Re: [Savannah-hackers-public] GHOST security vulnerability CVE-2015-0235
Thanks Bob. On Wed, Jan 28, 2015 at 3:25 PM, Bob Proulx b...@proulx.com wrote: For the record all of the VMs were patched with the officially released patches for the GHOST security issue. All were rebooted earlier today. All are operating normally. Bob
Re: [Savannah-hackers-public] tar.git/plain/NEWS?id=release_1_27 does not work
Hi Paul, Both links appear to be working for me. cgit uses caching which sometimes causes such problems temporarily. On Tue, Oct 29, 2013 at 2:07 PM, Paul Eggert egg...@cs.ucla.edu wrote: This web page: http://git.savannah.gnu.org/cgit/tar.git/tree/NEWS?id=release_1_27 has a link to this: http://git.savannah.gnu.org/cgit/tar.git/plain/NEWS?id=release_1_27 The latter web page should exist, but doesn't. Is there something busted with git.savannah.gnu.org, or with GNU Tar, or what?
Re: [Savannah-hackers-public] Old Wiki Is Offline
I spent a good hour on this last night; no success. On Wed, Aug 21, 2013 at 7:38 PM, Karl Berry k...@freefriends.org wrote: FYI, I poked around some with zope without success. Will pick it up again tomorrow. k
Re: [Savannah-hackers-public] Leaving Savannah Administration
Thanks for your time! On Wed, Apr 17, 2013 at 12:59 PM, Felipe Lopez felipe.lo...@gmail.comwrote: Hi Savannah hackers, I'm writing just to let you know that I'm leaving Savannah Administration. I completed all the tasks assigned to me to the best of my ability. I'm more into public domain works now and want to avoid the paperwork and restrictions brought by copyright and licenses. Best, -- Felipe Lopez http://sirgazil.info/
Re: [Savannah-hackers-public] Savannah VM system upgrades
On Sun, Mar 10, 2013 at 06:19:49PM -0600, Bob Proulx wrote: Karl Berry wrote in the other message: I think mgt might be the closest to savannah-hackers and the furthest from the public and therefore I propose that we upgrade it first. Also sounds good to me. Sounds good. Will upgrade mgt first and then decide what to do next. When would be a good time to perform this upgrade? I don't think it really matters, assuming downtime is basically a matter of a reboot. (Especially for mgt.) A VM should boot very quickly. As a risk management I will coordinate with sysadmin just in case something goes really bad and it needs a rescue. I don't expect that to be needed. But just in case. I would suggest posting a news item so users have a chance of knowing what is going on. https://savannah.gnu.org/news/?group=administration Okay. I see that there hasn't been a news posting there since May 2012. How long of a waiting time for major events such as this should we have between posting a proposal for action and then performing the action? Once Michael confirms, you're good :). Otherwise ... a few days at least? Your plan sounds good to me. I'll be on IRC if you have any problems.
Re: [Savannah-hackers-public] project rename incomplete
On 12/31/12 5:59 PM, Karl Berry wrote: Hi Michael - I renamed the new gnurc project to remotecontrol (after talking with Stephen): mgt# /opt/administration/maintenance/rename_project.sh gnurc remotecontrol It seems to have mostly done the job, but not completely ... on the project page, https://savannah.gnu.org/projects/remotecontrol, the links in the Development Tools section (Source Code Manager, Bug Tracker, etc.) still refer to gnurc rather than remotecontrol. What needs to happen to fix this? Thanks, karl Hi, Sorry for the belated response, I've been occupied with end-of-year activities. The rename script does not change these settings (yet), so I normally set them through the website manually after every project rename. Later tonight I'll finish up that rename so the paths are correct.
Re: [Savannah-hackers-public] cannot do git checkout of gawk from savannah
On 12/24/12 8:59 AM, arn...@skeeve.com wrote: Hi All. Ward Vandewege w...@fsf.org wrote: On Mon, Dec 24, 2012 at 01:46:07AM -0700, arn...@skeeve.com wrote: Ineiev ine...@gnu.org wrote: Hi, On 12/24/2012 12:28 PM, arn...@skeeve.com wrote: $ git clone git://git.savannah.gnu.org/gawk.git Works for me. Still fails for me. Perhaps there are some connectivity problems. I'm getting this from a machine in Boston and also from Israel. I'm seeing it too. Looks like cgit.cgi is segfaulting again. Thanks, Ward. Um, so any ETA as to when it might start working again? Thanks, Arnold I'm looking into it.
Re: [Savannah-hackers-public] Savannah maintenance wiki spammed heavily
On 8/5/12 6:33 AM, Thomas Schwinge wrote: Hi! The Savannah maintenance wiki is being spammed heavily, see http://savannah.gnu.org/maintenance/FrontPage/recentchanges. It also seems that existing pages' content has been reverted. Grüße, Thomas FYI, I reverted most of the wiki and disabled anonymous editing of pages, so unfortunately you'll need an account if you want to edit the wiki now. If you want one and currently don't have one, please let me know.
Re: [Savannah-hackers-public] Savannah maintenance wiki spammed heavily
On 8/5/12 6:33 AM, Thomas Schwinge wrote: Hi! The Savannah maintenance wiki is being spammed heavily, see http://savannah.gnu.org/maintenance/FrontPage/recentchanges. It also seems that existing pages' content has been reverted. Grüße, Thomas I think I'm going to disable anonymous comments on the wiki, to avoid the spam issue.
Re: [Savannah-hackers-public] ntpd?
On 07/09/2012 02:39 PM, Karl Berry wrote: Michael or anyone, why do we run ntpdate hourly instead of the usual ntp daemon? (dl:/etc/cron.d/ntpdate) Seems bizarre, to say the least. Just wondering. k This was manually setup some time back (not by me), since the DomUs time was drifting. ntp needs to update the time every hour to avoid being a second off, from time to time.
Re: [Savannah-hackers-public] task #12069: Submission of Sharper // Volunteer application
On 6/10/12 7:08 PM, Karl Berry wrote: Hi Marvin, Thanks much for volunteering! [2] https://savannah.gnu.org/task/?12069 I agree with your assessment :). I've made you an administrator of the administration project (username implementation, right?). If you log out and log back into savannah, you should be able to go ahead and reply to the submission and enable it. You seem to be quite capable of reading what documentation there is, but just email if any questions, of course. Thanks, Karl Hi, thanks for volunteering! :) If you have not already done so, please sign up for the applicable mailing lists mentioned here: http://savannah.gnu.org/maintenance/SavannahHackersCommunication If you have any further questions, just email the list. P.S. Michael and all: if anyone else can jump in when new volunteers come on board, that would be great.
Re: [Savannah-hackers-public] Copyright+License Headers in concatenated HTML files
On 6/11/12 12:55 PM, Tomasz Konojacki wrote: Hi! Well, in my opinion, these files should contain licensing headers to avoid confusion about copyright issues. Tomasz W dniu 2012-06-11 17:27, Marvin Cohrs pisze: Hi everygnu, I'm checking the project 'Machafuco' and wondered about the following issue: The program itself is correctly commented with GPL3+ headers, but it also carries some HTML files which seem to be concatenated at runtime. They don't contain such a header. The footer file includes a short line about CC-BY-NC, but all the other parts don't (most are longer than 10 lines). Is this sufficient? The runtime output will contain that line, but the source files don't. What shall I answer? Thank you for your advice! Yes, if the author is using multiple licenses it needs to be clearly stated. CC-BY-NC is OK for docs, such as html, though I'd recommend not using the NC component of it, since the GPL allows for commercial usage. It's worth noting that the CC-BY-* licenses are GPL incompatible if any code is using it (html is not compiled code, so that's ok)
Re: [Savannah-hackers-public] Copyright+License Headers in concatenated HTML files
On 6/11/12 5:04 PM, Karl Berry wrote: CC-BY-NC is OK for docs Michael, I don't believe that is correct, for the reason you point out: NC is no commercial, which is nonfree, hence must not be allowed on savannah under any circumstances. As a separate point, real documentation (a manual) is supposed to be released under the FDL (or compatible permissive licenses). A few html files probably don't qualify, though. It's worth noting that the CC-BY-* licenses are GPL incompatible Although I don't believe it's ever been officially determined, I personally believe CC-BY itself is compatible with the GPL, because it only requires attribution, that is, is a permissive license. You were probably knowingly excluding that with BY-* instead of BY* :), but just for putative clarity :). It's the SA (sharealike) versions which are copylefts and hence incompatible with the GPL (another copyleft). And, as discussed, the NC (and ND) versions are nonfree and shouldn't be used in savannah-hosted projects. Aside from all of the above, I agree with the point about it always being better to include a license statement in the source files, for clarity. karl Indeed, I stand corrected.
Re: [Savannah-hackers-public] Proposed review of xvidcap
Hi Aljosha, Thanks so much for helping out with project submissions. :) You now have permissions for accepting/rejecting projects on Savannah. You will want to subscribe to the applicable mailing lists mentioned here, on the wiki: http://savannah.gnu.org/maintenance/SavannahHackersCommunication When reviewing projects, ensure you're logged into the system as Super User, to do this click the Become Superuser link in the left-hand toolbar. When it's time for a project to be accepted or rejected, click the Group Administration link on the project submission, select the appropriate status, then click update. If a new project is being accepted, you'll also want to click the Send New Project Instruction Email and Trigger Project Creation (should be done only once) link. You may be interested in some emacs/vim scripts we have for expediting project submissions. They are mentioned here: http://savannah.gnu.org/maintenance/HowToBecomeASavannahHacker If you have any questions on reviews, or anything else for that matter, feel free to email savannah-hackers-public@gnu.org and someone will reply. Once again, thanks! Michael On 5/17/12 7:01 PM, Karl Berry wrote: [Back on list.] OK, I enhanced the review. Thanks. That looks very good. Would someone else mind posting this in the review of the submission? And giving Aljosha permissions for the future? I'm out of time for today. Thanks, Karl Date: Thu, 17 May 2012 12:24:12 +0200 From: Aljosha Papschpapsch...@googlemail.com To: Karl Berryk...@freefriends.org Subject: Re: [Savannah-hackers-public] Proposed review of xvidcap [...] Thanks for submitting your project to Savannah! I noticed several files which lack a license header and copyright notice. These files are non-trivial files which are longer than 10 lines. To fulfil the requirements please add these informations. On [http://www.gnu.org/licenses/gpl-howto.html] you find more information regarding this subject. You can also read about the requirements for projects on Savannah on [https://savannah.gnu.org/register/requirements.php.]. If you are not sure whether you missed anything else, I recommend [https://savannah.gnu.org/maintenance/HowToGetYourProjectApprovedQuickly]. The files missing the license header and copyright notice are: ChangeLog Makefile.am README TODO.tasks configure.in all the Makefile.am files ppm2mpeg.sh doc/omf.make doc/xmldocs.make doc/man/man.make xml files in doc/xvidcap src/gnome-xvidcap.glade src/xvidcap-intl.h src/dbus-server-object.h It is also important to include a copyright notice in image files. You can do this by either adding it directly to the image file or mention the copyright and license in the README. Please resubmit your tarball once you updated these files.
