[Bug 86392] Re: empty push performance slower than needed

2009-03-27 Thread Andrew Bennetts
Grr, Launchpad just ate my comment because I tried to change two statuses at once. In short: this is < 10 RPCs since sometime last year (and there's a test ensuring it stays that way in the test suite), so I think this is fixed. At 50ms latency, 10 RPCs should give less than a second of network t

[Bug 86392] Re: empty push performance slower than needed

2007-12-13 Thread John A Meinel
I wonder if a lot of this has now been addressed with pack and Branch6 formats Care to do a timing test? I'm guessing this is actually significantly better and this bug can be closed. ** Changed in: bzr Status: Confirmed => Incomplete -- empty push performance slower than needed https://

[Bug 86392] Re: empty push performance slower than needed

2007-05-01 Thread John A Meinel
see also bug 75721 -- empty push performance slower than needed https://bugs.launchpad.net/bugs/86392 You received this bug notification because you are a member of Ubuntu Bugs, which is the bug contact for Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.co

[Bug 86392] Re: empty push performance slower than needed

2007-02-20 Thread Robert Collins
% /usr/bin/time ssh HOST echo true true 0.00user 0.00system 0:00.17elapsed So the baseline to make new connections when there is a master channel is subsecond. ** Also affects: bzr (upstream) Importance: Undecided Status: Unconfirmed ** Changed in: bzr (upstream) Importance: Undecid