Re: [Sugar-devel] [Systems] Tidying up; images from 2010 to 2012
On 08/12/2018 10.35, James Cameron wrote: Several very old files are on the Sugar Labs download server; perhaps they should be removed; discuss? 1. the Dextrose fork, based on Fedora 11, sunjammer:/upload/dextrose (71 GB) http://download.sugarlabs.org/dextrose/ +1 to deleting the images, but maybe leave dextrose/patches/ for historical reference? Some XOs in the field might be still running those ancient images, and some of those patches were never accepted upstream. 2. a reference build for XO-1 and XO-1.5 laptops from November 2012, http://download.sugarlabs.org/images/reference-xo/ 3. DVD snapshots of activities.sugarlabs.org http://download.sugarlabs.org/images/ 4. a Sugar Creation Kit from 2011, http://download.sugarlabs.org/images/ (11 GB) -- _ // Bernie Innocenti \X/ https://codewiz.org/ ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Extended downtime for server maintenance
Today several Sugar Labs services will be down while we upgrade our hosts freedom and justice, switch hard drives, and move a few things around. We'll start in a few minutes, apologies for not notifying a few days in advance. For anything urgent, I'll be available on #sugar. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] [IAEP] Social Help [was Re: New Discourse version, update available]
Thankyou James for the summary and thankyou Sam for the quick response. As for Slack: I think it's used only by Samuel to mentor some new sysadmins. I have to admit I'm rarely on IRC these days due to lack of a good Android client, but I'd also recommend against adding new public communication channels for the project. Would someone volunteer to update chat.sugarlabs.org? Perhaps see if there are web IRC clients that are more mobile friendly than qwebirc? And Mailman3 with Hyperkitty would also be a big help for non-technical users (forum like interface for posting to lists). On September 29, 2017 6:36:19 PM EDT, Sam Parkinson wrote: >Thanks for replying James. Your pretty spot - that is a pretty good >summary of the thing. >I didn't get the email alerting about the 1-click upgrade. I'm >clicking to apply it now. >I've granted admin to quozl. If any sysadmin would like admin on the >service, please send your username and I can add you via the web >interface. If you admin, I think that will let you do 1 click >upgrades. >Sorry for dropping the ball on this one. >On Sat, 2017-09-30 at 06:46 +1000, James Cameron wrote: >> G'day Bernie, >> >> No, and no. >> >> Thanks for raising this. It is a risk to have a public service >> unmaintained, as it can lead to it being exploited. >> >> Discourse is integrated with Sugar through the Social Help feature >> [0], using the alt-shift-h shortcut or the frame icon. The feature >> was added [1, 2] in GSoC 2014 and merged in May 2015 for Sugar 0.106, >> with some good fixes since. >> >> Social Help is an underutilised feature. Rate of new posts is >> minimal, at about two per year, and I think I'm the only person >> responding [3]. >> >> Despite designed technical advantages [1] Social Help has not had the >> popularity or numerical success of Sugar Network, and with the >> board's >> decision [4] to remediate Sugar Network my guess is that the Social >> Help feature could be removed from Sugar. >> >> My guesses for why Social Help has not been used are; >> >> (a) the feature is undiscoverable, >> >> (b) deployments have set up their own forums, >> >> (c) independent users of Sugar 0.106 and later are minimal, >> >> (d) failed to gain critical mass. >> >> To answer your questions; >> >> 1. I'm not handling updates to Discourse; I've no access keys and >> don't know how to do it [5]. Sam Parkinson was handling updates at >> one stage. >> >> 2. I'm not using it for project communications, and I've asked those >> who try that to use mailing lists [6]. >> >> You mention Slack. I don't believe we have Slack. Slack isn't >> referenced in the Wiki, and hasn't been mentioned on mailing lists. >> If Slack is being used, it is a private group, and the knowledge >> hasn't been widely distributed yet. Please do that? >> >> We do have too many project communication options for the size of the >> project at the moment. I'd like to see some options close. >> >> +CC sugar-devel@ and iaep@ since I've raised several issues that may >> benefit from wider discussion. ;-) >> >> References: >> >> 0. https://wiki.sugarlabs.org/go/Features/Social_Help >> 1. https://wiki.sugarlabs.org/go/Summer_of_Code/2014/AnaBalica/Socia >> l_Help#Comparison_between_Discourse_and_Sugar_Network >> 2. https://wiki.sugarlabs.org/go/Summer_of_Code/2014/Prasoon2211/Soc >> ial_Help >> 3. https://socialhelp.sugarlabs.org/ >> 4. https://wiki.sugarlabs.org/go/Oversight_Board/Decisions#2017-04-2 >> 1_2 >> 5. https://wiki.sugarlabs.org/go/Service/Social_Help#Todo >> 6. https://socialhelp.sugarlabs.org/t/sugar-collaboration-xsce-iiab- >> xmpp-jabber-ejabberd/226/5 >> >> On Fri, Sep 29, 2017 at 01:31:46PM -0400, Bernie Innocenti wrote: >> > Is anyone handling updates for discourse? >> > >> > Is it being used for project communication? I ask because we also >> > have >> > IRC, Slack and mailing lists. >> > >> > On 29/09/17 07:57, socialh...@sugarlabs.org wrote: >> > > Hooray, a new version of Discourse <http://www.discourse.org> is >> > > available! >> > > >> > > Your version: 1.6.8 >> > > New version: *1.8.8* >> > > >> > > * >> > > >> > > Upgrade using our easy *one-click browser upgrade [...] >> > > >> > > * >> > > >>
Re: [Sugar-devel] [Systems] Maintenance downtime today for OS upgrades
On 10/07/16 12:57, Bernie Innocenti wrote: > On 09/07/16 19:15, Bernie Innocenti wrote: > >> Most notably, activities.sugarlabs.org is still down. It probably >> doesn't like PHP 7... The wiki is also very slow because I had to >> disable APC until I understand how to make Mediawiki use the new APCu. > > I'm still working to restore ASLO. It's an old codebase that was still > using PHP features that were deprecated in PHP 5.5 and removed in PHP 7... I finally got ASLO to display its front page, but activities are not being listed. I had to switch the DB backend to mysqli because the mysql extension has been removed in PHP7. Aleksey, since you're familiar with ASLO's codebase, would you be able to look into the outstanding issues? -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Maintenance downtime today for OS upgrades