[Savannah-hackers-public] [task #11980] Update look and feel of Savannah to match GNU.org
URL: http://savannah.gnu.org/task/?11980 Summary: Update look and feel of Savannah to match GNU.org Project: Savannah Administration Submitted by: mjflick Submitted on: Tue 03 Apr 2012 11:46:52 PM EDT Should Start On: Tue 03 Apr 2012 12:00:00 AM EDT Should be Finished on: Sun 03 Jun 2012 12:00:00 AM EDT Category: System Priority: 5 - Normal Status: None Privacy: Public Percent Complete: 0% Assigned to: mjflick Open/Closed: Open Discussion Lock: Any Effort: 0.00 ___ Details: The look and feel of savannah needs updated to reflect the look and feel of gnu.org. ___ Reply to this item at: http://savannah.gnu.org/task/?11980 ___ Message sent via/by Savannah http://savannah.gnu.org/
Re: [Savannah-hackers-public] userlist search
On 03/11/2012 06:01 AM, James Anslow wrote: Hi Michael, Yes please do add me to the project. I'm happy to review and work on it. KR, James You've been added to the project. Happy hacking :)
Re: [Savannah-hackers-public] userlist search
On 3/9/12 7:27 PM, Karl Berry wrote: Hi Michael/anyone -- how hard would it be to actually make the username search field work on https://savannah.gnu.org/siteadmin/userlist.php? It is a drag to scroll through a bunch of useless pages of (for instance) j user names to get to (for instance) jsg :). Just wondering. Thanks, karl Added to to do list.
Re: [Savannah-hackers-public] userlist search
On 3/10/12 1:54 PM, James Anslow wrote: It's not a complicated bit of work. Is the site written in PHP? I'd happily volunteer to code this functionality if so. James Hi, The code for the Savane cleanup-project is located here: https://savannah.nongnu.org/p/savane-cleanup If you're interested I'll add you to the project. :)
Re: [Savannah-hackers-public] savannah access from fencepost
On 3/2/12 7:59 PM, Karl Berry wrote: Hi Michael, Evidently something is now blocking access to *.sv.gnu.org from fencepost, e.g., fencepost$ ssh mgt.sv.gnu.org -l root ssh: connect to host mgt.sv.gnu.org port 22: Connection timed out Intentional change or related to the move? Not sure if this is something under your control, thought I'd ask before going to the sysadmins ... thanks, karl Yep, related to server migration (fencepost changed IPs). It's fixed now.
Re: [Savannah-hackers-public] gawk git repo not letting me in...
On 2/27/12 3:58 PM, Aharon Robbins wrote: Hi. I'm suddenly getting this: $ git push Permission denied (publickey). fatal: The remote end hung up unexpectedly Here is what's in my .git/config (hasn't changed): [remote origin] url = arn...@git.sv.gnu.org:/srv/git/gawk.git fetch = +refs/heads/*:refs/remotes/origin/* Help? Thanks, Arnold Robbins (gawk maintainer) Investigating.
Re: [Savannah-hackers-public] gawk git repo not letting me in...
On 2/27/12 5:27 PM, Michael J. Flickinger wrote: On 2/27/12 3:58 PM, Aharon Robbins wrote: Hi. I'm suddenly getting this: $ git push Permission denied (publickey). fatal: The remote end hung up unexpectedly Here is what's in my .git/config (hasn't changed): [remote origin] url = arn...@git.sv.gnu.org:/srv/git/gawk.git fetch = +refs/heads/*:refs/remotes/origin/* Help? Thanks, Arnold Robbins (gawk maintainer) Investigating. Resolved for now. This was caused by dns not resolving on the host.
Re: [Savannah-hackers-public] volunteering
Hi, Thanks for your interesting in helping out. :) On 1/28/12 6:53 AM, Brandon Invergo wrote: Hello Savannah Hackers, I've been wanting to help out with the GNU project for a while now but I've been unsure of where to help. I assume that Savannah still needs help, despite the resurrection back in November. Of course, I understand that this would mostly entail reviewing project submissions, so, as instructed, I've included what my response to a submission (#11765 CDEsql) below. The review looks fine. For future reference, we have two scripts to assist with project review response crafting. http://git.savannah.gnu.org/cgit/administration.git/tree/approvals One for emacs and one for vim. However, what I would be more interested in doing, if you guys approve, would be to help clean out old, inactive things. In particular, the Help Wanted area has posts going back a decade. I think it's safe to assume that many of those requests may be considered closed and removed at this point. Another, possibly touchier subject, would be to flag old, inactive and incomplete projects for removal. For both cases, I've written a suggested protocol below. Let's focus on the help wanted area. There's already a support request, https://savannah.gnu.org/support/?106581, which we can use to discuss the issue (feel free to post your ideas there). I agree, having open help wanted items from 2001 just isn't helpful. This isn't so much about space, which I'm sure isn't a huge problem. It's that, for me, as a visitor to Savannah, when I saw the Help Wanted area full of old posts and many of the projects apparently abandoned, it made the site feel a bit stale or inactive. I think cleaning out some of the cobwebs wouldn't hurt... This would require a lot more care and attention. First of all, only projects which have been inactive for at least X years (3? 5?) would be considered. In all cases, the project would not be immediately deleted. Instead, it would be flagged for review by another admin, and then we would attempt to contact the project owner to verify that the project is abandoned. It's important not to be overly zealous with this since even inactive projects can be of value to people starting new projects. Some rules for deciding to flag a project for removal (again, only after it's been inactive for X years): - no source code has been uploaded - it's determined that project activity is occurring on a different hosting service - development status 2(?) That's an interesting idea. Some projects only use Savannah for webhosting though, so they have no uploaded code. I think it's better to focus on cleaning up the old help wanted posts first. What's your username on Savannah, is it brandon?
Re: [Savannah-hackers-public] /root/administration - /opt
On 1/15/12 7:16 PM, Karl Berry wrote: I just now have replaced /root/administration with a symlink to /opt/administration, as we discussed. I left a tarball of what was /root/administration in mgt:/root/deleted-projects/administration-root.tar.gz in case there was anything unique there. (I should have checked, but didn't.) Please, let's never create multiple independent copies of manually maintained files! Too many cases of that already ... Thanks, karl thanks, thought this was already done.
Re: [Savannah-hackers-public] /root/administration vs. /opt/administration
On 1/3/12 6:24 AM, Ryan Doyle wrote: On 03/01/12 11:06, Karl Berry wrote: On mgt, there is both /root/administration and /opt/administration, which appear to be essentially duplicates, but /opt has later dates. What's up? Can we remove one or the other? And as I recall, I updated a bunch of references to use /root/administration, so I'd like to keep that path working, even if it becomes a symlink. Thanks, k Hi Karl, I added the /opt/administration as some lower privileged accounts (namely nagios) needed access. If /root/administration is the standardized location, I can update the symlink in /etc/nagios3 and change any permissions needed in /root/administration. Regards, Ryan If lower privileged accounts need access, I think it's best to symlink to make /root/administration a symlink to /opt/administration.
Re: [Savannah-hackers-public] Projects for Android in GNU Savannah
On 12/24/11 6:46 PM, Tomasz Konojacki wrote: Provided Savannah's maintainers have the capacity, projects running on Replicant may be hosted on Savannah. Projects having dependencies on non-free software, such as proprietary software drivers or AndroidOS, are not permissible. Shouldn't there be for instead of or? I was referring to drivers or the non-free operating system.
Re: [Savannah-hackers-public] Projects for Android in GNU Savannah
On 12/22/11 10:59 PM, Richard Stallman wrote: Do we have enough personnel and machine capacity on Savannah to handle all the projects that want hosting and DO run on GNU? In practice, yes. In that case, we can also host packages that run on Replicant. But this one won't run on Replicant, because of the camera problem, so it can't be hosted here. In that case, I'll add it to Savannah's hosting policy. -- Michael J. Flickinger
Re: [Savannah-hackers-public] Projects for Android in GNU Savannah
On 12/20/11 6:34 PM, Karl Berry wrote: Should we accept software targeted for Android in GNU Savannah?. Currently there is https://savannah.gnu.org/task/?11659 I am not sure. On the one hand, fsf.org has a page about free software on Android: http://www.fsf.org/working-together/next-steps/free-software-for-android. That would imply it is ok for us to host such free software projects. On the other hand, rms's essay (http://www.gnu.org/philosophy/android-and-users-freedom.html) concludes Android remains effectively nonfree. Which implies we shouldn't do any hosting. I guess I will ask Brett and see if he can advise us. I have formerly approved similar projects Regardless of what happens in the future, I would feel very bad about kicking out previously-approved projects due to our mistake, especially in such a borderline case of this. I don't think we've ever done that and I don't think we should start. Thanks, karl I agree, I suppose if we have accepted projects (in borderline cases), which we'd no longer accept today, I don't have any problem in grandfathering them. -- Michael J. Flickinger
Re: [Savannah-hackers-public] Task can not be assigned to Project Manager
On 12/17/11 7:46 AM, Joerg Kohne wrote: Hi Michael! In project's task manager tracker, the project manager can't assign a task with his user name itself (for others for information). Can this feature be made available? none is not optimally. Hope you can help me (even if this is the wrong mailing list). Thanks. Sincerely, Joerg Hi, On the project's Set Permissions page (https://savannah.gnu.org/project/admin/userperms.php?group=www-de) only the name of 'technicians' show in the assignment list. So, you want the permission 'tech manager' set, so you can assign and be assigned tasks. I updated this permission for you, so it should work now. -- Michael J. Flickinger
[Savannah-hackers-public] authorized_keys retrieval
Regarding ssh public key authentication: Keys are now automatically pulled from the database, so there's no more authorized_keys cronjob. I'll update the wiki in a little with details on the new setup. -- Michael J. Flickinger
Re: [Savannah-hackers-public] sysadmin help
Fantastic, if you're interested in helping out first review the maintenance wiki here: http://savannah.gnu.org/maintenance/SavannahArchitecture http://savannah.gnu.org/maintenance/SavannahHackersCommunication * (you'll want to subscribe to the emailing lists) I'll be in touch with you, via email, to discuss getting you started. :) Best, Michael J. Flickinger On 12/13/2011 07:18 AM, Ryan Doyle wrote: Hi, I'd like to offer to help with sysadmin of Savannah. Relevant skills include: * GNU/Linux - (8y) * PHP (3y), Ruby/Rails (6m), Python (1y), Perl (3y), Bash (3y) * Puppet (1y) I'm a sysadmin by trade and according to gnu.org/help you are looking for technical sysadmin volunteers so I'd be happy to help in my free time. Cheers, Ryan
Re: [Savannah-hackers-public] excessively high load avg, again (daily?)
