and gives the following

[David du Colombier]

> why did it write anything wnen nothing was changed? just
> login/fshalt...

It is likely to be atime updates.

--
David du Colombier

Re: [9fans] issuing fshalt shuts down rio and gives the following

[erik quanstrom]

> why did it write anything wnen nothing was changed? just login/fshalt...
>

atime updates.

i'm not convinced that everything is fine.  are you running
the current version of fshalt, fossil?  there are a couple
of expect scripts hiding in fshalt, and they could easily
be confused.

- erik

Re: [9fans] issuing fshalt shuts down rio and gives the following

[tlaronde]

On Thu, Jan 19, 2012 at 09:38:40AM -0500, erik quanstrom wrote:
> > why did it write anything wnen nothing was changed? just login/fshalt...
> >
>
> atime updates.
>
> i'm not convinced that everything is fine.  are you running
> the current version of fshalt, fossil?  there are a couple
> of expect scripts hiding in fshalt, and they could easily
> be confused.

Since I have a vanilla fresh install, I can tell that what is described
is the default behavior. The fs does sync and halt.

The only curious thing is that "fshalt -r" does reboot, but reboots say
Plan9, and not the computer (restarts from 9load it seems). [I'm not
saying it should not or whatever; it is a constatation and it probably
makes sense for a Plan9 terminal.]
--
        Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
                      http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

Re: [9fans] issuing fshalt shuts down rio and gives the following

[David du Colombier]

> The only curious thing is that "fshalt -r" does reboot, but reboots
> say Plan9, and not the computer (restarts from 9load it seems). [I'm
> not saying it should not or whatever; it is a constatation and it
> probably makes sense for a Plan9 terminal.]

It reloads the current kernel using #c/reboot.

--
David du Colombier

Re: [9fans] issuing fshalt shuts down rio and gives the following

[erik quanstrom]

On Thu Jan 19 10:06:50 EST 2012, 0intro@gmail.com wrote:
> > The only curious thing is that "fshalt -r" does reboot, but reboots
> > say Plan9, and not the computer (restarts from 9load it seems). [I'm
> > not saying it should not or whatever; it is a constatation and it
> > probably makes sense for a Plan9 terminal.]
>
> It reloads the current kernel using #c/reboot.

not correct.  it loads the current default kernel,
from /n/boot/$cputype/9xxx.  where xxx is determined
by peeking at $terminal.

- erik

Re: [9fans] issuing fshalt shuts down rio and gives the following

[Peter A. Cejchan]

[-- Attachment #1: Type: text/plain, Size: 412 bytes --]

... there've been some issues around pull considering files as locally
modified while they weren't... thus, I have some outdated versions of
everything ... I'll try to edit the pull's output to pull -s list, then
we'll see.
Although unhappy with that, I am spending much time now on linux, writing
in 'Go', as I depend on packages using features introduced after Go 60.2,
sigh...
Thanks, regards,
++pac

[-- Attachment #2: Type: text/html, Size: 451 bytes --]

Re: [9fans] issuing fshalt shuts down rio and gives the

[ROuNIN]

Why can we not re-instate the following messages at least the "done
halting":

term% fshalt
syncing.../srv/fscons...prompt: venti...
halting.../srv/fscons...
prompt:
done halting

ROuNIN

Re: [9fans] issuing fshalt shuts down rio and gives the

[Nicolas Bercher]

On 30/01/2012 10:44, ROuNIN wrote:
> Why can we not re-instate the following messages at least the "done
> halting":
>
> term% fshalt
> syncing.../srv/fscons...prompt: venti...
> halting.../srv/fscons...
> prompt:
> done halting

That an important point to me: how can I be sure all of the fshalt stuff
is done for good?  I'm currently setting up a Plan 9 cpu/fs/auth with
recent iso and this really confused to me at first.

Actually, after (too) many reboots, I saw "done halting" only once.

Nicolas