On 09/07/16 19:15, Bernie Innocenti wrote: > Most notably, activities.sugarlabs.org is still down. It probably > doesn't like PHP 7... The wiki is also very slow because I had to > disable APC until I understand how to make Mediawiki use the new APCu. I'm still working to restore ASLO. It's an old codebase that was still using PHP features that were deprecated in PHP 5.5 and removed in PHP 7... -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Maintenance downtime today for OS upgrades
On 09/07/16 19:15, Bernie Innocenti wrote: > Samuel is currently restoring docker images, and meanwhile I'm upgrading > freedom. Correction: I'm currently upgrading justice. Freedom was already done earlier today. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Maintenance downtime today for OS upgrades
On 09/07/16 14:31, Bernie Innocenti wrote: > Greetings, > > today Sam and I are upgrading the SL infrastructure. Multiple servers > and virtual machines will have to be rebooted and will suffer a few > minutes downtime. > > We'll send an update after everything is done. We finished upgrading sunjammer and freedom to Ubuntu 16.04 and we're currently working on restoring services and post-upgrade cleanups. Most notably, activities.sugarlabs.org is still down. It probably doesn't like PHP 7... The wiki is also very slow because I had to disable APC until I understand how to make Mediawiki use the new APCu. Samuel is currently restoring docker images, and meanwhile I'm upgrading freedom. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Maintenance downtime today for OS upgrades
Greetings, today Sam and I are upgrading the SL infrastructure. Multiple servers and virtual machines will have to be rebooted and will suffer a few minutes downtime. We'll send an update after everything is done. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Planned system maintenance on Sunjammer starting NOW
On 02/08/2016 08:04 PM, Ignacio Rodríguez wrote: > OT: > > ignacio@*sunjammer*:~$uptime > > 20:03:55 up 1 day, 1:09, 4 users, load average: 0.42, 0.38, 0.37 > > ignacio@*sunjammer*:~$ > > :( I saved this memento: bernie@sunjammer:~$ uptime 06:35:21 up 1011 days, 16:55, 2 users, load average: 2.98, 2.34, 2.00 :-) > Ignacio Rodríguez > SugarLabs at Facebook > <https://www.facebook.com/pages/SugarLabs/187845102582> > > 2016-02-08 21:58 GMT-03:00 Samuel Cantero <mailto:scante...@gmail.com>>: > > -iaep > > Thanks for reporting. This has been fixed now. > > Best Regards, > > Samuel C. > > On Mon, Feb 8, 2016 at 2:25 PM, Jerry Vonau <mailto:m...@jvonau.ca>> wrote: > > http://lists.sugarlabs.org/archive/sugar-devel/ returns "403 > Forbidden" > > Jerry > > > On February 8, 2016 at 12:37 AM Bernie Innocenti > mailto:ber...@codewiz.org>> > > wrote: > > > > > > Sunjammer is now back online with Ubuntu 14.04 LTS. All hosted > websites > > and services appear to be running correctly, but please let us > know if > > we missed something (*). > > > > We also intended to switch to Trisquel 7, which is equivalent > to Ubuntu > > 14.04 minus the non-free packages (of which we don't use any), > but we > > ran out of time for tonight. This will require one last > reboot, but the > > outage should be very short. > > > > A big thank you to Rubén Rodríguez of the Free Software > Foundation for > > staying online all Sunday to support us during the transition. > Also big > > kudos to Samuel Cantero who stayed up until 1AM to lead the > second stage > > of the upgrade and cleanup the considerable fallout. > > > > > > (*) There are sysadmin contacts in the wiki, so please don't > send me > > private email. > > > > On 07/02/16 14:26, Bernie Innocenti wrote: > > > Today Sunjammer will undergo a long overdue maintenance > session to > > > bring > > > the OS up to date. > > > > > > The following services are affected: > > > > > > - Mailing lists > > > - Main wiki > > > - Local Labs wikies > > > - ASLO > > > - sugarlabs.org <http://sugarlabs.org> email delivery > > > - planet.sugarlabs.org <http://planet.sugarlabs.org> > > > > > > And a few minor things: > > > > > > - secondary nameserver > > > - download.sugarlabs.org <http://download.sugarlabs.org> > > > - Munin > > > - shell accounts (including any IRC bouncers and personal > sites) > > > > > > We'll have to upgrade the base system in stages, from one > LTS release > > > to > > > the next, so things will be going up and down for the entire > afternoon. > > > At the end, we'll be running Trisquel 7 (roughly equivalent > to Ubuntu > > > 14.04). We'll perform another upgrade to Trisquel 8 when it > becomes > > > available later this year. > > > > > > My apologies for the very short notice: Ruben, Samuel and I > have been > > > planning this since last week, but I forgot to send out a public > > > notification. > > > > -- > > _ // Bernie Innocenti > > \X/ http://codewiz.org > > ___ > > Sugar-devel mailing list > > Sugar-devel@lists.sugarlabs.org > <mailto:Sugar-devel@lists.sugarlabs.org> > > http://lists.sugarlabs.org/listinfo/sugar-devel > ___ > Systems mailing list > syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/systems > > > > ___ > Systems mailing list > syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/systems > > > > > ___ > Systems mailing list > syst...@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/systems > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Planned system maintenance on Sunjammer starting NOW