On 12/09/2011 07:49 PM, Karl Berry wrote: This high load may be due to the nightly rsync you see below: Looks more like a runaway python to me. PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 1927 www-data 39 19 1090m 740m 3596 S0 12.1 425:43.54 python 2691 nobody20 0 109m 58m 33m S0 1.0 0:02.24 git 10840 root 20 0 107m 48m 1556 D1 0.8 1:19.83 rsync (BTW, not to be teaching my betters, but I suggest top c to show command lines and not just names.) But anyway ... Michael, what is the rsync job doing? It is easy to make it consume less resources in exchange for more time, --bwlimit=100 (Kbytes/sec) or whatever number turns out to be good. It should probably also be running with nice -19. I nice'd rsync to 19, The python process, loggerhead, is already nice'd to 19 as well.
Re: [Savannah-hackers-public] excessively high load avg, again (daily?)
On 12/8/11 7:18 AM, Jim Meyering wrote: Since there were some git-daemon processes dating back to November and since we have a limit on those (at least I think that's what Michael said), I've just killed those November git-daemon processes. I just added a timeout for them, just like we do for bzr. Perhaps related, there are lots of these in dmesg. We get from 1 to ~6 per hour: [2572324.224898] cgit.cgi[29777]: segfault at 863d000 ip b768c810 sp bfe9da2c error 6 in libc-2.11.2.so[b7619000+14] I'll investigate this further later today... In the near future I'll upgrade the version of git we're using, along with upgrading cgit.cgi. -- Michael J. Flickinger
Re: [Savannah-hackers-public] excessively high load avg, again (daily?)
On 12/8/11 3:30 PM, Jim Meyering wrote: Michael J. Flickinger wrote: On 12/8/11 7:18 AM, Jim Meyering wrote: Since there were some git-daemon processes dating back to November and since we have a limit on those (at least I think that's what Michael said), I've just killed those November git-daemon processes. I just added a timeout for them, just like we do for bzr. Perhaps related, there are lots of these in dmesg. We get from 1 to ~6 per hour: [2572324.224898] cgit.cgi[29777]: segfault at 863d000 ip b768c810 sp bfe9da2c error 6 in libc-2.11.2.so[b7619000+14] I'll investigate this further later today... In the near future I'll upgrade the version of git we're using, along with upgrading cgit.cgi. Great. Thanks for pursuing it. Upgraded git: git version 1.7.2.5 Upgraded to latest version of cgit: 0.9.0.2 I'm not really sure I like the idea of C cgi programs. Hopefully the segfaults will stop. If they don't, I'll debug it when I have some time... -- Michael J. Flickinger
Re: [Savannah-hackers-public] mirmon web page access
On 12/2/11 8:09 PM, Michael J. Flickinger wrote: On 12/2/11 2:24 PM, Karl Berry wrote: The basemirror, previously gnu-mirmon.basemirror.de, is now hosted on Savannah's download instance Hi Michael -- is it possible to access dl:/var/www/ftpmirror/mirmon/index.html from the web? It would be useful so the webmasters can follow up on dead mirrors. I couldn't find any url that worked, since theDocumentRoot in /etc/apache2/sites-enabled/download is /var/www/download. (http://dl.sv.gnu.org/mirmon/ gets to the savannah(nongnu) mirmon page.) Thanks, karl I'll make it accessible via the web as soon as I can. Done. http://download.savannah.gnu.org/mirmon/ -- Michael J. Flickinger
Re: [Savannah-hackers-public] mirmon web page access
On 12/2/11 2:24 PM, Karl Berry wrote: The basemirror, previously gnu-mirmon.basemirror.de, is now hosted on Savannah's download instance Hi Michael -- is it possible to access dl:/var/www/ftpmirror/mirmon/index.html from the web? It would be useful so the webmasters can follow up on dead mirrors. I couldn't find any url that worked, since theDocumentRoot in /etc/apache2/sites-enabled/download is /var/www/download. (http://dl.sv.gnu.org/mirmon/ gets to the savannah(nongnu) mirmon page.) Thanks, karl I'll make it accessible via the web as soon as I can. -- Michael J. Flickinger
Re: [Savannah-hackers-public] FYI: git was not responding: fixed
On 12/2/11 4:54 AM, Jim Meyering wrote: About an hour ago, for more than 20 minutes, git.sv.gnu.org's git server was dead. A restart attempt failed: $ service xinetd restart Stopping internet superserver: xinetd. Starting internet superserver: xinetd failed! Logs gave no indication of why it failed. A little later I tried a simple start, and that worked. Go figure. $ service xinetd start Starting internet superserver: xinetd. This has happened before when a bunch of stale git processes piled up. Did you check how many git processes, via xinetd, were running? -- Michael J. Flickinger
[Savannah-hackers-public] frontend repository
In an effort to provide isolation for Savannah-specific changes to Savane, the frontend now pulls from the administration/savane.git repository. Web-viewable here: http://git.savannah.gnu.org/cgit/administration/savane.git General fixes for savane should still be applied to the savane-cleanup project.
Re: [Savannah-hackers-public] bzr access problems
James Cloos wrote: MP == Martin Pool m...@canonical.com writes: MP Was bzr upgraded, or the server...? Someone recently posted that sv had been updated to bzr 2.4. -JimC Bazaar is at (bzr) 2.3.1 Does bzr 2.4 fix any bugs which could have potentially caused this?
Re: [Savannah-hackers-public] Savannah bzr server errors out
Hmm, all on the machine's side looks good. Low load, no stale bzr serve processes, nothing blowing up... Is this still a problem for you right now? On 10/10/2011 05:27 AM, Eli Zaretskii wrote: From: Martin Poolm...@canonical.com Date: Mon, 10 Oct 2011 18:43:13 +1100 Cc: savannah-hackers-public@gnu.org Getting a log with -Dhpss will tell you more. Did that, results below. Let me know if you need more data from my .bzr.log than I post below. And this time, it happened after only 20 minutes, so I doubt if a 1-hour timeout is involved. Here's the last portion of .bzr.log, including 2 transactions that look OK (there are many more like that before): 1249.000 RemoteTCPTransport.readv 2 offsets = 2 coalesced = 1 requests (2) 1249.000 hpss call w/readv: 'readv','/emacs/.bzr/repository/packs/cbc955c848fe27435bbc01a59b4808b0.pack' 1249.00020 bytes in readv request 1249.157 result: ('readv',) 1251.579284719 body bytes read 1253.765 RemoteTCPTransport.readv 2 offsets = 2 coalesced = 1 requests (2) 1253.765 hpss call w/readv: 'readv','/emacs/.bzr/repository/packs/cbc955c848fe27435bbc01a59b4808b0.pack' 1253.76520 bytes in readv request 1253.922 result: ('readv',) 1258.121454111 body bytes read 1258.750 RemoteTCPTransport.readv 2 offsets = 2 coalesced = 1 requests (2) 1258.750 hpss call w/readv: 'readv','/emacs/.bzr/repository/packs/859edd62277c23ca4a6dc8b1cc508429.pack' 1258.75019 bytes in readv request 1290.518 decoder state: buf[:10]='', state_accept=_state_accept_expecting_protocol_version 1290.612 Transferred: 86781kB (67.3kB/s r:86740kB w:41kB) 1290.612 Traceback (most recent call last): File bzrlib\commands.pyo, line 946, in exception_to_return_code File bzrlib\commands.pyo, line 1150, in run_bzr File bzrlib\commands.pyo, line 699, in run_argv_aliases File bzrlib\commands.pyo, line 721, in run File bzrlib\cleanup.pyo, line 135, in run_simple File bzrlib\cleanup.pyo, line 165, in _do_with_cleanups File bzrlib\builtins.pyo, line 1307, in run File bzrlib\bzrdir.pyo, line 453, in sprout File bzrlib\cleanup.pyo, line 131, in run File bzrlib\cleanup.pyo, line 165, in _do_with_cleanups File bzrlib\bzrdir.pyo, line 494, in _sprout File bzrlib\repository.pyo, line 724, in fetch File bzrlib\decorators.pyo, line 217, in write_locked File bzrlib\vf_repository.pyo, line 2499, in fetch File bzrlib\fetch.pyo, line 75, in __init__ File bzrlib\fetch.pyo, line 102, in __fetch File bzrlib\fetch.pyo, line 130, in _fetch_everything_for_search File bzrlib\vf_repository.pyo, line 1962, in insert_stream File bzrlib\vf_repository.pyo, line 2026, in insert_stream_without_locking File bzrlib\groupcompress.pyo, line 1661, in insert_record_stream File bzrlib\groupcompress.pyo, line 1751, in _insert_record_stream File bzrlib\repofmt\groupcompress_repo.pyo, line 1248, in wrap_and_count File bzrlib\groupcompress.pyo, line 1474, in get_record_stream File bzrlib\groupcompress.pyo, line 1627, in _get_remaining_record_stream File bzrlib\groupcompress.pyo, line 1188, in yield_factories File bzrlib\groupcompress.pyo, line 1430, in _get_blocks File bzrlib\repofmt\pack_repo.pyo, line 2027, in get_raw_records File bzrlib\pack.pyo, line 271, in iter_records File bzrlib\pack.pyo, line 312, in _read_format File bzrlib\pack.pyo, line 239, in _read_line File bzrlib\pack.pyo, line 202, in readline File bzrlib\pack.pyo, line 187, in _next File bzrlib\transport\remote.pyo, line 367, in _readv File bzrlib\smart\client.pyo, line 175, in call_with_body_readv_array File bzrlib\smart\client.pyo, line 81, in _call_and_read_response File bzrlib\smart\message.pyo, line 299, in read_response_tuple File bzrlib\smart\message.pyo, line 264, in _wait_for_response_args File bzrlib\smart\message.pyo, line 286, in _read_more ConnectionReset: Connection closed: Unexpected end of message. Please check connectivity and permissions, and report a bug if problems persist. 1290.612 return code 3 [ 3288] 2011-10-10 10:49:04.815 INFO: HPSS calls: 249 (249 vfs) bzrlib.smart.medium.SmartTCPClientMedium object at 0x01593EF0
[Savannah-hackers-public] Fwd: BerliOS will be closed on 31.12.2011
BerliOS is closing... Original Message Subject: BerliOS will be closed on 31.12.2011 Date: Fri, 30 Sep 2011 16:53:24 +0200 (CEST) From: ad...@berlios.de To: ad...@berlios.de Dear BerliOS developers and users, BerliOS was founded 10 years ago as one of the first repositories in Europe. It was developed and maintained by Fraunhofer FOKUS. As an European, non-proprietary project BerliOS pursued the goal to support the various open-source players and provide a neutral mediator function. In 2011 over 4710 projects have been hosted on BerliOS, with 50,000 registered users and over 2.6 million file downloads each month. We are proud that with BerliOS we have brought the idea of an OSS repository to Europe. Meanwhile, the concept has prevailed and there are many good alternatives. Unfortunately, as a research institute Fraunhofer FOKUS has only few opportunities to operate a repository like BerliOS. Such a project will only work with a follow-up financing, or with sponsors or partners taking over the repository. In the field of OSS this is a difficult undertaking. In a recent survey the community indicated some support in funds and manpower which we would like to thank you for. Unfortunately, the result is not enough to put the project on a sustainable financial basis. In addition the search for sponsors or partners was unsuccessful. Open Source is understood by Fraunhofer FOKUS as a paradigm for future-oriented intelligent use of IT. It hurts us all the more that we are forced to discontinue the hosting for BerliOS by 31.12.2011. * As a developer, you should export your BerliOS project into another repository. Alternatives see http://en.wikipedia.org/wiki/Comparison_of_open_source_software_hosting_facilities * On our site you will find a guide on how to get your project data out of the portal and migrate it in a different platform, see http://developer.berlios.de/docman/display_doc.php?docid=2056group_id=2 Fraunhofer FOKUS has a strong commitment to open source and interoperability, and is involved in numerous successful OSS projects. The institute focuses on the development of quality standards for open source software and in particular on the technical, semantic and organizational interoperability between open source software components and between open source and closed source software. Example of our OSS activities including our management of the German Competence Center QualiPSo. We thank all who have used BerliOS over the years. Fraunhofer FOKUS www.fokus.fraunhofer.de Sehr geehrte BerliOS Entwickler und Anwender, BerliOS wurde vor 10 Jahren als eines der ersten Repositories in Europa gegründet. Es wurde von Fraunhofer FOKUS entwickelt und gepflegt. Als ein europäisches, nicht proprietäres Projekt verfolgt BerliOS das Ziel, die verschiedenen Open-Source-Akteure zu unterstützen und eine neutrale Vermittlerfunktion zu bieten. 