You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Julian Foad <ju...@btopenworld.com> on 2015/02/13 19:07:05 UTC
1.7.x backport r1643074 - load with invalid mergeinfo - resolve the veto
Ben, you currently have a veto recorded against 1.7.x backport r1643074 - load with invalid mergeinfo:
* r1643074
Don't let invalid mergeinfo stop 'svnadmin load' from working.
Part of issue #4476 "Mergeinfo containing r0 makes svnsync and svnadmin
dump fail".
Justification:
The presence of 'bad' metadata in a repository should not prevent
backing up and restoring from a backup, given that this data does not
make the repository totally unusable.
Notes:
The 'svnadmin dump' part of this issue is already approved for backport.
The 'svnsync' part of this issue is not yet fixed.
Branch:
^/subversion/branches/1.7.x-r1643074
Votes:
+1: julianfoad
-1: breser (no changes on branch, attempt to merge something for this
was mismerged on release branch in r1643189, the merge that
was made broke swig-rb)
I resolved the problems (and have just synchronized the branch); could you check and at least remove your veto please if not also review it?
If anyone else can review it too, that would be great.
The same fix is proposed on 1.8.x, not vetoed, approved by me and Philip, and just needs one more vote. I have just sync'd that one's branch too.
- Julian
Re: 1.7.x backport r1643074 - load with invalid mergeinfo - resolve the veto
Posted by Julian Foad <ju...@btopenworld.com>.
Branko Čibej wrote:
> On 13.02.2015 19:57, Ben Reser wrote:
>> On 2/13/15 10:07 AM, Julian Foad wrote:
>>> Ben, you currently have a veto recorded against 1.7.x backport r1643074
>>> - load with invalid mergeinfo:
>>
>> I removed my veto but it doesn't merge:
>>
>> $ svn merge ^/subversion/branches/1.7.x-r1643074 .
>> svn: E195016: Reintegrate can only be used if revisions 1643188 through 1659644
>> were previously merged from
>> https://svn.apache.org/repos/asf/subversion/branches/1.7.x to the reintegrate
>> source, but this is not the case:
>>
>> subversion/branches/1.7.x-r1643074/subversion/tests/libsvn_repos/repos-test.c
>> Missing ranges:
>>
>> /subversion/branches/1.7.x/subversion/tests/libsvn_repos/repos-test.c:1643189,1643986
>
> The svn-backport-conflicts-1.7.x builder is red now; possibly related?
The merge problem probably originates from me having a sparse WC when I did the merges on the branch. But after a few quick tries just now I wasn't able to untangle it: simply making the WC complete (svn up --set-depth infinity) and doing another catch-up merge didn't behave as I expected and instead created conflicts in files such as repos-test.c that are already caught up. I'm not sure what's going on.
For now, I've veto'd the nomination again because of this.
I may have to recreate the branch from scratch. That will probably have to wait till Monday.
- Julian
Re: 1.7.x backport r1643074 - load with invalid mergeinfo - resolve
the veto
Posted by Branko Čibej <br...@wandisco.com>.
On 13.02.2015 19:57, Ben Reser wrote:
> On 2/13/15 10:07 AM, Julian Foad wrote:
>> Ben, you currently have a veto recorded against 1.7.x backport r1643074 - load with invalid mergeinfo:
>>
>>
>> * r1643074
>> Don't let invalid mergeinfo stop 'svnadmin load' from working.
>> Part of issue #4476 "Mergeinfo containing r0 makes svnsync and svnadmin
>> dump fail".
>> Justification:
>> The presence of 'bad' metadata in a repository should not prevent
>> backing up and restoring from a backup, given that this data does not
>> make the repository totally unusable.
>> Notes:
>> The 'svnadmin dump' part of this issue is already approved for backport.
>> The 'svnsync' part of this issue is not yet fixed.
>> Branch:
>> ^/subversion/branches/1.7.x-r1643074
>> Votes:
>> +1: julianfoad
>> -1: breser (no changes on branch, attempt to merge something for this
>> was mismerged on release branch in r1643189, the merge that
>> was made broke swig-rb)
>>
> I removed my veto but it doesn't merge:
>
> $ svn merge ^/subversion/branches/1.7.x-r1643074 .
> svn: E195016: Reintegrate can only be used if revisions 1643188 through 1659644
> were previously merged from
> https://svn.apache.org/repos/asf/subversion/branches/1.7.x to the reintegrate
> source, but this is not the case:
> subversion/branches/1.7.x-r1643074/subversion/tests/libsvn_repos/repos-test.c
> Missing ranges:
> /subversion/branches/1.7.x/subversion/tests/libsvn_repos/repos-test.c:1643189,1643986
The svn-backport-conflicts-1.7.x builder is red now; possibly related?
-- Brane
Re: 1.7.x backport r1643074 - load with invalid mergeinfo - resolve
the veto
Posted by Ben Reser <be...@reser.org>.
On 2/13/15 10:07 AM, Julian Foad wrote:
> Ben, you currently have a veto recorded against 1.7.x backport r1643074 - load with invalid mergeinfo:
>
>
> * r1643074
> Don't let invalid mergeinfo stop 'svnadmin load' from working.
> Part of issue #4476 "Mergeinfo containing r0 makes svnsync and svnadmin
> dump fail".
> Justification:
> The presence of 'bad' metadata in a repository should not prevent
> backing up and restoring from a backup, given that this data does not
> make the repository totally unusable.
> Notes:
> The 'svnadmin dump' part of this issue is already approved for backport.
> The 'svnsync' part of this issue is not yet fixed.
> Branch:
> ^/subversion/branches/1.7.x-r1643074
> Votes:
> +1: julianfoad
> -1: breser (no changes on branch, attempt to merge something for this
> was mismerged on release branch in r1643189, the merge that
> was made broke swig-rb)
>
I removed my veto but it doesn't merge:
$ svn merge ^/subversion/branches/1.7.x-r1643074 .
svn: E195016: Reintegrate can only be used if revisions 1643188 through 1659644
were previously merged from
https://svn.apache.org/repos/asf/subversion/branches/1.7.x to the reintegrate
source, but this is not the case:
subversion/branches/1.7.x-r1643074/subversion/tests/libsvn_repos/repos-test.c
Missing ranges:
/subversion/branches/1.7.x/subversion/tests/libsvn_repos/repos-test.c:1643189,1643986