Sunjammer is now back online with Ubuntu 14.04 LTS. All hosted websites and services appear to be running correctly, but please let us know if we missed something (*). We also intended to switch to Trisquel 7, which is equivalent to Ubuntu 14.04 minus the non-free packages (of which we don't use any), but we ran out of time for tonight. This will require one last reboot, but the outage should be very short. A big thank you to Rubén Rodríguez of the Free Software Foundation for staying online all Sunday to support us during the transition. Also big kudos to Samuel Cantero who stayed up until 1AM to lead the second stage of the upgrade and cleanup the considerable fallout. (*) There are sysadmin contacts in the wiki, so please don't send me private email. On 07/02/16 14:26, Bernie Innocenti wrote: > Today Sunjammer will undergo a long overdue maintenance session to bring > the OS up to date. > > The following services are affected: > > - Mailing lists > - Main wiki > - Local Labs wikies > - ASLO > - sugarlabs.org email delivery > - planet.sugarlabs.org > > And a few minor things: > > - secondary nameserver > - download.sugarlabs.org > - Munin > - shell accounts (including any IRC bouncers and personal sites) > > We'll have to upgrade the base system in stages, from one LTS release to > the next, so things will be going up and down for the entire afternoon. > At the end, we'll be running Trisquel 7 (roughly equivalent to Ubuntu > 14.04). We'll perform another upgrade to Trisquel 8 when it becomes > available later this year. > > My apologies for the very short notice: Ruben, Samuel and I have been > planning this since last week, but I forgot to send out a public > notification. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Planned system maintenance on Sunjammer starting NOW
Today Sunjammer will undergo a long overdue maintenance session to bring the OS up to date. The following services are affected: - Mailing lists - Main wiki - Local Labs wikies - ASLO - sugarlabs.org email delivery - planet.sugarlabs.org And a few minor things - secondary nameserver - download.sugarlabs.org - Munin - shell accounts (including any IRC bouncers and personal sites) We'll have to upgrade the base system in stages, from one LTS release to the next, so things will be going up and down for the entire afternoon. At the end, we'll be running Trisquel 7 (roughly equivalent to Ubuntu 14.04). We'll perform another upgrade to Trisquel 8 when it becomes available later this year. My apologies for the very short notice: Ruben, Samuel and I have been planning this since last week, but I forgot to send out a public notification. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Error on trac reading a bug #4863
By the way, found this relevant article today on Hacker News: http://sirupsen.com/production-docker/ The comments are also interesting: https://news.ycombinator.com/item?id=9961537 On 28/07/15 19:48, Bernie Innocenti wrote: On 07/28/2015 09:27 AM, Samuel Cantero wrote: Hi All, On Tue, Jul 28, 2015 at 6:52 AM, Sam P. mailto:sam@sam.today>> wrote: Hi Bernie, On Tue, Jul 28, 2015 at 2:12 PM Bernie Innocenti mailto:ber...@codewiz.org>> wrote: 1. We need monitoring for containers health with notifications when parameters go off thresholds. Similar to what Munin does. I'm not sure how that would help us. A container running out of resources doesn't crash - parsoid runs almost using all the allocated memory and parsoid+trac+others hit their cpu quotas sometimes and just keep running. A container running out of memory dies. By default, the memory limit value for the container is set to twice as much as the memory parameter we specify while starting a container. The limit value is the sum of memory and swap. We do not have swap accounting enabled yet on freedom, but for sure we are going to have it at the next reboot. In the best Google tradition, freedom doesn't have any swap space at all :-) You can test it with the stress image (as we did before for the CPU): sudo docker run --rm -m 128m --name testmem scg/stress --vm 1 --vm-bytes 256M If you allocate over twice the amount of memory we assign you will get the following error: stress: info: [1] dispatching hogs: 0 cpu, 0 io, 1 vm, 0 hdd stress: FAIL: [1] (415) <-- worker 7 got signal 9 stress: WARN: [1] (417) now reaping child worker processes stress: FAIL: [1] (421) kill error: No such process stress: FAIL: [1] (451) failed run completed in 3s By default, Docker kills processes in a container if an out-of-memory (OOM) error occurs. We can change this behaviour on a container but we are not doing it now. See my response to Sam P.: killing the entire container is almost always the right response to an OOM. Maybe it would be more useful to monitor client metrics, like can we reach the service over http? I guess It would be useful. We can do it with nagios. Cool, does it support Docker? I used Nagios a long time ago at the FSF and it was pretty good. The reason I was using munin at SL is that it was easier to setup and I didn't know Nagios back then. Moreover, Munin seems a bit of a dead project nowadays, so I'd be happy to switch. Can we run the Nagios frontend in a container too? That is a feature built into docker. I've enabled it in container.yml (as a global default) and restarted trac with the new parameters. Great. Maybe we need to enable historical resource usage in cadvisor. This would help us to understand if a container has crashed because it has run out of resources. Cool, another useful feature I didn't know about. Yes, please, let's enable resource graphing. Is there an http frontend to look at the numbers? Thanks, Sam 3. Since high-profile services have been deployed using Docker for a while, software for doing (1) and (2) certainly exists already :-) On 27/07/15 22:32, Samuel Cantero wrote: > I didn't shut it down. The trac container was already off when I checked > it. > > On Mon, Jul 27, 2015 at 9:40 PM, Sam P. <mailto:sam@sam.today <mailto:sam@sam.today>>> wrote: > > Did you check the logs before you shut it down? > > > On Tue, 28 Jul 2015 9:38 am Samuel Cantero mailto:s...@sugarlabs.org> > <mailto:s...@sugarlabs.org <mailto:s...@sugarlabs.org>>> wrote: > > Hi Gonzalo, > > The trac container was down. I just recently turned it on. > > Somebody turned it off for some reason? If it not the case, > maybe the container has reached its maximum memory or processor > limit. The docker container stops in that case. > > Greetings, > > > > On Mon, Jul 27, 2015 at 5:01 PM, Gonzalo Odiard > mailto:godi...@sugarlabs.org> <mailto:godi...@sugarlabs.org <mailto:godi...@sugarlabs.org>>> wrote: > > Today, I get a error > > 502 Bad Gateway > nginx/1.8.0 > > when try access trac. > > Gonzalo > > On Sat, Jul 25, 2015 at 8:40 PM, Gonzalo Odiard > mai