2011 wurden 4710 Projekte auf BerliOS gehosted, mit 50.000 registrierten Nutzern und über 2,6 Millionen Dateien Downloads jeden Monat. Wir sind stolz, dass wir mit BerliOS die Idee eines OSS-Repository nach Europa gebracht haben. Mittlerweile hat sich das Konzept durchgesetzt und es gibt zahlreiche gute Alternativen. Leider hat ein Forschungsinstitut wie Fraunhofer FOKUS nur wenig Möglichkeiten, langfristig ein Repository wie BerliOS zu betreiben. Ein solches Projekt funktioniert nur, wenn es gelingt, eine Anschlussfinanzierung zu finden, bzw. Sponsoren oder Partner zu gewinnen, die das Repository übernehmen. Das ist im OSS-Bereich ein schwieriges Unterfangen. In einer kürzlich durchgeführten Umfrage haben wir zwar Unterstützung an Geldmitteln und Arbeitsleistung signalisiert bekommen, für die wir uns bedanken. Leider reicht das Ergebnis aber nicht aus, um das Projekt auf eine nachhaltige finanzielle Basis zu stellen. Auch die Suche nach Sponsoren oder Partnern war leider erfolglos. Open Source wird bei Fraunhofer FOKUS als Paradigma für zukunftsweisenden intelligenten IT-Einsatz verstanden. Es schmerzt uns deshalb um so mehr, dass wir gezwungen sind, den Betrieb von BerliOS zum 31.12.2011 einzustellen. * Als Entwickler sollten Sie Ihre BerliOS Projekte in ein anderes Repository exportieren. Alternativen siehe http://en.wikipedia.org/wiki/Comparison_of_open_source_software_hosting_facilities * Auf unserer Website finden Sie einen Leitfaden, wie Sie Ihre Projektdaten aus dem Portal exportieren und in einer anderen Plattform überführen können, siehe http://developer.berlios.de/docman/display_doc.php?docid=2055group_id=2 Fraunhofer FOKUS hat nach wie vor ein starkes Engagement für Open Source und Interoperabilität, engagiert sich erfolgreich in zahlreichen OSS-Projekten. Das Institut konzentriert sich auf die Entwicklung von Qualitätsstandards für Open Source Software und dabei insbesondere auf die technische, semantische und organisatorische Interoperabilität zwischen einzelnen Open Source Software
Re: [Savannah-hackers-public] (No subject header)
rhe...@realtime.net wrote: What is Savannah backers? Tell the Castro person he or she should change that name. Bad connotation. I'm going to make the supposition you meant hacker. In the mainstream media, the word hacker does have a pejorative connotation. However, among computer programmers, at large, it has a positive one. Searching for the definition of hacker on most search engines will quickly reveal the word isn't limited to unethical activities. http://catb.org/jargon/html/H/hacker.html 1. A person who enjoys exploring the details of programmable systems and how to stretch their capabilities, as opposed to most users, who prefer to learn only the minimum necessary. RFC1392, the Internet Users' Glossary, usefully amplifies this as: A person who delights in having an intimate understanding of the internal workings of a system, computers and computer networks in particular. 2. One who programs enthusiastically (even obsessively) or who enjoys programming rather than just theorizing about programming. Best regards.
Re: [Savannah-hackers-public] captcha insufficient
On 9/18/11 2:29 PM, Karl Berry wrote: Michael, all, Sadly, it seems the captcha is insufficent (not surprising, of course -- if human spammers signed up before, we can't stop them). User robomo: https://savannah.gnu.org/siteadmin/usergroup.php?user_id=85295 signed up yesterday and posted a spam on a freetype bug report: https://savannah.nongnu.org/bugs/?32245 (I spamified it.) I wonder about doing both the captcha and asking for year GNU was announced + today's month number. I'm a little confused, you want this on the registration page? If the captcha isn't stopping them, then I don't think also asking for the current month will either. :( Are you suggesting adding a captcha every time a new patch/bug/support item is submitted or updated? If so, I think that may get a little annoying to users. If the spammers are created by humans, but auto-spam then we'll need captchas on every form submission, which just seems cumbersome. What do you think the ideal solution is? I don't expect to completely stop spammers from signing up but I fear that it happening a day or two after the captcha means that soon there will be a flood. Wdyt? karl P.S. To get recent user ids, I've been running: echo select user_id from user where add_date1316185718; | mysql savane where the big integer is a time_t for, e.g., yesterday about this time.
Re: [Savannah-hackers-public] news ui removal changes
Karl Berry wrote: To do the news ui removal that I just posted about, what I did was edit in-place /usr/src/savane/frontend/php/forum/forum.php /usr/src/savane/frontend/php/include/news/forum.php on frontend. I didn't see any bits in the database which would support making it conditional, and presumably upstream savane doesn't want these changes as-is. (Sylvain asked me to revert an earlier commit I made.) You could of used an on/off variable for this, specified in savane.ini. I'll make a change for this, leaving it on by default. We somehow have to disentangle ourselves from upstream so we can commit changes without messing them up -- our situation is only going to get worse. There are already tons of uncommitted changes, and committed changes not live on our host. But I don't understand git anywhere near well enough to do that. Meanwhile, I'll try to figure out the sql for deleting the actual spam items. Unfortunately it seems some early follow-ups are real. karl
Re: [Savannah-hackers-public] curl.txt, mirrors-contacts.txt, infra
On 09/07/2011 11:05 AM, Karl Berry wrote: Hi Michael, In the colonialone days, there were some files under /root that we used and maintained. For example: /root/curl.txt /root/mirrors-contacts.txt /root/infra/ I can't find these now on mgt, internal, or anywhere else. Am I missing them, or did they not get transferred, or? We need them :). Thanks, Karl They weren't transferred. :( I guess you can ask FSF sysadmin to retrieve them. :(
Re: [Savannah-hackers-public] Migrating items from one tracker to another...?
On 09/03/2011 06:11 PM, Karl Berry wrote: What I'd really like to do is move the patches and support tracker contents into the bug tracker ... Does anyone have any idea how to do such a thing? Maybe with some magic Savannah admin script or something? I have never seen anything like that. Resolving conflicts and merging info from different trackers sounds rather intimidating. Paul, if you want to hack on it yourself you can surely have the data, of course. Or, is there a way to make trackers read-only to the public, All I know about is the on/off activation, which isn't what you want. Michael? Sorry, karl There's no magical way to import them. The good news is that support, bug, and patch items all share a nearly identical schema, so it would be possible to write a script to move tracker items. I'm adding this to my to do list, but there's no real way I'll get around to it soon.
[Savannah-hackers-public] New Volunteer
I'd like to welcome Ben Asselstine to the Savannah Hackers, as a project reviewer. I've added him to the administration group where he'll be helping with project applications. -- Michael
[Savannah-hackers-public] Bzr bug
For reference: I reported a bug regarding an issue with bzr's hpss today here: https://bugs.launchpad.net/bzr/+bug/824797 This bug has caused previous load problems on Savannah.
Re: [Savannah-hackers-public] Emacs-diffs
On 08/03/2011 11:58 AM, Eli Zaretskii wrote: Date: Tue, 02 Aug 2011 11:47:34 -0400 From: Michael J. Flickingermjfl...@gnu.org CC: savannah-hackers-public@gnu.org On 07/31/2011 11:14 AM, Eli Zaretskii wrote: Date: Sat, 30 Jul 2011 11:37:10 -0400 From: Michael J. Flickingermjfl...@gnu.org CC: savannah-hackers-public@gnu.org On 07/30/2011 02:55 AM, Eli Zaretskii wrote: There were no diffs sent from the commits to the Emacs repository since July 22. Are the bzr-related scripts that send the diffs upon commit still functional? Can someone please take a look at this? TIA Should be working now. Let me know if this still fails. I waited for more than 24 hours, but still didn't see even a single message, neither in my inbox (I subscribe to emacs-diffs) nor in the mailman archives. I think I fixed the problem. Should be working now. I'm sorry, but it doesn't. No messages were sent in the past 24 hours, and you can see in the archives that the last message is still from July 21. Mail configuration problem. Should be working now. Sorry for all the recent hangups, they are primarily due to infrastructure changes.
Re: [Savannah-hackers-public] Anonymous Git access troubles
On 08/02/2011 01:48 AM, Eli Zaretskii wrote: Date: Tue, 02 Aug 2011 05:51:31 +0300 From: Eli Zaretskiie...@gnu.org Cc: b...@gnu.org, savannah-hackers-public@gnu.org Reply-To: Eli Zaretskiie...@gnu.org Thanks. Apparently, the current limit is still too low, or maybe some other factor is at work here, because one of the users who had this problem reports: Doubled the limit (again). A few days back the anon bzr was causing an intense amount of load on the server. Bzr required a great deal more resources than the other version control systems when it runs. Works on Windows where the branch is only a couple of revisions behind. On GNU/Linux (Arch, bzr v2.3.4) where the trunk branch is at 104259 it starts pulling the revisions but after a while stops with the same error as before. Another data point: using nosmart+bzr:// protocol works, but using bzr:// fails: bzr: ERROR: Connection closed: Unexpected end of message. Please check connectivity and permissions, and report a bug if problems persist. I tried both on fencepost, which probably excludes any network issues.
Re: [Savannah-hackers-public] Emacs-diffs
On 07/31/2011 11:14 AM, Eli Zaretskii wrote: Date: Sat, 30 Jul 2011 11:37:10 -0400 From: Michael J. Flickingermjfl...@gnu.org CC: savannah-hackers-public@gnu.org On 07/30/2011 02:55 AM, Eli Zaretskii wrote: There were no diffs sent from the commits to the Emacs repository since July 22. Are the bzr-related scripts that send the diffs upon commit still functional? Can someone please take a look at this? TIA Should be working now. Let me know if this still fails. I waited for more than 24 hours, but still didn't see even a single message, neither in my inbox (I subscribe to emacs-diffs) nor in the mailman archives. I think I fixed the problem. Should be working now.
