Pushing to launchpad tells you it is stacking when it isn't

[John Arbash Meinel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Barry Warsaw wrote:
> On Nov 12, 2008, at 8:30 AM, Aaron Bentley wrote:
> 
>> Barry Warsaw wrote:
>>> On Nov 12, 2008, at 12:21 AM, John Arbash Meinel wrote:
>>>
>>>> I just pushed up a new branch of bzr-gtk so that I could start hacking
>>>> on a feature. As part of the creation, it led me to believe that it
>>>> would stack, though the formats did not support it.
>>>
>>> Which reminds me of a question I had yesterday.  If you push a stacked
>>> branch, but then push --overwrite, does it remain stacked?
> 
>> Yes.
> 
> Cool.
> 
>>> I ask
>>> because I did this yesterday and the second push took for evah!
> 
>> It could be one of the push bugs we're working on.  What branch was this?
> 
> ~barry/launchpad/bug-293128
> 
> Note: I kill that push, renamed the local directory and pushed it to
> ~barry/launchpad/bug-293128x.  That one went blindingly fast (and is the
> one that pqm landed last night).  The original is still there because of
> the merge proposal.
> 
> Cheers,
> -Barry
> 

Just to mention there was a bug when pushing to a stacked branch that it
failed to set the "cache size", which caused the push to write every
little request as a round trip. Andrew was debugging that one (the
normal behavior is to build up 1MB before sending it).

I'm pretty sure he has a fix for it, I don't know if he's sent it into
PQM yet.

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkkbVpcACgkQJdeBCYSNAANjgQCfUtPIjTpdwnni3UIheVSlput9
wbQAnRG3fJFp4qQv1YXu4sv4lGfVSNaa
=/PK9
-----END PGP SIGNATURE-----