Re: [Sugar-devel] [Systems] Error on trac reading a bug #4863
On 07/28/2015 09:27 AM, Samuel Cantero wrote: > Hi All, > > On Tue, Jul 28, 2015 at 6:52 AM, Sam P. <mailto:sam@sam.today>> wrote: > > Hi Bernie, > > On Tue, Jul 28, 2015 at 2:12 PM Bernie Innocenti <mailto:ber...@codewiz.org>> wrote: > > 1. We need monitoring for containers health with notifications when > parameters go off thresholds. Similar to what Munin does. > > > I'm not sure how that would help us. A container running out of > resources doesn't crash - parsoid runs almost using all the > allocated memory and parsoid+trac+others hit their cpu quotas > sometimes and just keep running. > > > A container running out of memory dies. By default, the memory limit > value for the container is set to twice as much as the memory parameter > we specify while starting a container. The limit value is the sum of > memory and swap. > > We do not have swap accounting enabled yet on freedom, but for sure we > are going to have it at the next reboot. In the best Google tradition, freedom doesn't have any swap space at all :-) > You can test it with the stress image (as we did before for the CPU): > > sudo docker run --rm -m 128m --name testmem scg/stress --vm 1 --vm-bytes > 256M > > If you allocate over twice the amount of memory we assign you will get > the following error: > > stress: info: [1] dispatching hogs: 0 cpu, 0 io, 1 vm, 0 hdd > stress: FAIL: [1] (415) <-- worker 7 got signal 9 > stress: WARN: [1] (417) now reaping child worker processes > stress: FAIL: [1] (421) kill error: No such process > stress: FAIL: [1] (451) failed run completed in 3s > > By default, Docker kills processes in a container if an out-of-memory > (OOM) error occurs. We can change this behaviour on a container but we > are not doing it now. See my response to Sam P.: killing the entire container is almost always the right response to an OOM. > Maybe it would be more useful to monitor client metrics, like can we > reach the service over http? > > I guess It would be useful. We can do it with nagios. Cool, does it support Docker? I used Nagios a long time ago at the FSF and it was pretty good. The reason I was using munin at SL is that it was easier to setup and I didn't know Nagios back then. Moreover, Munin seems a bit of a dead project nowadays, so I'd be happy to switch. Can we run the Nagios frontend in a container too? > That is a feature built into docker. I've enabled it in > container.yml (as a global default) and restarted trac with the new > parameters. > > Great. Maybe we need to enable historical resource usage in cadvisor. > This would help us to understand if a container has crashed because it > has run out of resources. Cool, another useful feature I didn't know about. Yes, please, let's enable resource graphing. Is there an http frontend to look at the numbers? > Thanks, > Sam > > > > 3. Since high-profile services have been deployed using Docker for a > while, software for doing (1) and (2) certainly exists already :-) > > > On 27/07/15 22:32, Samuel Cantero wrote: > > I didn't shut it down. The trac container was already off when > I checked > > it. > > > > On Mon, Jul 27, 2015 at 9:40 PM, Sam P. > <mailto:sam@sam.today <mailto:sam@sam.today>>> wrote: > > > > Did you check the logs before you shut it down? > > > > > > On Tue, 28 Jul 2015 9:38 am Samuel Cantero > mailto:s...@sugarlabs.org> > > <mailto:s...@sugarlabs.org <mailto:s...@sugarlabs.org>>> wrote: > > > > Hi Gonzalo, > > > > The trac container was down. I just recently turned it on. > > > > Somebody turned it off for some reason? If it not the > case, > > maybe the container has reached its maximum memory or > processor > > limit. The docker container stops in that case. > > > > Greetings, > > > > > > > > On Mon, Jul 27, 2015 at 5:01 PM, Gonzalo Odiard > > mailto:godi...@sugarlabs.org> > <mailto:godi...@sugarlabs.org <mailto:godi...@sugarlabs.org>>> > wrote: > > > > Today, I get a error > > > > 502 Bad Gateway >
Re: [Sugar-devel] [Systems] Error on trac reading a bug #4863
On 07/28/2015 06:52 AM, Sam P. wrote: > Hi Bernie, > > On Tue, Jul 28, 2015 at 2:12 PM Bernie Innocenti <mailto:ber...@codewiz.org>> wrote: > > 1. We need monitoring for containers health with notifications when > parameters go off thresholds. Similar to what Munin does. > > > I'm not sure how that would help us. A container running out of > resources doesn't crash - parsoid runs almost using all the allocated > memory and parsoid+trac+others hit their cpu quotas sometimes and just > keep running. > > Maybe it would be more useful to monitor client metrics, like can we > reach the service over http? I wouldn't expect processes to die when they hit their quota of cpu time, network and disk I/O bandwidth. But the only possible outcome of an OOM on a Linux system is killing the process group. The reason why is that most processes are not prepared to handle malloc() returning NULL and they would just crash or misbehave. Restarting everything is better than leaving half of the processes limping around in the container in a crippled state in which all user queries would fail anyway (no database, no frontend http server...). When the filesystem fills, the appropriate action depends on whether the container contains volatile or permanent data. Frontend containers typically have no local data and can just be restarted from a clean image. Storage is only used to store logs, coredumps and stuff like that. > 2. Containers should autorestart if they crash for any reason (with the > usual measures to deal with services which crash repeatedly). > > > That is a feature built into docker. I've enabled it in container.yml > (as a global default) and restarted trac with the new parameters. Cool. Hey, thank you very much for being always so responsive in fixing things. Much appreciated. > Thanks, > Sam > > > > 3. Since high-profile services have been deployed using Docker for a > while, software for doing (1) and (2) certainly exists already :-) > > > On 27/07/15 22:32, Samuel Cantero wrote: > > I didn't shut it down. The trac container was already off when I > checked > > it. > > > > On Mon, Jul 27, 2015 at 9:40 PM, Sam P. > <mailto:sam@sam.today <mailto:sam@sam.today>>> wrote: > > > > Did you check the logs before you shut it down? > > > > > > On Tue, 28 Jul 2015 9:38 am Samuel Cantero <mailto:s...