[Savannah-hackers-public] lists.gnu.org - allow ssh from 140.186.70.75
After we migrated the internal savannah domU to a real ip address, 140.186.70.75, the mailman sync functionality has broken. lists.gnu.org does not accept traffic from 140.186.70.75, which is the internal domU ip. Could you please allow ssh from 140.186.70.75 to lists?
Re: [Savannah-hackers-public] Source IP for VCS commit notifications
On 07/31/2011 05:01 PM, Sylvain Beucler wrote: Hi, At Gna! the -commits mailing lists (in particular, savane-comm...@gna.org) use an IP-based restriction to prevent SPAM: only known sender IP addresses are accepted (in particular, Gna! et Savannah). Today I received the message below which indicated a change in configuration: the source IP appears to be 140.186.70.72 (vcs.savannah.gnu.org) rather than the previous 140.186.70.51. Do you confirm? Yes. Incidentally, I'm curious on why you moved away from the mail smarthost on 'internal' :) It used to centralize all the mail aliasing. Yes, some email providers flagged mail sent from the smarthost as span, since the internal.in.sv.gnu.org domU had a 10.1.0.101 for an IP address. It was easiest to just setup mail on vcs. - Sylvain - Forwarded message from Mail Delivery Systemmailer-dae...@vcs.savannah.gnu.org - Date: Sun, 31 Jul 2011 20:29:35 + From: Mail Delivery Systemmailer-dae...@vcs.savannah.gnu.org To: b...@gnu.org Subject: Mail delivery failed: returning message to sender This message was created automatically by mail delivery software. A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed: savane-comm...@gna.org SMTP error from remote mail server after RCPT TO:savane-comm...@gna.org: host mail.gna.org [78.40.125.82]: 550 Administrative prohibition -- This is a copy of the message, including all the headers. -- Return-path:b...@gnu.org Received: from Beuc by vcs.savannah.gnu.org with local (Exim 4.72) (envelope-fromb...@gnu.org) id 1QncdM-0003Uh-S9 for savane-comm...@gna.org; Sun, 31 Jul 2011 20:29:13 + Date: Sun, 31 Jul 2011 20:29:12 + Message-Id:e1qncdm-0003uh...@vcs.savannah.gnu.org To: savane-comm...@gna.org Subject: [SCM] Savane framework branch, master, updated. cba498bc6372c0357f5006cd08ae342ba69fbe19 X-Git-Refname: refs/heads/master X-Git-Reftype: branch X-Git-Oldrev: 3fa889d285dbf29d5941739051a662a3a6d8073c X-Git-Newrev: cba498bc6372c0357f5006cd08ae342ba69fbe19 From: Sylvain Beuclerb...@gnu.org This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project Savane framework. The branch, master has been updated via cba498bc6372c0357f5006cd08ae342ba69fbe19 (commit) from 3fa889d285dbf29d5941739051a662a3a6d8073c (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log - http://git.savannah.gnu.org/cgit/savane.git/commit/?id=cba498bc6372c0357f5006cd08ae342ba69fbe19 commit cba498bc6372c0357f5006cd08ae342ba69fbe19 Author: Sylvain Beuclerb...@beuc.net Date: Sun Jul 31 22:27:15 2011 +0200 Django 1.3 new blocktrans syntax (1.3 has incomplete 1.2 syntax support) diff --git a/templates/svmain/homepage.html b/templates/svmain/homepage.html index 6263252..80e8385 100644 --- a/templates/svmain/homepage.html +++ b/templates/svmain/homepage.html @@ -13,10 +13,10 @@ {% blocktrans %}{{site_name}} statistics{% endblocktrans %}/a/div div class=smaller div class={% cycle 'boxitemalt' 'boxitem' as rowcolor %} -{% blocktrans count nb_users as count and 'strong'|add:nb_users|add:'/strong'|safe as html %}{{html}} registered user{% plural %}{{html}} registered users{% endblocktrans %} +{% blocktrans with html='strong'|add:nb_users|add:'/strong'|safe count count=nb_users %}{{html}} registered user{% plural %}{{html}} registered users{% endblocktrans %} /div div class={% cycle rowcolor %} -{% blocktrans count nb_groups as count and 'strong'|add:nb_groups|add:'/strong'|safe as html %}{{html}} hosted project{% plural %}{{html}} hosted projects{% endblocktrans %} +{% blocktrans with html='strong'|add:nb_groups|add:'/strong'|safe count count=nb_groups %}{{html}} hosted project{% plural %}{{html}} hosted projects{% endblocktrans %} /div {% for conf in group_confs %} div class={% cycle rowcolor %} --- Summary of changes: templates/svmain/homepage.html |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) hooks/post-receive
Re: [Savannah-hackers-public] Emacs-diffs
On 07/30/2011 02:55 AM, Eli Zaretskii wrote: There were no diffs sent from the commits to the Emacs repository since July 22. Are the bzr-related scripts that send the diffs upon commit still functional? Can someone please take a look at this? TIA Should be working now. Let me know if this still fails.
[Savannah-hackers-public] Creating additional git repositories
If you've ever wondered how to create additional git repositories, as this is a common support request, wonder no more! It's documented here now: http://savannah.gnu.org/maintenance/Git
[Savannah-hackers-public] management vm
Previously, we used to run maintenance scripts directly from colonialone (savannah's dom0). Since we'll no longer have access to the dom0, I'll setup the scripts on the builder vm, which will be renamed to mgt (for management). I'll let you all know once the scripts have been updated to work properly. Also, if we want to add any additional monitoring functionality, we can set it up on the mgt vm -- Michael
Re: [Savannah-hackers-public] help needed with savannah
On 06/12/2011 05:12 PM, Karl Berry wrote: Glad to see progress on the script ... why not just use the administration project https://savannah.gnu.org/projects/administration which already exists, instead of making a new one? You can add it here: http://bzr.savannah.gnu.org/lh/administration/changes There's also an empty git repository: https://savannah.gnu.org/git/?group=administration
Re: [Savannah-hackers-public] bzr.savannah.gnu.org unusable
Hi Eli, The 24 hour+ outage was caused by upgrades to the machine Savannah runs on. Since the system upgrade, the load on vcs-noshell has been low, so nothing else is weighing down the machine now. On 6/4/11 2:48 AM, Eli Zaretskii wrote: It gets stuck even in a simple bzr update (for the Emacs repository). This is after more than 24 hours of total outage, so I have quite a few commits that wait to be done. Could someone please take a look??
Re: [Savannah-hackers-public] Savannah slow again?
Eli Zaretskii wrote: Is something wrong with bzr.savannah.gnu.org? It takes forever to resync with the Emacs repository (4 minutes and counting), where normally it takes 20 seconds. viewvc is currently causing a high load. I'll post promptly with a followup.
Re: [Savannah-hackers-public] Savannah slow again?
Michael J. Flickinger wrote: Eli Zaretskii wrote: Is something wrong with bzr.savannah.gnu.org? It takes forever to resync with the Emacs repository (4 minutes and counting), where normally it takes 20 seconds. viewvc is currently causing a high load. I'll post promptly with a followup. I posted the cause of the crash to the savannah-hackers-private list, since it could become a minor security concern until it's fixed.
Re: [Savannah-hackers-public] Loggerhead down?
Eli Zaretskii wrote: http://bzr.savannah.gnu.org/lh/emacs says The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. What happened? Earlier this week loggerhead crashed the vcs domU. I'm working on re-enabling it right now with some constraints to ensure that doesn't happen again.
Re: [Savannah-hackers-public] Loggerhead down?
Michael J. Flickinger wrote: Eli Zaretskii wrote: http://bzr.savannah.gnu.org/lh/emacs says The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. What happened? Earlier this week loggerhead crashed the vcs domU. I'm working on re-enabling it right now with some constraints to ensure that doesn't happen again. Ok, it's back up. I limited the amount of concurrent connections and enhanced logging, in an effort to avoid that from happening again. Thanks
Re: [Savannah-hackers-public] Loggerhead down?
Eli Zaretskii wrote: Date: Fri, 27 May 2011 13:00:17 -0400 From: Michael J. Flickinger mjfl...@gnu.org CC: savannah-hackers-public@gnu.org Earlier this week loggerhead crashed the vcs domU. I'm working on re-enabling it right now with some constraints to ensure that doesn't happen again. Ok, it's back up. I limited the amount of concurrent connections and enhanced logging, in an effort to avoid that from happening again. Thanks, it works fine. Great, sorry for the inconvenience.
Re: [Savannah-hackers-public] vcs-noshell hangup
Michael J. Flickinger wrote: Michael J. Flickinger wrote: vcs-noshell hung due to cgit. attached is output from the vcs-noshell console. I'll follow-up with more information when I get a chance. I disabled cgit until I can dig into this further. Update: Doesn't look like cgit (I panicked when I saw all the cgit segfaults...) It looks more like loggerhead spawning too many processes...
Re: [Savannah-hackers-public] more than one git module per savannah project?
On 05/06/2011 12:47 AM, Simon Josefsson wrote: How does it work? How will the pages under Source Code look like? Currently, these repositories are viewable in this format: http://git.savannah.gnu.org/cgit/hurd I created a new repository for your project, which you can see here: http://git.savannah.gnu.org/cgit/libidn (this page should show both repositories) This should let you clone your repository: git clone usern...@git.savannah.gnu.org/libidn/libidn2.git At some point, hopefully before 2012, I'd like to create an interface to manage repositories, so that projects can do these things themselves. Can I get another git module called 'libidn2' connected to the 'libidn' project? Thanks Karl for asking. /Simon Michael J. Flickingermjfl...@gnu.org writes: At present, they have to be requested via support request. Aside from that, there's no web-based support for them (yet). On 5/5/11 6:29 PM, Karl Berry wrote: Michael, Jim, someone, Does savannah support more than one git module per project? (Question came up in discussion with Simon.) (Sorry I don't know the answer, but I don't, and don't have any idea where to even begin looking.) Thanks, karl
[Savannah-hackers-public] New Volunteer
I'd like to welcome Shaunak Saha to the Savannah Hackers. I've added him to the administration group where he'll be initially helping with project applications. :) Shaunak, be sure to join this list (savannah-hackers-public), as well as other applicable lists mentioned in the wiki. (http://savannah.gnu.org/maintenance/SavannahHackersCommunication) Further, If you have any questions regarding reviewing project applications, or anything else for that matter, feel free to write the savannah-hackers-public list. Thanks so much for your help, Shaunak. -- Michael
Re: [Savannah-hackers-public] more than one git module per savannah project?
