9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* Re: [9fans] stealing hard drive's contents
@ 2005-11-21  4:07 YAMANASHI Takeshi
  2005-11-21 13:12 ` Russ Cox
  0 siblings, 1 reply; 6+ messages in thread
From: YAMANASHI Takeshi @ 2005-11-21  4:07 UTC (permalink / raw)
  To: 9fans

On Mon Nov 21 12:13:19 JST 2005, Russ Cox wrote:
> > But the network has to be stable.  If venti/copy stops in the way,
> > bad things can happen. like, you've got the root score copied to
> > the local venti but not the whole tree.
> 
> Did this happen to you?  The venti/copy code intends this
> to be impossible.

I think it happened.

vtWrite in venti/copy.c always gets called regardless to
the result of walk()s in the switch clause starting at
/sys/src/cmd/venti/copy.c:70.  Or am I misreading the source?
-- 
"on travel, off the network ... and a fossil in my pocket"




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9fans] stealing hard drive's contents
  2005-11-21  4:07 [9fans] stealing hard drive's contents YAMANASHI Takeshi
@ 2005-11-21 13:12 ` Russ Cox
  0 siblings, 0 replies; 6+ messages in thread
From: Russ Cox @ 2005-11-21 13:12 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

> vtWrite in venti/copy.c always gets called regardless to
> the result of walk()s in the switch clause starting at
> /sys/src/cmd/venti/copy.c:70.  Or am I misreading the source?

This is true, and that's intentional.  If the block is missing
on the one server, venti/copy prints an error and then
keeps going.  That's so that if there is a block missing,
you still copy what you can.

If the network flakes out mid-copy, it won't just affect
one or two blocks and then come back.  The session
is being run over a single tcp connection with no
redialing, so if you lose the network during one block,
you're not going to be able to copy any of the rest of the
blocks either.

Oh!  I see what happened.  If there is a read error due to
network failure (as opposed to do the block just being
corrupt on the remote server) then venti/copy needs to
call sysfatal to avoid vtwriting the block we were in the
middle of.

I'll fix this.  Thanks for the report.

Russ


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9fans] stealing hard drive's contents
  2005-11-21  2:58 YAMANASHI Takeshi
  2005-11-21  3:13 ` Russ Cox
@ 2005-11-21  9:41 ` Dave Eckhardt
  1 sibling, 0 replies; 6+ messages in thread
From: Dave Eckhardt @ 2005-11-21  9:41 UTC (permalink / raw)
  To: 9fans

> But the network has to be stable.  If venti/copy stops in the way,
> bad things can happen.  like, you've got the root score copied to
> the local venti but not the whole tree.

I had a venti/copy go bad on me once, and it didn't seem to me as if
there were network problems at the time.  I *think* I used copy -f;
as a sanity check I tried to tar some random old dump trees to
/dev/null and found blocks missing.  So I fell back to copying
each arena from the old machine to the new one and feeding it to
the new venti with wrarena (the new venti had different sizes for
everything, so I couldn't just use the old arenas).

In hindsight I probably should have reported the problem; I didn't
because of uncertainty that what I was trying to do made sense
and because of "new venti soon" rumors...

Dave Eckhardt


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9fans] stealing hard drive's contents
  2005-11-21  2:58 YAMANASHI Takeshi
@ 2005-11-21  3:13 ` Russ Cox
  2005-11-21  9:41 ` Dave Eckhardt
  1 sibling, 0 replies; 6+ messages in thread
From: Russ Cox @ 2005-11-21  3:13 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

> But the network has to be stable.  If venti/copy stops in the way,
> bad things can happen. like, you've got the root score copied to
> the local venti but not the whole tree.

Did this happen to you?  The venti/copy code intends this
to be impossible.

Russ


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [9fans] stealing hard drive's contents
@ 2005-11-21  2:58 YAMANASHI Takeshi
  2005-11-21  3:13 ` Russ Cox
  2005-11-21  9:41 ` Dave Eckhardt
  0 siblings, 2 replies; 6+ messages in thread
From: YAMANASHI Takeshi @ 2005-11-21  2:58 UTC (permalink / raw)
  To: 9fans

> plan9.ucalgary.ca is going away and i need to move its venti to a
> newer, better hard drive. given that i'm some 3000km away from it and
> there's nobody there to fix things if they hit the fan, what's the
> best way of moving the data onto a local venti?

I just venti/copy'ied from BL to tip9ug venti.  I didn't care about
if the time is off-peak or not. :)

But the network has to be stable.  If venti/copy stops in the way,
bad things can happen. like, you've got the root score copied to
the local venti but not the whole tree.
-- 
"on travel, off the network ... and a fossil in my pocket"




^ permalink raw reply	[flat|nested] 6+ messages in thread

* [9fans] stealing hard drive's contents
@ 2005-11-21  2:47 andrey mirtchovski
  0 siblings, 0 replies; 6+ messages in thread
From: andrey mirtchovski @ 2005-11-21  2:47 UTC (permalink / raw)
  To: Fans of the OS Plan 9 from Bell Labs

it is indeed a silly question already answered in the man pages, but i
need a short tutorial because i really don't want to screw things up.

plan9.ucalgary.ca is going away and i need to move its venti to a
newer, better hard drive. given that i'm some 3000km away from it and
there's nobody there to fix things if they hit the fan, what's the
best way of moving the data onto a local venti? is there a simple
command that will let me move out several blocks at a time during the
off-peak hours, say, vac?

the disk has grown over the past few years and has information i would
not like to lose. has anyone else had a success story? please share!

andrey

ps: if you ever see david leymbach get him to give you some of the
Verrari PR mints he has. best mint ever! :)


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2005-11-21 13:12 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-11-21  4:07 [9fans] stealing hard drive's contents YAMANASHI Takeshi
2005-11-21 13:12 ` Russ Cox
  -- strict thread matches above, loose matches on Subject: below --
2005-11-21  2:58 YAMANASHI Takeshi
2005-11-21  3:13 ` Russ Cox
2005-11-21  9:41 ` Dave Eckhardt
2005-11-21  2:47 andrey mirtchovski

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).