@sugarlabs.org> > > <mailto:s...@sugarlabs.org <mailto:s...@sugarlabs.org>>> wrote: > > > > Hi Gonzalo, > > > > The trac container was down. I just recently turned it on. > > > > Somebody turned it off for some reason? If it not the case, > > maybe the container has reached its maximum memory or > processor > > limit. The docker container stops in that case. > > > > Greetings, > > > > > > > > On Mon, Jul 27, 2015 at 5:01 PM, Gonzalo Odiard > > mailto:godi...@sugarlabs.org> > <mailto:godi...@sugarlabs.org <mailto:godi...@sugarlabs.org>>> wrote: > > > > Today, I get a error > > > > 502 Bad Gateway > > nginx/1.8.0 > > > > when try access trac. > > > > Gonzalo > > > > On Sat, Jul 25, 2015 at 8:40 PM, Gonzalo Odiard > > mailto:godi...@sugarlabs.org> > <mailto:godi...@sugarlabs.org <mailto:godi...@sugarlabs.org>>> wrote: > > > > Thanks Sams :) > > > > On Sat, Jul 25, 2015 at 3:58 PM, Samuel Cantero > > mailto:s...@sugarlabs.org> > <mailto:s...@sugarlabs.org <mailto:s...@sugarlabs.org>>> wrote: > > > > Hi All, > > > > I just recently check this email. The good news is > > that Sam P already dockerized this service and now > > we have the 1.0.7 Trac version. > > > > Greetings, > > > > On Fri, Jul 24, 2015 at 11:06 AM, Gonzalo Odiard > > <mailto:godi...@sugarlabs.org> > > <mailto:godi...@sugarlabs.org > <mailto:godi...@sugarlabs.org>>> wrote: > > > > I
Re: [Sugar-devel] [Systems] Error on trac reading a bug #4863
1. We need monitoring for containers health with notifications when parameters go off thresholds. Similar to what Munin does. 2. Containers should autorestart if they crash for any reason (with the usual measures to deal with services which crash repeatedly). 3. Since high-profile services have been deployed using Docker for a while, software for doing (1) and (2) certainly exists already :-) On 27/07/15 22:32, Samuel Cantero wrote: I didn't shut it down. The trac container was already off when I checked it. On Mon, Jul 27, 2015 at 9:40 PM, Sam P. mailto:sam@sam.today>> wrote: Did you check the logs before you shut it down? On Tue, 28 Jul 2015 9:38 am Samuel Cantero mailto:s...@sugarlabs.org>> wrote: Hi Gonzalo, The trac container was down. I just recently turned it on. Somebody turned it off for some reason? If it not the case, maybe the container has reached its maximum memory or processor limit. The docker container stops in that case. Greetings, On Mon, Jul 27, 2015 at 5:01 PM, Gonzalo Odiard mailto:godi...@sugarlabs.org>> wrote: Today, I get a error 502 Bad Gateway nginx/1.8.0 when try access trac. Gonzalo On Sat, Jul 25, 2015 at 8:40 PM, Gonzalo Odiard mailto:godi...@sugarlabs.org>> wrote: Thanks Sams :) On Sat, Jul 25, 2015 at 3:58 PM, Samuel Cantero mailto:s...@sugarlabs.org>> wrote: Hi All, I just recently check this email. The good news is that Sam P already dockerized this service and now we have the 1.0.7 Trac version. Greetings, On Fri, Jul 24, 2015 at 11:06 AM, Gonzalo Odiard mailto:godi...@sugarlabs.org>> wrote: I am pretty sure dnarvaez updated the trac instance a time ago. On Fri, Jul 24, 2015 at 12:03 PM, Bernie Innocenti mailto:ber...@codewiz.org>> wrote: On 07/24/2015 10:34 AM, Gonzalo Odiard wrote: > When try to read bughttp://bugs.sugarlabs.org/ticket/4863 > get the following error:"Genshi UnicodeEncodeError error while rendering > template (unknown template location)" > > A quick google search show this bug [1] > > The error is originated by simply enter a link like this > "<http://trac.edgewall.org/>" > to a ticket. According to this [2] the bug is already solved on Trac 1.0.2 > > Then: > Developers: Please don't add links between <> until the error is solved. > Systems: I don't know what version of trac we have running, could be > possible update? Who's maintaining Trac these days? The wiki points at dnarvaez and me, but the former is MIA and the latter has installed it in 2008 and not touched it ever since: https://wiki.sugarlabs.org/go/Service/bugs Perhaps one of the two Sams would like to upgrade Trac in-place or move it into a docker container? -- _ // Bernie Innocenti \X/ http://codewiz.org -- Gonzalo Odiard SugarLabs - Software [for | by] children learning ___ Systems mailing list syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> http://lists.sugarlabs.org/listinfo/systems -- Gonzalo Odiard SugarLabs - Software [for | by] children learning -- Gonzalo Odiard SugarLabs - Software [for | by] children learning ___ Systems mailing list syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> http://lists.sugarlabs.org/listinfo/systems _________
Re: [Sugar-devel] [Systems] Error on trac reading a bug #4863
On 07/24/2015 10:34 AM, Gonzalo Odiard wrote: > When try to read bug http://bugs.sugarlabs.org/ticket/4863 > get the following error:"Genshi UnicodeEncodeError error while rendering > template (unknown template location)" > > A quick google search show this bug [1] > > The error is originated by simply enter a link like this > "<http://trac.edgewall.org/>" > to a ticket. According to this [2] the bug is already solved on Trac 1.0.2 > > Then: > Developers: Please don't add links between <> until the error is solved. > Systems: I don't know what version of trac we have running, could be > possible update? Who's maintaining Trac these days? The wiki points at dnarvaez and me, but the former is MIA and the latter has installed it in 2008 and not touched it ever since: https://wiki.sugarlabs.org/go/Service/bugs Perhaps one of the two Sams would like to upgrade Trac in-place or move it into a docker container? -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Announcing] an extra UNSTABLE 0.105.3 release (code-frozen)