At present, they have to be requested via support request. Aside from that, there's no web-based support for them (yet). On 5/5/11 6:29 PM, Karl Berry wrote: Michael, Jim, someone, Does savannah support more than one git module per project? (Question came up in discussion with Simon.) (Sorry I don't know the answer, but I don't, and don't have any idea where to even begin looking.) Thanks, karl
[Savannah-hackers-public] Savannah VPN
Bernie and I setup a VPN for Savannah, so that Savannah and other applicable servers may be accessed via the VPN, rather than directly or via fencepost proxying. If you would like access to the vpn, so that you may access colonialone and the mailserver trivially, please send me an email so that I may setup a key for you. A page with information on how to connect to the vpn is available here: * http://savannah.gnu.org/maintenance/VPN Cheers, Michael
[Savannah-hackers-public] Upgraded vcs-noshell to debian squeeze
Today I upgraded the vcs-noshell from debian lenny to debian squeeze. There's not too much to note about the upgrade, rather than some minor fallout thanks to a bug with libnss-mysql (someone else also had this problem): http://us.generation-nt.com/answer/bug-611019-libnss-pgsql2-does-not-correctly-handle-empty-string-query-result-help-201878892.html Everything else should be working as usual. ** https://savannah.gnu.org/forum/forum.php?forum_id=6772
Re: [Savannah-hackers-public] Re: savannah call for help
Hi Paul, It's also worth noting that there's an IRC channel on channel #savannah at irc.oftc.net, if you're interested. Some savannah hackers idle in there, if you want instantaneous help. Thanks, Michael On 03/31/2011 07:10 PM, Karl Berry wrote: Hi Paul, I'd be willing to help out. That is great! We would be most grateful to have you join us. Your background is surely plenty sufficient. (I clued in the other current sv folks.) Right now, there are no tools for reviewing project submissions. It's a matter of laboriously checking all the files for license headers, among the other more interesting criteria. http://savannah.gnu.org/maintenance/HowToGetYourProjectApprovedQuickly https://savannah.gnu.org/maintenance/ApprovingSubmission I added you to the administration project as an admin, and to the two important savannah-*-private mailing lists. https://savannah.gnu.org/projects/administration/ In addition to new project submissions, addressing support requests (wherever possible) is another useful and important thing to do. https://savannah.gnu.org/support/?group=administration Please dive in, and of course write any time with questions/problems/whatever ... Thanks again, Karl
Re: [Savannah-hackers-public] Joe Kierpaul
Hi Mario, Joe sent an email to the savannah-help-public address saying he wanted to help out. I replied to him, cc'ing the list, and we later met-up on IRC. In the past we haven't really sent out formal introductory emails when someone has started working on project applications. In fact, in the past, there hasn't always been an email sent out when Sylvain gave someone root on the dom0. This wouldn't be a bad idea in the future though, so next time someone starts an introductory email will be sent. I've been working with Joe to understand how to do project applications, at first. My apologies for any confusion. Mario Castelan Castro wrote: 2011-03-29 in savannah-hackers-public@gnu.org thread Joe Kierpaul. Hello. I have noticed this person has assigned himself and approved (?) some projects. I'm totally confussed because of the lack of transparency. Is he a new volunteer in GNU Savannah?. Have someone (Alex?, Michael?, Karl?) told him what to do?. Who gave him administration permissions inmediatly and why?. And more importantly: where is the report of the new member and his inclusion?. Hope everything is ok. Regards.
Re: [Savannah-hackers-public] spam on sv lists
On 03/29/2011 12:23 PM, Jim Meyering wrote: Karl Berry wrote: Michael, Jim, all, One other thing Sylvain was doing was manually deleting spam from the savannah-*-private lists. I am not excited about deleting spam manually when listhelper will do it for free. I don't see that the extra mail hops significantly increase the chance of exposure of any private information. Ok to install listhelper on those lists? I agree. Good idea. Likewise, agree.
Re: [Savannah-hackers-public] savannah-cvs and wiki changes?
Karl Berry wrote: Maybe we should subscribe the savannah-cvs list to wiki changes? Seems like they go together ... Michael? When changes are made in the maintenance wiki they already go to the savannah-cvs list.
Re: [Savannah-hackers-public] git repositories with no description, FYI: fixed
On 03/25/2011 06:12 PM, Karl Berry wrote: Does anyone know an easy way to determine which is which? savannah.gnu.org:/var/www/overlay/cooperation/groups.tsv is a .tsv dump of sv projects with their type (gnu, nongnu, etc.). If not, I'll just redo it without the GNU prefix. I think that would be better anyway. Less information, less chance for it to be wrong :). I agree with Karl. Just use the project name, it's the easiest and least buggy way.
Re: [Savannah-hackers-public] backup redirector for ctan
Hi Karl, I still haven't forgotten about this. Could you please tell me how you'd like this implemented, at a dns level? I could setup something like ctan-mirror.sv.gnu.org or you guys could setup a cname entry mapping to dl.savannah.gnu.org. Just ping me back with whatever you'd prefer. Cheers, Michael Michael J. Flickinger wrote: Hi Karl, I don't see any problem hosting another redirector on Savannah--especially since this one would actually come with a mirror.txt file. I'll dabble into this tonight or tomorrow night. -- Michael Karl Berry wrote: Michael and all, In another part of life I work on TeX stuff. Randy had (not coincidentally :) implemented the same sort of redirector for CTAN (http://mirror.ctan.org) as for GNU. The CTAN folks have reimplemented it (I'm cc-ing Rainer Schoepf of CTAN who was the principal person doing it, hi Rainer). Now they are wondering about having a standby available as a backup. I suggested that savannah would be a possibility, since of course all the necessary infrastructure is already there. If you/we (GNU hat) are willing, the two files are available from rsync://comedy.dante.de/MirMon -- mirmon.state and mirror.txt. Wdyt? If the answer is no for whatever reason, that's ok, we (TeX hat) can always figure out somewhere else. Just thought I'd ask here first. Thanks, Karl
Re: [Savannah-hackers-public] browsing webpages repo
Fixed by making a trivial code change in savane-cleanup. (I implemented the logic that was used in the project menubar, where the Browse Web Pages Repository link was correctly showing.) That said, the 'www' project should now display the link correctly. Thanks, Michael On 03/08/2011 06:09 PM, Michael J. Flickinger wrote: I'll investigate this later tonight. I'm assuming it has to do with something in the database not being properly set (or a bug in savane.) Either way, it should prove trivial to fix. -- Michael Karl Berry wrote: Michael or anyone, When I look at https://savannah.gnu.org/projects/gnulib the last line is a link to Browse Web Pages Repository - http://web.cvs.savannah.gnu.org/viewvc/?root=gnulib (and it works fine) But when I look at https://savannah.gnu.org/projects/www there is no such corresponding line about browsing the web repo, even though http://web.cvs.savannah.gnu.org/viewvc/?root=www does in fact work (yay). Why doesn't that link show up on the www project page? It would be helpful to webmasters, especially the new ones always coming along, if the link was visible instead of secret knowledge :). Thanks, Karl
Re: [Savannah-hackers-public] help for savannah
On 03/12/2011 10:40 AM, Karl Berry wrote: I'd love to get more volunteers helping with support requests, but it's not entirely trivial to get a new volunteer working on those issues Some support requests don't require root/shell access, just the web interface. New volunteers could start with those (plus submission reviewing). That's a good point. I'm not so sure focusing on a rewrite now, I fully agree with that. Anyway, how about something like this for an FSF/GNU blog post: News from Savannah: Sylvain Beucler, who was instrumental in modernizing and maintaining Savannah (http://savannah.gnu.org) for the last seven years, has decided to step down and look for new challenges. Many, many, thanks to Sylvain and best wishes. Michael Flickinger, a long-time Savannah administrator, has resumed work on the Savannah back end. Of course, new volunteers would be greatly appreciated. Some specific tasks: - Help with reviewing new project submissions. http://savannah.gnu.org/maintenance/HowToGetYourProjectApprovedQuickly - Work on bazaar/loggerhead integration at Savannah, e.g., right now we cannot offer web browsing of source repositories. http://savannah.gnu.org/maintenance/Bzr - Help with handling support requests. https://savannah.gnu.org/support/?group=administration For those who want to explore, the main Savannah documentation (architecture, procedures, faq's) is at http://savannah.gnu.org/maintenance. Wdyt? Sound good.
Re: [Savannah-hackers-public] help for savannah
Alex Fernandez wrote: Hi Michael, On Fri, Mar 11, 2011 at 4:38 PM, Michael J. Flickinger mjfl...@gnu.org wrote: Mario and Alex are doing a great job with project submissions (and I cannot thank them enough!) So, there's no real backlog of them. (Thanks Mario and Alex!) More reviewers are always a good thing and welcomed though. Well, thanks! I would like to tackle a very interesting project: automatic compliance checks via a Python or Lua script (Perl has also been suggested). I have not had the time though. Yes, having some sort of script that could inspect tarballs for both current releases and submitted projects would be a totally awesome idea. Alex Fernandez wrote: I'm hoping that perhaps Mario and Alex can start taking more of these.. :) Unfortunately I don't have the knowledge near the level necessary to do what you do... Alex. That's ok... I'm more than willing to help you, as much as I can. Aside from the maintenance wiki (which I'm trying to keep updated--(I see karl's been updating things too there), you can always send me an email or stop by the IRC chat and ask me how to do something.
Re: [Savannah-hackers-public] browsing webpages repo
I'll investigate this later tonight. I'm assuming it has to do with something in the database not being properly set (or a bug in savane.) Either way, it should prove trivial to fix. -- Michael Karl Berry wrote: Michael or anyone, When I look at https://savannah.gnu.org/projects/gnulib the last line is a link to Browse Web Pages Repository - http://web.cvs.savannah.gnu.org/viewvc/?root=gnulib (and it works fine) But when I look at https://savannah.gnu.org/projects/www there is no such corresponding line about browsing the web repo, even though http://web.cvs.savannah.gnu.org/viewvc/?root=www does in fact work (yay). Why doesn't that link show up on the www project page? It would be helpful to webmasters, especially the new ones always coming along, if the link was visible instead of secret knowledge :). Thanks, Karl
Re: [Savannah-hackers-public] Re: bzr + savannah.gnu.org slow again
On 03/06/2011 07:42 PM, Alex Fernandez wrote: Hi all, On Mon, Feb 28, 2011 at 2:02 PM, Jim Meyeringj...@meyering.net wrote: Glenn Morris wrote: At 23.30 PST on Sunday 27 Feb, it is very slow again. ~ 35 mins for a ~ 5 line commit. It was alright mid-afternoon PST. Can you not just disable the cgit service (it seems to still be available) until there is a solution for this problem? Being able to commit is more important. Hi Glenn, I've diagnosed and fixed the problem. git.sv is now running a new binary. The trigger was a buggy spider that was provoking an infloop in cgit. Once there were 15-20 cgit.cgi process stuck in this infloop, the system would become essentially unusable. I am having problems with my CVS commits right now, it is taking many minutes to update a few web pages. Apparently it is stuck at the Triggering webpages update... phase. In fact it just crashed: IOError: [Errno 4] Interrupted system call but the webpage update apparently succeeded. The machine cvs.savannah.nongnu.org seems to be under heavy load, with lots of svn blame, viewvc.cgi and cvs processes competing for i/o. Let me know if I can be of any help. Alex. load average: 105.00, 106.65, 83.33 Looks like it's under high load... investigating.
