From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!not-for-mail From: Tim Cross Newsgroups: gmane.emacs.devel Subject: Re: bzr not working? Date: Thu, 20 Oct 2011 08:33:41 +1100 Message-ID: References: <83y5wtxzwv.fsf@gnu.org> <83ty7ewqeu.fsf@gnu.org> NNTP-Posting-Host: lo.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Trace: dough.gmane.org 1319060037 23686 80.91.229.12 (19 Oct 2011 21:33:57 GMT) X-Complaints-To: usenet@dough.gmane.org NNTP-Posting-Date: Wed, 19 Oct 2011 21:33:57 +0000 (UTC) Cc: emacs-devel@gnu.org To: Eli Zaretskii Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Wed Oct 19 23:33:53 2011 Return-path: Envelope-to: ged-emacs-devel@m.gmane.org Original-Received: from lists.gnu.org ([140.186.70.17]) by lo.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1RGdlp-0003kn-7I for ged-emacs-devel@m.gmane.org; Wed, 19 Oct 2011 23:33:53 +0200 Original-Received: from localhost ([::1]:51177 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RGdlo-0005ZU-7Z for ged-emacs-devel@m.gmane.org; Wed, 19 Oct 2011 17:33:52 -0400 Original-Received: from eggs.gnu.org ([140.186.70.92]:42620) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RGdlk-0005ZP-JL for emacs-devel@gnu.org; Wed, 19 Oct 2011 17:33:50 -0400 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RGdlj-0005k3-D1 for emacs-devel@gnu.org; Wed, 19 Oct 2011 17:33:48 -0400 Original-Received: from mail-iy0-f169.google.com ([209.85.210.169]:51438) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RGdlg-0005jZ-7G; Wed, 19 Oct 2011 17:33:44 -0400 Original-Received: by iagf6 with SMTP id f6so3118296iag.0 for ; Wed, 19 Oct 2011 14:33:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=NfazvII+Ht8f8OzASzU5CBrNBBtR3PoL8NNxd2nn52M=; b=Mz/Q1M2OBNSe9Fplw3802CQKGTeNJ7onqv7vO2f0KONJMsJhyqo5FJ+htRW4qVX9Ak K3PzTvSEyBFIG7Tc6+GceNmWdFT0+5JsgLFwv69QrrqWDvIGs33zJRHYOz2GNGV33vpU fd4Nq7Mos6jyWokPksP6ow4zqx0vx5Fl61Jg8= Original-Received: by 10.231.50.202 with SMTP id a10mr3754514ibg.39.1319060021783; Wed, 19 Oct 2011 14:33:41 -0700 (PDT) Original-Received: by 10.231.12.67 with HTTP; Wed, 19 Oct 2011 14:33:41 -0700 (PDT) In-Reply-To: X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.6 (newer, 2) X-Received-From: 209.85.210.169 X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Original-Sender: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Xref: news.gmane.org gmane.emacs.devel:145354 Archived-At: On Thu, Oct 13, 2011 at 10:13 AM, Tim Cross wrote: > On Wed, Oct 12, 2011 at 7:43 PM, Eli Zaretskii wrote: >>> Date: Wed, 12 Oct 2011 10:35:13 +1100 >>> From: Tim Cross >>> Cc: emacs-devel@gnu.org >>> >>> Has anyone heard anything more regarding a resolution for this issue? >>> I've heard nothing from the savannah admins and stil cannot access the >>> bzr repository. >> >> You didn't subscribe to savannah-hackers-public, and evidently don't >> read the archives, so you are unaware of the continuing discussion on >> this matter. =A0Please look in the archives, if you are interested. >> >> In a nutshell, after a day and a half of testing on 3 different >> machine, my conclusion is that the problem I succeeded to reproduce is >> a different problem. =A0It evidently causes any bzr connection that >> takes longer than 1 hour to be forcibly dropped after precisely 1 >> hour. =A0(Due to the long time, I can only reproduce this with an >> initial "bzr branch" command, and only on machines with slow >> connection.) =A0There's probably some timeout somewhere on savannah, and >> savannah admins will be looking for it. >> >> But that is not your problem, because you reported a much faster >> disconnect. =A0So I suggest that you run "bzr -Dhpss pull" on the >> machine where this happens, and post the last portion of the .bzr.log >> file as a followup to your original message. =A0That should give the >> savannah admins information about what happens in the transport that >> causes the problem. >> >> TIA >> >> > > I have sent another message to the savannah list with the output from > -Dhpss. I also tried accessing another bzr repository on savannah > (color-theme) and got the same error. However, I can access bzr > repositories on other hosts, such as launchpad.net with no problems, > so it does seem to be savannah specific. I can also use the git > mirrors on savannah with no problems, which is what I'm doing now to > update my emacs code. > > While I would prefer to use bzr as it is the 'official' repository, > using git is actually more convenient, seems faster and has better > emacs integration. Now that the git mirror is being updated regularly, > I'm happy to use it for the time being, so this issue is less urgent > for me than it was. > > Tim > This issue has now been resolved. A quick update for the record and in case anyone else encounters a similar issue. The root cause of the problem was the deep packet inspection used by the firewall used by my employer. The firewall was configured to drop connections which were using 'unknown' protocols. At some point, either the bzr protocols changed or the vendor changed the signatures used to identify protocols (it looks like bzr may have changed, but not certain about this). When this change occurred, the bzr protocol was no longer recognized and the firewall dropped the connection. The confusing/misleading aspect of this is that the nature of some firewalls which use deep packet inspection is that they initially allow some traffic through in order to perform the analysis and then drop the connection if unrecognized. This gives the impression you have gotten 'out' of your network and are connected tot he remote server and that it is the remote server which drops the connection. I have assisted our network admins in gathering sample bzr protocl data, which they will submit to the firewall vendor so that they can update their signatures for identifying bzr protocls. This should then allow explicit configuration of what to do with bzr connections i.e. allow/block depending on policy etc. There is also an unrelated issue of savannah's bzr dropping connections after 1 hour. This was done by savannah sys admins to deal with runaway bzr processes - they configured a 'hard' 60min timeout for bzr connections. I believe this has been increased to 120min and is being reviewed. Apparently, newer versions of bzr have both built-in connection timeouts and improved detection for idle connections. Therefore, it is likely this hard coded limit will be removed once savannah has updated bzr to a sufficiently recent version. I have no details of time lines or even firm confirmation if/when these updates/changes will be performed and only mention it so that anyone on a slow connection is aware that bzr://bzr.savannah.gnu.org will timeout after 120min. I believe this only affects users of the native bzr protocol and not those accessing bzr repositories over http. Contact savannah if you have concerns/issues. Tim --=20 Tim Cross