We might be having trouble with DKIM: the key might be expired or something. I'm in Peru and cant even ssh in to check. Perhaps SamP can help with this? On July 1, 2015 8:38:56 AM GMT-05:00, Gonzalo Odiard wrote: >Found in spam folder (again). > >There are something wrong with the way our mailing lists or servers are >configured? > >This is what gmail say about why this mail is considered spam: > >https://support.google.com/mail/answer/81126?hl=en#authentication > >As a rule, I find emails with links triggered the spam filter more >often. > >Gonzalo > >On Tue, Jun 30, 2015 at 7:16 PM, Martin Abente < >martin.abente.lah...@gmail.com> wrote: > >> Hello everyone, >> >> I just added one extra UNSTABLE release (0.105.3), basically to ease >our >> testing before the final release on next Monday (July 6). If you are >> interested in helping out, the tarballs can be downloaded from: >> >>- >> >http://download.sugarlabs.org/sources/sucrose/glucose/sugar/sugar-0.105.3.tar.xz >>- >> >http://download.sugarlabs.org/sources/sucrose/glucose/sugar-toolkit-gtk3/sugar-toolkit-gtk3-0.105.3.tar.xz >>- >> >http://download.sugarlabs.org/sources/sucrose/glucose/sugar-artwork/sugar-artwork-0.105.3.tar.xz >>- >> >http://download.sugarlabs.org/sources/sucrose/glucose/sugar-datastore/sugar-datastore-0.105.3.tar.xz >>- >> >http://download.sugarlabs.org/sources/sucrose/glucose/sugar-runner/sugar-runner-0.105.3.tar.xz >> >> Consider the source code *frozen [1]*! We will only merge changes >under >> exceptions now. >> >> Regards, >> Martin. >> >> Refs >> [1] http://wiki.sugarlabs.org/go/0.106/Roadmap >> >> ___ >> Sugar-devel mailing list >> Sugar-devel@lists.sugarlabs.org >> http://lists.sugarlabs.org/listinfo/sugar-devel >> >> > > >-- >Gonzalo Odiard > >SugarLabs - Software [for | by] children learning -- Sent from my Android device with K-9 Mail. Please excuse my brevity.___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Sugar Contribution
Forwarding to the development team. On 17/02/15 20:52, Alexandra Castelli wrote: > Hello! > > We have a team that is trying to contribute to the Sugar Calculate > activity. If this email is not directed to the correct person, would you > mind pointing me in the right direction? > > We submitted a bug a week ago, do you know how long will it take to get > it approved? We haven’t seen it show up on the bug list yet. We have > also already solved it and we can update the Calculate activity once our > bug is accepted. > > Let me know, thanks! > Alex -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] New ASLO Bundles on the Mirror
On 01/06/2015 08:20 PM, Sam P. wrote: > - Binaries built for different architectures (i686, x86_64, arm...) > > Do you have an example for this? I tried physics, but it does not have > a native part. I will try and use git-annex or a similar solution, as > discussed on irc. IIRC, the Physics bundle contains multiple binaries for box2D, one per architecture. So actually we've been doing multi-arch bundles, a bit like OSX, rather than one bundle per architecture. I'm not sure this scales well, but if we only support 3 archs that's not a concern. However, building multiarch bundles from source can get tricky because they need multiple cross-toolchains and all the target libraries. I bet the Physics' XO bundle was built manually, i.e. by taking pre-built binaries of the box2D from different places and dropping them into the right folder before invoking setup.py. This is far from being ideal, but the complexity of build systems tends to explode when you want to cover all the weird corner cases. > - Bundles manually pinned by deployments. This is crucial for > deployments that do their own QA and can't tolerate breakage during the > school year. The ASLO updater never supported this very well, and thus > many large deployments kept using the "microformat" updater along with a > wiki page or a static html page hosted on their infrastructure. > > IMHO, deployment can easily deploy their own new ASLO copy and manually > edit or rollback or not update the data files. It is relatively easy to > deploy your own new ASLO frontend and updater dataset (part of > frontend). I should probably reach out to developments doing this. Ask Walter, he knows them all. > Sorry for dropping so many new requirements on you, but... Replacing a > production system is a lot harder than designing something anew :-) > > That's fine :) > > Thanks, > Sam -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] New ASLO Bundles on the Mirror
On 01/06/2015 07:59 PM, Sam P. wrote: > But that is only 1/2 the bundles. For every commit that the bots are > notified about, the new ASLO builds a bundle as a development/latest > copy. These are available in the new ASLO UI as devel version bundles. > I plan to purge these on a cron job, so there is only 1 on the sugarlabs > servers for each activity. But this does mean that we have lots of > bundles that are not on the old ASLO. We also need a way to keep around multiple bundle versions for the same activities. There are various reasons for this: - Bundles for older versions of Sugar (initially we may need to support only the latest Sugar which gets the new bundles, but eventually we'll end up with the same situation we have today in the old ASLO) - Binaries built for different architectures (i686, x86_64, arm...) - Bundles manually pinned by deployments. This is crucial for deployments that do their own QA and can't tolerate breakage during the school year. The ASLO updater never supported this very well, and thus many large deployments kept using the "microformat" updater along with a wiki page or a static html page hosted on their infrastructure. Sorry for dropping so many new requirements on you, but... Replacing a production system is a lot harder than designing something anew :-) -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] New ASLO Bundles on the Mirror
