<br><div class="gmail_quote">On Wed, Jun 24, 2009 at 4:31 AM, Vincent Ladeuil <span dir="ltr"><<a href="mailto:v.ladeuil%2Blp@free.fr">v.ladeuil+lp@free.fr</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div class="im">>>>>> "Maritza" == Maritza Mendez <<a href="mailto:martitzam@gmail.com">martitzam@gmail.com</a>> writes:<br> <br> </div> Maritza> Hi Vincent,<br> Maritza> Thanks again for your interest in this topic. Here's what I have done:<br> <br> Thanks for your work :)<br> </blockquote><div><br>My pleasure. <br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Maritza> 1. boot the jaunty box<br> Maritza> 2. attempt to run 'bzr selftest test_http.TestActivity.test_get -v' <==<br> Maritza> that's the first test that fails, every time<br> <br> Hmm, ok, I get the failure again and I know why (see below).<br> <br> I also know why I failed to reproduce it this week-end during my<br> other tests: even when using the installed bzr, I was still using<br> the local bzrlib from my sources.<br> <br> Maritza> 3. instead of running selftest, bzr produces a dbus<br> Maritza> exception -- terminal outout included in archive as<br> Maritza> file 'dbus'<br> <br> bzr-gtk related, reported upstream (dbus), fixed there, will find<br> its way back.<br> </blockquote><div><br>Ah. Good. Thanks. <br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Maritza> 4. attempt to run same command again ==> runs<br> Maritza> without excpetion, but stalls on<br> Maritza> test_http.TestActivity.test_get(urllib,https,HTTP/1.0)<br> <br> Maritza> The bit about bzr selftest throwing a dbus exception<br> Maritza> on the first try after login seems to be fairly<br> Maritza> repeatable. I see a lot of speculation about this<br> Maritza> in bug reports, but I have no idea if it is<br> Maritza> related...<br> <br> Not related to our problem.<br> </blockquote><div><br>Ok. <br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Maritza> 5. strace bzr selftest test_http.TestActivity.test_get -v with bzr-gtk(and a<br> Maritza> few other typical plugins) installed ==> strace_gtk_installed=yes.txt<br> Maritza> 6. test hangs in the usual place. ctrl-C to kill strace.<br> Maritza> 7. remove *all* plugins (including gtk) from my plugins directory, so that<br> Maritza> only system plugins are present (bzrtools, lanuchpag and<br> Maritza> netrc_credential_store)<br> Maritza> 8. strace again, same options ==> strace_gtk_installed=no.txt<br> Maritza> 9. test hangs in the usual place. ctrl-C to kill stace.<br> <br> Ok. Sorry for my insistence there but an *ssh* password from<br> tests related to *http* was non-sense.<br> </blockquote><div><br>No problem. I agree that it is non-sense. I saw it only once. <br>Normally I do not report non-reproducible things. <br>But it was such bad non-sense that it had to be mentioned. :) <br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> So I know realize that the ssh related one is certainly another<br> bug, I focus on the http hangs here.<br> <br> <snip/><br> <br> Maritza> which suggests to me that it is the read() itself that is stalling.<br> <br> Maritza> Thoughts?<br> <br>