On Sun, Nov 14, 2010 at 2:16 AM, wrodrigues201 <wrodrigues201_at_gmail.com> wrote:
> On 11/08/2010 10:35 PM, Campbell Allan wrote:
>> On Monday 08 Nov 2010, wrodrigues201 wrote:
>>> Our subversion (1.4.3-r23084 on windows 2003) was holding around 1.6 TB
>>> of data and one user has accidentally deleted a directory of 1 TB. I
>>> have done a svn export from the previous version and have the data. Do I
>>> have to add and again commit this data ? Will it use up 1 TB of disk
>>> space on the svn server ? Is there any way i can restore the data from
>>> the previous version without using up 1 TB of disk space ?
>>> Thanks in advance.
>> If I understand correctly, nothing has been deleted from the server it
>> isn't anymore in the working copy? If that is the case then assuming it
>> wasn't too long ago or you do not mind redoing/merging the commits then
>> could take a copy of the trunk/branch prior to the delete and rename this
>> back. The delete would still have occurred but on the branch that no
>> matters and once you're happy that can be deleted too. This will only take
>> the space required for a few copies of the parent, nowhere near the 1TB of
>> the content. Something like
>> svn copy https://svnserver/svn/project/trunk@12345 \
>> svn move https://svnserver/svn/project/trunk \
>> svn move https://svnserver/svn/project/trunkcopy \
> Hello Allan,
> Thanks your suggestions worked like a charm.
And it saves the repository space of importing 1 Terabyte of raw data
that is already present, with history.
This sort of accident is why it's often a good idea to restrict
"users" to working in branches, and having only an administrator able
to commit things to trunks. It's also why it's so useful to have
occasional, locked down tags, to provide good points to revert big
accidental changes from.
Received on 2010-11-15 00:17:09 CET