On 01/06/2015 07:30 PM, Sam P. wrote: > Slow propagation will cause some issues. So just SSHFS it? sshfs introduces too much latency and is unsuitable for serving. Instead, run an scp (or rsync over ssh) every time something changes, like for example as a post-upload hook. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] New ASLO Bundles on the Mirror
+sugar-devel On 01/06/2015 04:52 AM, Sam P. wrote: > Hi All, > > I was wondering about the possibility of putting the new ASLO bundles on > the download.sl.o mirror. I was going to put them into a separate > folder (named "activities2"), so there may be duplication from the old > ASLO bundles, resulting in more space usage. This will hopefully be > short term :) Go ahead, all our mirrors should be fine carrying a few hundred MBs of data. However, I don't believe you for a moment when you say it's going to be short term ;-) By the way, is the old Sugar updater compatible with the new aslo design? If it's not, then we'll have to keep around the old ASLO for years to support the user base. > I was also wondering about how to move the files from the (new... not > yet running) Bot Master on Freedom to the mirror, which is on > Sunjammer. I was just planning on using SSHFS, but if you people know > anything better I would like to use that. If propagation latency isn't an issue, you could export the files with rsync from sunjammer from a cronjob every 10 minutes. You could either use a public rsyncd, or ssh with a role account not the same account who owns the files). Use a forced command in authorized_keys for extra security. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] requesting the control of sugarlabs twitter account
On 27/12/14 12:57, samson goddy wrote: > Hello > > I was asking for the password of the sugarlabs twitter account so that > i can start updating it. I told Walter Bender but and he asked me to > mail you guys, so that i can receive the password and also link the > Sugarlabs facebook page with the twitter. I didn't even know Sugar Labs had a twitter account! -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] MediaWiki upgrade on wiki.sugarlabs.org
On 04/11/14 20:37, Bernie Innocenti wrote: > On 04/11/14 19:44, Bernie Innocenti wrote: >> Hello, >> >> I'll start upgrading the main wiki soon. This will hopefully fix >> creating new user accounts from OpenID, which is currently the only way >> to log into the wiki due to persistent spammers. > > Done. Please check if you can still login and edit pages. > > Test edits go here: http://wiki.sugarlabs.org/go/SandBox Correction: http://wiki.sugarlabs.org/go/Sandbox -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] MediaWiki upgrade on wiki.sugarlabs.org
On 04/11/14 19:44, Bernie Innocenti wrote: > Hello, > > I'll start upgrading the main wiki soon. This will hopefully fix > creating new user accounts from OpenID, which is currently the only way > to log into the wiki due to persistent spammers. Done. Please check if you can still login and edit pages. Test edits go here: http://wiki.sugarlabs.org/go/SandBox -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] MediaWiki upgrade on wiki.sugarlabs.org
Hello, I'll start upgrading the main wiki soon. This will hopefully fix creating new user accounts from OpenID, which is currently the only way to log into the wiki due to persistent spammers. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Broken ASLO mirrors
+rralcala On 08/11/2014 03:48 AM, Sam P. wrote: > Hi James, > > I tried downloading the file from my mirror (sgp1.old.aslo.cf > <http://sgp1.old.aslo.cf>) and it worked a charm. The Arrnet http > mirror is more than 1 month old! Maybe change your code (just for this > run) to use the address of my mirror instead of download.sugarlabs.org > <http://download.sugarlabs.org> so you don't get arrnet. Stats (now > correct!) are here: http://mirrors.sugarlabs.org > > +bernie, I thought mirrorbrain managed this stuff :( I thought so too! Roberto, since you're now familiar with mirrorbrain, would you mind checking what's going on? If it's not easy to figure out, just disable the outdated mirror. > Sam > > On Aug 11, 2014 1:45 PM, "James Cameron" <mailto:qu...@laptop.org>> wrote: > > On Sat, Aug 09, 2014 at 06:37:43PM -0400, Bernie Innocenti wrote: > > On 08/09/2014 06:26 PM, Bernie Innocenti wrote: > > > On 08/09/2014 06:16 PM, Bernie Innocenti wrote: > > >> (dropping a bunch of redundant addresses from the cc list) > > >> > > >> I noticed a strange pattern: all the rsync mirrors appear to be > broken, > > >> but in fact they're all responsive and up to date (the > timestamp.txt > > >> file is fresh). > > >> > > >> Looks like Mirmon's health check over rsync is broken in some way. > > > > > > Ok, found the cause with strace: > > > > > > $ strace -f -e execve mirmon -v -get all -c /etc/mirmon.conf > > > [...] > > > [pid 11135] execve("/usr/bin/wget", ["/usr/bin/wget", "-q", > "-O", "-", > > > "-T", "30", "-t", "1", "rsync://mirror.aarnet.edu.au/pub > <http://mirror.aarnet.edu.au/pub>"...], [/* 40 > > > vars */]Process 11124 detached > > > > > > It's using wget with an rsync url :-( > > > > Ok, looks like the default /etc/mirmon.conf shipped with the Ubuntu > > package doesn't support rsync. After reading the f...ine manual, I > > changed it like so: > > > > probe /usr/bin/probe -q -t %TIMEOUT% %URL%timestamp.txt > > > > This seems to do the right thing. Mirrors with rsync probing urls have > > started to go green, and will go back online in a few hours > (mirmon does > > not declare them healthy until it has seen them up for long enough). > > Thanks. I still have a problem. > > In Australia, and on sunjammer, a wget of > http://download.sugarlabs.org/activities/4315/words-21.xo is being 302 > redirected to mirror.aarnet.net.au <http://mirror.aarnet.net.au> but > fails with 404. > > Now, I'm not sure if Gonzalo's link for words-21 is correct: > http://wiki.laptop.org/go/Activities/Sugarlabs/0.102 > > This is hindering olpc-os-builder. ;-) I don't yet know how to > workaround it short of duplicating Gonzalo's activity list page. > > -- > James Cameron > http://quozl.linux.org.au/ > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Broken ASLO mirrors