Re: [Savannah-hackers-public] Re: bzr + savannah.gnu.org slow again
On 03/06/2011 08:35 PM, Michael J. Flickinger wrote: On 03/06/2011 07:42 PM, Alex Fernandez wrote: Hi all, On Mon, Feb 28, 2011 at 2:02 PM, Jim Meyeringj...@meyering.net wrote: Glenn Morris wrote: At 23.30 PST on Sunday 27 Feb, it is very slow again. ~ 35 mins for a ~ 5 line commit. It was alright mid-afternoon PST. Can you not just disable the cgit service (it seems to still be available) until there is a solution for this problem? Being able to commit is more important. Hi Glenn, I've diagnosed and fixed the problem. git.sv is now running a new binary. The trigger was a buggy spider that was provoking an infloop in cgit. Once there were 15-20 cgit.cgi process stuck in this infloop, the system would become essentially unusable. I am having problems with my CVS commits right now, it is taking many minutes to update a few web pages. Apparently it is stuck at the Triggering webpages update... phase. In fact it just crashed: IOError: [Errno 4] Interrupted system call but the webpage update apparently succeeded. The machine cvs.savannah.nongnu.org seems to be under heavy load, with lots of svn blame, viewvc.cgi and cvs processes competing for i/o. Let me know if I can be of any help. Alex. load average: 105.00, 106.65, 83.33 Looks like it's under high load... investigating. This was caused by someone crawling viewvc too fast. I restarted apache (two hours back) and it looks like we're down to a sane load now. Most of those repository viewers are pretty heavy.
Re: [Savannah-hackers-public] #savannah on irc.gnu.org
Hi Alfred, I'm going to have to agree here. It's very annoying to have all the GNU channels on one network (Freenode) and the savannah channel residing on another. Further, since irc.gnu.org points to irc.freenode.net, a lot of users login to Freenode's savannah channel with questions that often go ignored. I think since Freenode is GNU's official IRC server that we should move the channel back. Simply put, to users, I think going to irc.gnu.org #savannah should be the proper channel. Is anyone opposed to this? -- Michael Alfred M. Szmidt wrote: Hi, I find it immensly annoying that irc.gnu.org isn't hosting #savannah, while I understand Sylvain's concerns regarding TOR, it does more harm than good to have some IRC channels (specially ones as important as #savannah, which has been invaluable when shit hits the fan) on random servers. If freenode for whatever reason cannot support our needs, then maybe we should switch. But I think that #savannah should exist, and be active when you connect to irc.gnu.org. Cheers, Alfred
Re: [Savannah-hackers-public] /usr/src/savane on frontend stale
I don't see anything there that looks problematic. I applied the changes to my local copy of savannah and everything is working just fine. So, you could have pull'ed with confidence. Thanks, Michael Karl Berry wrote: I believe the active checkout of the savane-cleanup git repo for savannah is /usr/src/savane on frontend. I just made a tiny tweak to frontend/php/my/groups.php. I was scared to blindly run git pull, so Jim M (hi Jim) kindly provided this recipe to see what would be changed: git fetch git diff master origin/master I'll append the output, plus the git status output for good measure. (Looks like a couple files should be checked in.) And indeed, there are a few changes besides mine, notably one that looks, well, crucial -- admin password setting. (I've replaced the actual hash values with obvious placeholders.) The other non-me changes don't look problematic to my uneducated eye, but I have no clue what to do about the adminpw thing. Michael, Jim, anyone, help? Thanks, Karl diff --git a/db/mysql/bootstrap.sql b/db/mysql/bootstrap.sql index 639099b..9f50191 100644 --- a/db/mysql/bootstrap.sql +++ b/db/mysql/bootstrap.sql @@ -6,7 +6,7 @@ -- user 'admin', password 'admin' -- (account/register.php) INSERT INTO user (user_name, user_pw, add_date, status, realname) -VALUES ('admin', 'SHORT-HEX-PW', UNIX_TIMESTAMP(NOW()), 'A', 'Administrator'); +VALUES ('admin', 'LONG-MD5-PW', UNIX_TIMESTAMP(NOW()), 'A', 'Administrator'); -- siteadmin project -- (register/*) diff --git a/frontend/php/include/init.php b/frontend/php/include/init.php index eba72dc..4c3b70b 100644 --- a/frontend/php/include/init.php +++ b/frontend/php/include/init.php @@ -226,6 +226,8 @@ if ($sys_debug_on == true) { case E_USER_NOTICE: print User Notice;break; case E_STRICT:print Strict Notice; break; case E_RECOVERABLE_ERROR: print Recoverable Error; break; +/* E_DEPRECATED - PHP = 5.3 : */ +case 8192:return false; // too much noise default: print Unknown error ($errno); break; } print '/strong'; @@ -413,7 +415,7 @@ user_guess(); # if we got an item_id and no group_id we need to get the appropriate # group_id -if (!isset($group_id) !isset($group) isset($item_id)) +if (!isset($group_id) !isset($group) isset($item_id) in_array(ARTIFACT, array('bugs', 'patch', 'task', 'cookbook', 'support'))) { $result = db_execute(SELECT group_id FROM .ARTIFACT. WHERE bug_id=?, array($item_id)); if (db_numrows($result)) diff --git a/frontend/php/my/groups.php b/frontend/php/my/groups.php index f7f6009..ab293e7 100644 --- a/frontend/php/my/groups.php +++ b/frontend/php/my/groups.php @@ -294,7 +294,7 @@ print $HTML-box_top(_(Request for Inclusion),'',1); print 'div class=boxitem'.\n; print 'p'; #print _(If there is a project - or several - you would like to be member of, to be able to fully contribute, it is possible to search for the names in the whole group database with the following search tool. A list of groups will be generated, depending on the word(s) typed in this form.).\n; -print _(Type below the name of the project you want to contribute to. Joining a project means getting write access to the code repositories and involves responsibilities. Usually you will first contact the project developers (e.g. using the project mailing list) before requesting formal inclusion using this form.).\n; +print _(Type below the name of the project you want to contribute to. Joining a project means getting write access to the project repositories and trackers, and involves responsibilities. Therefore, usually you would first contact the project developers (e.g., using a project mailing list) before requesting formal inclusion using this form.).\n; print '/p'; print ' @@ -525,4 +525,4 @@ print html_splitpage(3); $HTML-footer(array()); -? \ No newline at end of file +? # On branch master # Your branch and 'origin/master' have diverged, # and have 35 and 3 different commit(s) each, respectively. # # Changed but not updated: # (use git add file... to update what will be committed) # (use git checkout -- file... to discard changes in working directory) # # modified: frontend/php/account/lostpw.php # modified: frontend/php/include/trackers_run/index.php # # Untracked files: # (use git add file... to include in what will be committed) # # curpatch # diff # frontend/php/register2.bak/ no changes added to commit (use git add and/or git commit -a)
Re: [Savannah-hackers-public] /usr/src/savane on frontend stale
It's a schema change when installing a new copy of savane-cleanup. Karl Berry wrote: I don't see anything there that looks problematic. So what is that admin password change in bootstrap.sql about? It makes me fear that the next time savannah reboots the sql admin pw will be wrong. Or something. Unless it's just two different representations of the same actual pw. So, you could have pull'ed with confidence. I couldn't have been confident, but I could have been lucky :). Anyway, I did the pull now. Thanks, k
Re: [Savannah-hackers-public] randy kobes passed away - ftpmirror redirector
A cronjon now exists on the download vm for keeping an up-to-date list: /root/bin/gnu-mirror-geoip-sync.pl I still need to implement a file-refreshing feature for the Apache2::Geo::Mirror module. Karl Berry wrote: One problem here is that the Apache2::Geo::Mirror module only loads the mirror list upon initialization of Apache, Wow. That would explain a lot of things where users keep reporting outdated mirrors no longer in the list, etc. I think the most sane thing to do here is subclass Apache2::Geo::Mirror and add a simple feature that checks the last modified time of the /usr/local/share/GeoIP/gnu-ftpmirror.txt file; if it changes, reload the We'd want to do the same thing for the existing non-gnu redirector too, of course. Ideally even pass that fix back upstream, if there still is an upstream ... Thanks for all, k
Re: [Savannah-hackers-public] randy kobes passed away - ftpmirror redirector
Implementing this now.. this will be testable on gnu-ftpmirror.savannah.gnu.org. Once it's setup, we can have the dns for ftpmirror.gnu.org changed. On 02/26/2011 06:41 PM, Karl Berry wrote: I belatedly learned that Randy Kobes, the person who was running the ftpmirror.gnu.org redirector, passed away last September. http://perlbuzz.com/2010/11/passing-of-randy-kobes.html Fortunately, Sylvain implemented the redirector for dl.sv.gnu.org on savannah and left some notes about it in download.txt (in bzr). The question is about the redirector for http://ftpmirror.gnu.org, which has always been running on Randy's machine. It seems the path of least resistance to now also implement that on savannah, since sv already has the necessary modules, etc. Sound plausible? If anyone else is interested in taking this on, great, otherwise I guess I feel duty-bound to be the one ... ? Thanks, k
Re: [Savannah-hackers-public] randy kobes passed away - ftpmirror redirector
Currently the list is hosted on the download vm with the mirror list here: /usr/local/share/GeoIP/gnu-ftpmirror.txt, as you've already found. I'll write a script to update the mirror daily (later tonight, I hope.) One problem here is that the Apache2::Geo::Mirror module only loads the mirror list upon initialization of Apache, so every time the list is updated, apache would need restarted, which I'd like to avoid. I think the most sane thing to do here is subclass Apache2::Geo::Mirror and add a simple feature that checks the last modified time of the /usr/local/share/GeoIP/gnu-ftpmirror.txt file; if it changes, reload the file. Karl Berry wrote: Hi Michael, Ok, this is testable now on http://gnu-ftpmirror.savannah.gnu.org Wow, incredibly quick work, thanks. It works fine for me too, right now. Did you set up some method for updating it, though? I did not see anything new in the cron jobs, but I certainly could have missed it. The mirror list changes all the time (e.g., today the artfiles.org url changed from what I see in /usr/local/share/GeoIP/gnu-ftpmirror.txt). Randy's redirector worked off the mirmon installed at http://gnu-mirmon.basemirror.de (by other volunteers). You can get that mirmon status file from http://gnu-mirmon.basemirror.de/status.txt, also to be fed into /usr/local/bin/mirmon2geoip.pl. Then there's the matter of also getting the mirmon conf file for mirmon2geoip. The ultimate source is gnu.org/prep/FTP (or ftp.html). The basemirror guys (contact info at the end of /gd/gnuorg/web/FTP.contacts, I am hesitant to post email addresses publicly) wrote another script to massage that into the mirmon conf format. We could also run that script, or (my preference, to ensure that the mirmon data and site list are in sync) just get their resulting file. Unfortunately the script is written in an unreadable (to me anyway) style, but here it is FWIW. wget http://www.gnu.org/prep/ftp.html -O /root/mirmon-1.37/gnu-org-ftp.html # grep -i LI /root/mirmon-1.37/gnu-org-ftp.html | perl -lane 'm~^(\s*)LI(.+)~i;($sp,$l)=($1,$2);exit if(length($sp)2);if($sp eq q[]){$r=$l}if($r=~m~United States~i){$r.=q[ ]}else{$r=q[]}if($sp eq q[ ]){$c=$l}if($l=~m~HREF=([^]+)~i){($l)=($1);}$l=~s~\s*.*~~;if($sp eq q[ ]){$rc=$r.$c;$rcm=$rc;$rcm=~s~[^a-z]+~-~gi;$rc=~s~\s{2,}~, ~;push @a,sprintf qq[%s %s\n],$rcm,$l;$h-{qq[$rcm]}=$rc;}sub END{open(MLGF,,/root/mirmon-1.37/mirror-list-gnu-ftp.txt);print MLGF sort @a;open(CL,,/root/mirmon-1.37/countries.list);foreach(sort keys %{$h}){printf CL qq[%s %s\n],$_,$h-{qq[$_]}}}' ; cat /root/mirmon-1.37/mirror-list-gnu-ftp.txt # /root/mirmon-1.37/mirmon -v -get update -c /root/mirmon-1.37/mirmon-gnu-ftp.conf Thanks, karl
Re: [Savannah-hackers-public] Stepping Down of Savannah Maintenance
Sylvain, I'd love for you to stay, if you decide to change your mind... In the case that you actually want to stay away from the project, I'll devote more time to sysadmin'ing Savannah and step in where you left off. I've been getting back into the groove of getting back-end support requests done anyways. Hope all is well, Michael Sylvain Beucler wrote: Hi, I hereby step down from my unofficial position of Savannah maintainer. I've been contributing for 7 years and now I feel the need to look for new challenges. The system architecture is documented at: http://savannah.gnu.org/maintenance/SavannahArchitecture and the up-to-date core config files can be found at: bzr branch bzr://bzr.savannah.gnu.org/administration I'll be reasonably available to answer questions people have on the system when the above documentation is not enough. Cheers!
