[p4] Deleting original depot without affecting branch

Gabor Maghera gmaghera at gmail.com
Wed Mar 5 13:33:45 PST 2008


Thanks for pointing out that article, Oren.  I now think I understand where
the confusion stems from.

Pre 2006.2:

   - When obliterate finds archive files that are referenced by lazy
   copies, the lazy copies are replaced with real files before the archive
   files are deleted from the server.


2006.2 and on:

   - When obliterate finds archive files that are referenced by lazy
   copies, the archive files are not deleted from the server.
   - Lazy copies are no longer replaced with real files because they will
   not be adversely affected by removal of the original archive files they were
   branched from.

Thanks to all who helped,
Gabor

On Tue, Mar 4, 2008 at 10:49 PM, Oren Shemesh (oshemesh) <oshemesh at cisco.com>
wrote:

> Having just tested this (On Server Version: 2007.2), I agree with
> William.
>
> This article perfectly explains the changes that confused some of us:
>
> http://kb.perforce.com/P4dServerReference/P4ClientReference/Obliterate
>
> Oren.
>
> -----Original Message-----
> From: Ivey, William [mailto:william_ivey at bmc.com]
> Sent: Tuesday, March 04, 2008 11:37 PM
> To: Gabor Maghera; Oren Shemesh (oshemesh); perforce-user at perforce.com
> Subject: RE: [p4] Deleting original depot without affecting branch
>
> You don't want to run p4 snap before using p4 obliterate, they
> are unrelated commands.
>
> P4 obliterate will correctly handle lazy copies in the most
> efficient way. P4 snap, on the other hand, will cause all
> lazy copies found in the given path to be replaced by real file
> copies. This will eat up repository space; possibly a great
> deal of space.
>
> P4 snap is overkill in most cases. Only if you want to make
> sure that the files are redundantly stored would you want to
> use it. This is not something most people will want to do
> for most files.
>
> -Wm
>
>
> -----Original Message-----
> From: perforce-user-bounces at perforce.com
> [mailto:perforce-user-bounces at perforce.com] On Behalf Of Gabor Maghera
> Sent: Tuesday, March 04, 2008 2:07 PM
>
> I am little bit confused about the necessity of running p4 snap when
> using
> p4 obliterate.  The help on p4 obliterate for version 2007.2, does not
> state
> anything about having to run p4 snap prior to using p4 obliterate.
>
> So, I wonder, in order to preserve revisions for which you have lazy
> copies
> in other branches, do you have to first run p4 snap?  Has there been a
> change to the obliterate algorithm going from p4d 2007.2 to p4d 2007.3,
> perhaps?
>
> Thanks,
> Gabor
>
> On Thu, Feb 28, 2008 at 7:33 AM, Oren Shemesh (oshemesh)
> <oshemesh at cisco.com>
> wrote:
>
> > Thanks Paul, indeed the latest 'p4 help obliterate' shows that quoting
> > from memory is a bad habit (at least when it comes to MY memory).
> >
> > Read 'p4 help snap'  and 'p4 help obliterate' - they explain
> everything
> > very clearly.
> >
> > Oren.
> >
> > -----Original Message-----
> > From: paul goffin [mailto:paul.goffin at dsl.pipex.com]
> > Sent: Thursday, February 28, 2008 12:17 PM
> > To: Oren Shemesh (oshemesh)
> > Cc: Jeff Desharnais; perforce-user at perforce.com
> > Subject: Re: [p4] Deleting original depot without affecting branch
> >
> > Not with recent versions of the server it doesn't.
> >
> > See "p4 help snap".
> >
> > Paul.
> >
> > Quoting "Oren Shemesh (oshemesh)" <oshemesh at cisco.com>:
> >
> > > Obliterate automatically does the work '-z' would do, if needed.
> > > You can safely use obliterate, just as you understood in the past.
> > >
>



More information about the perforce-user mailing list