Re: [Savannah-hackers-public] [gnu.org #649274] colonialone ailing
Hi, You are likely subscribed to the Savannah-hackers-public mailing list. If you wish to unsubscribe, you may do so here: http://lists.gnu.org/mailman/listinfo/savannah-hackers-public -- Michael J. Flickinger On 12/20/2010 06:23 PM, Ozcan Tercan via RT wrote: REPLIES GO TO REQUESTORS BY DEFAULT. URL: http://rt.gnu.org/Ticket/Display.html?id=649274 Hey Guys, Can some one take me off this mailing list. Would really appreciate it. Thanks. On Mon, Dec 20, 2010 at 8:18 PM, Bernie Innocenti via RT sysad...@gnu.orgwrote: REPLIES GO TO REQUESTORS BY DEFAULT. URL: http://rt.gnu.org/Ticket/Display.html?id=649274 [peabo - Mon Dec 20 13:21:11 2010]: So it appears that if we accept 4 hours of downtime (no VMs running) we can add one disk to the RAID. Adding two disks would take longer, but probably not a lot longer. Trying to add disks without an explanation of why the overloads take place seems to be risky. Ok, we replaced the two suspect drives with two new ones. We did not add them to the arrays yet. -- Bernie Innocenti Systems Administrator, Free Software Foundation Hey Guys, Can some one take me off this mailing list. Would really appreciate it. Thanks. On Mon, Dec 20, 2010 at 8:18 PM, Bernie Innocenti via RT sysad...@gnu.org mailto:sysad...@gnu.org wrote: REPLIES GO TO REQUESTORS BY DEFAULT. URL: http://rt.gnu.org/Ticket/Display.html?id=649274 [peabo - Mon Dec 20 13:21:11 2010]: So it appears that if we accept 4 hours of downtime (no VMs running) we can add one disk to the RAID. Adding two disks would take longer, but probably not a lot longer. Trying to add disks without an explanation of why the overloads take place seems to be risky. Ok, we replaced the two suspect drives with two new ones. We did not add them to the arrays yet. -- Bernie Innocenti Systems Administrator, Free Software Foundation
Re: [Savannah-hackers-public] FSF fundraising widget
On 12/14/2010 12:51 PM, John Sullivan wrote: Savannah hackers, In previous years (last year IIRC), Savannah has displayed the FSF fundraising widget for the months of December and January, to promote our appeal and help raise resources -- some of which are of course spent on equipment and staff resources related to Savannah. Can you do that again? The widget code is at https://my.fsf.org/associate/widget/. The slim version for use in the sidebar might be the best choice. Thanks, Hi John, I added this to Savannah's left menu. Feel free to comment if you don't like its position or have any other concerns. Thanks, Michael J. Flickinger
[Savannah-hackers-public] Re: [Savannah-cvs] [Beuc] spam cleanup
Cleaned up this spam this morning. On Fri, Feb 15, 2008 at 06:46:23PM +, simon wrote: Hi Beuc.. I see you got spammed across quite a few pages. If you don't already know, there are some useful expunge* methods in ZWiki/Edit.py. Happy to help on freenode #zwiki if you need. -- forwarded from https://savannah.gnu.org/maintenance/[EMAIL PROTECTED]://savannah.gnu.org/maintenance ___ Savannah-cvs mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/savannah-cvs -- Michael J. Flickinger
Re: [Savannah-hackers-public] Re: [Savannah-cvs] [Beuc] spam cleanup
No, I didn't... just removed the spam with undo. On Fri, Feb 15, 2008 at 09:29:24PM +0100, Sylvain Beucler wrote: Hi, Did you use the expunge function on the 2 spammer logins? (just to know if that was enough, I never really tried it) Too bad the savannah-cvs archives get spammed too :/ -- Sylvain On Fri, Feb 15, 2008 at 02:12:24PM -0500, Michael J. Flickinger wrote: Cleaned up this spam this morning. On Fri, Feb 15, 2008 at 06:46:23PM +, simon wrote: Hi Beuc.. I see you got spammed across quite a few pages. If you don't already know, there are some useful expunge* methods in ZWiki/Edit.py. Happy to help on freenode #zwiki if you need. -- forwarded from https://savannah.gnu.org/maintenance/[EMAIL PROTECTED]://savannah.gnu.org/maintenance ___ Savannah-cvs mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/savannah-cvs -- Michael J. Flickinger -- Michael J. Flickinger
[Savannah-hackers-public] Re: [gnu.org #278693] Lists web interface occasionally broken
Looks like the list axiom-mail@nongnu.org won't quit work right. In regard to https://savannah.gnu.org/support/?func=detailitemitem_id=105218. Same issue? On Mon, Mar 13, 2006 at 01:53:47PM -0500, Justin Baugh via RT wrote: [mjflick - Sun Mar 12 20:50:16 2006]: It appears that some lists won't update from mailman's web interface. For example, bug-hurd won't update from mailman, but savannah-announce will. Could you please look into this. Michael, 23 lists are affected: gnustep-webmasters bug-hurd info-gnu-emacs bug-gnu-emacs bug-gnustep bug-guile info-gnu-events info-digitalspeech bug-oleo glibc-sc chinese-volunteers info-fsf help-global bug-gnu-chess dmca-activists bug-gnu-sql bug-panorama help-glpk commit-grub bug-glpk This is the result of filesystem corruption on lists.gnu.org related to the regular problems we were experiencing. We are attempting to restore the affected configurations from tape, but this takes a *very* long time. I am hoping to have the problem fixed by this evening or tomorrow morning at the latest. We apologize, once again, for the inconvenience. -Justin -- Justin Baugh (baughj at gnu dot org) Systems Administrator Free Software Foundation -- Michael J. Flickinger
[Savannah-hackers-public] Re: [gnu.org #278693] Lists web interface occasionally broken
Looks like the list axiom-mail@nongnu.org won't quit work right. In regard to https://savannah.gnu.org/support/?func=detailitemitem_id=105218. Same issue? On Mon, Mar 13, 2006 at 01:53:47PM -0500, Justin Baugh via RT wrote: [mjflick - Sun Mar 12 20:50:16 2006]: It appears that some lists won't update from mailman's web interface. For example, bug-hurd won't update from mailman, but savannah-announce will. Could you please look into this. Michael, 23 lists are affected: gnustep-webmasters bug-hurd info-gnu-emacs bug-gnu-emacs bug-gnustep bug-guile info-gnu-events info-digitalspeech bug-oleo glibc-sc chinese-volunteers info-fsf help-global bug-gnu-chess dmca-activists bug-gnu-sql bug-panorama help-glpk commit-grub bug-glpk This is the result of filesystem corruption on lists.gnu.org related to the regular problems we were experiencing. We are attempting to restore the affected configurations from tape, but this takes a *very* long time. I am hoping to have the problem fixed by this evening or tomorrow morning at the latest. We apologize, once again, for the inconvenience. -Justin -- Justin Baugh (baughj at gnu dot org) Systems Administrator Free Software Foundation -- Michael J. Flickinger
[Savannah-hackers-public] Lists web interface occasionally broken
It appears that some lists won't update from mailman's web interface. For example, bug-hurd won't update from mailman, but savannah-announce will. Could you please look into this. Thanks, -- Michael J. Flickinger
[Savannah-hackers-public] GNU Arch
Wait no longer, support for GNU Arch is finally here! We currently are supporting GNU Arch via sftp with support for email commit notifications. Currently, GNU Arch services are by request only, please request an Arch repository in a support request. For more information, see http://arch.sv.gnu.org/ ___ Message sent via/by Savannah http://savannah.gnu.org/
Re: [Savannah-hackers-public] Accessing savannah
On Tuesday 01 November 2005 10:49 am, Steven Rubin wrote: I've got a project at GNU (Electric, a circuit-design system) and a Savannah project for it. This project has both the source-code and the GNU web page. Unfortunately, I haven't accessed this in such a long time that all of my access methods are stale. I think the upgrade from SSH1 to SSH2 is part of the problem. In any case, I need renewed access to Savannah for myself and my co-developer, Dmitry Nadezhin. How do we do this? I assume that we have to submit new keys, but how? Thanks! -Steven Rubin To submit new keys, simply login to Savannah, then go to the edit keys page. (https://savannah.gnu.org/account/editsshkeys.php) After you submit your new keys, they should be functional within an hour. For more information, please read the FAQ on updating SSH keys. (https://savannah.gnu.org/faq/?group_id=5802question=User_Account_-_How_do_I_configure_my_SSH_access.txt) -- Michael J. Flickinger
[Savannah-hackers-public] Reboot Savannah
Savannah stalled earlier today, I noticed this at around noon (EST). I issued a reboot and it seems to be working fine now. -- Michael J. Flickinger