On 08/09/2014 06:26 PM, Bernie Innocenti wrote: > On 08/09/2014 06:16 PM, Bernie Innocenti wrote: >> (dropping a bunch of redundant addresses from the cc list) >> >> I noticed a strange pattern: all the rsync mirrors appear to be broken, >> but in fact they're all responsive and up to date (the timestamp.txt >> file is fresh). >> >> Looks like Mirmon's health check over rsync is broken in some way. > > Ok, found the cause with strace: > > $ strace -f -e execve mirmon -v -get all -c /etc/mirmon.conf > [...] > [pid 11135] execve("/usr/bin/wget", ["/usr/bin/wget", "-q", "-O", "-", > "-T", "30", "-t", "1", "rsync://mirror.aarnet.edu.au/pub"...], [/* 40 > vars */]Process 11124 detached > > It's using wget with an rsync url :-( Ok, looks like the default /etc/mirmon.conf shipped with the Ubuntu package doesn't support rsync. After reading the f...ine manual, I changed it like so: probe /usr/bin/probe -q -t %TIMEOUT% %URL%timestamp.txt This seems to do the right thing. Mirrors with rsync probing urls have started to go green, and will go back online in a few hours (mirmon does not declare them healthy until it has seen them up for long enough). >> On 08/09/2014 06:08 PM, Bernie Innocenti wrote: >>> Our CDN works like this: >>> >>> 1. mirmon checks periodically if mirrors are responding and up to date >>> >>> 2. clients send queries to download.sugarlabs.org (sunjammer) >>> >>> 3. mod_mirrorbrain in apache uses geoip to find the closest *WORKING* mirror >>> >>> 4. apache sends back a 302 redirect to the chosen mirror >>> >>> 5. user downloads the file from the mirror >>> >>> >>> So, no matter how many mirrors are broken, you should *never* get >>> redirected to an error page. Can anyone show me a "wget --save-headers" >>> showing the problem? >>> >>> >>> On 08/07/2014 08:32 AM, Martin Abente wrote: >>>> >>>> >>>> On Wed, Aug 6, 2014 at 5:54 PM, Sam Parkinson >>> <mailto:sam.parkins...@gmail.com>> wrote: >>>> >>>> We really should perge the mirrors list of dead mirrors. >>>> http://mirrors.sugarlabs.org >>>> >>>> >>>> +1, I was thinking more like disabling temporarily those who haven't >>>> been working since a few days. >>>> >>>> Maybe we should contact arrnet (the one in question) - they might >>>> fix it if you ask nicely :) >>>> >>>> I will try to setup a mirror myself - I have like 1tb unused >>>> internet on my new aslo server (digitalocean in Singapore). >>>> >>>> On Aug 7, 2014 7:50 AM, "Ignacio Rodríguez" >>> <mailto:igna...@sugarlabs.org>> wrote: >>>> >>>> Sebastian, did you remember in sugar camp? >>>> >>>> We got the same problem.. >>>> >>>> You fixed it? >>>> >>>> Or that was automatic? >>>> >>>> Ignacio Rodríguez >>>> >>>> >>>> 2014-08-06 18:44 GMT-03:00 Sebastian Silva >>>> mailto:sebast...@fuentelibre.org>>: >>>> >>>> Hi, >>>> I have experienced the same thing. Not sure it was the same >>>> mirror. >>>> >>>> Regards, >>>> Sebastian >>>> >>>> El mié, 6 de ago 2014 a las 2:47 PM, Martin Abente >>>> >>> <mailto:martin.abente.lah...@gmail.com>> escribió: >>>> >>>>> Hello everyone, >>>>> >>>>> Has anybody else noticed that some ASLO mirrors are broken? >>>>> >>>>> In example: >>>>> >>>>> The >>>>> URL >>>>> http://download.sugarlabs.org/activities/4040/finance-11.xo, >>>>> sometimes redirects >>>>> to >>>>> http://mirror.aarnet.edu.au/pub/sugarlabs/activities/4040/finance-11.xo >>>>> which is broken. >>>>> >>>>> Ideas? >>>>> >>>>> tch. >>>> >>>> ___ >>>> Systems mailing list >>>> syst...@lists.sugarlabs.org >>>> <mailto:syst...@lists.sugarlabs.org> >>>> http://lists.sugarlabs.org/listinfo/systems >>>> >>>> >>>> >>>> ___ >>>> Sugar-devel mailing list >>>> Sugar-devel@lists.sugarlabs.org >>>> <mailto:Sugar-devel@lists.sugarlabs.org> >>>> http://lists.sugarlabs.org/listinfo/sugar-devel >>>> >>>> >>> >>> >> >> > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Broken ASLO mirrors
On 08/09/2014 06:16 PM, Bernie Innocenti wrote: > (dropping a bunch of redundant addresses from the cc list) > > I noticed a strange pattern: all the rsync mirrors appear to be broken, > but in fact they're all responsive and up to date (the timestamp.txt > file is fresh). > > Looks like Mirmon's health check over rsync is broken in some way. Ok, found the cause with strace: $ strace -f -e execve mirmon -v -get all -c /etc/mirmon.conf [...] [pid 11135] execve("/usr/bin/wget", ["/usr/bin/wget", "-q", "-O", "-", "-T", "30", "-t", "1", "rsync://mirror.aarnet.edu.au/pub"...], [/* 40 vars */]Process 11124 detached It's using wget with an rsync url :-( > On 08/09/2014 06:08 PM, Bernie Innocenti wrote: >> Our CDN works like this: >> >> 1. mirmon checks periodically if mirrors are responding and up to date >> >> 2. clients send queries to download.sugarlabs.org (sunjammer) >> >> 3. mod_mirrorbrain in apache uses geoip to find the closest *WORKING* mirror >> >> 4. apache sends back a 302 redirect to the chosen mirror >> >> 5. user downloads the file from the mirror >> >> >> So, no matter how many mirrors are broken, you should *never* get >> redirected to an error page. Can anyone show me a "wget --save-headers" >> showing the problem? >> >> >> On 08/07/2014 08:32 AM, Martin Abente wrote: >>> >>> >>> On Wed, Aug 6, 2014 at 5:54 PM, Sam Parkinson >> <mailto:sam.parkins...@gmail.com>> wrote: >>> >>> We really should perge the mirrors list of dead mirrors. >>> http://mirrors.sugarlabs.org >>> >>> >>> +1, I was thinking more like disabling temporarily those who haven't >>> been working since a few days. >>> >>> Maybe we should contact arrnet (the one in question) - they might >>> fix it if you ask nicely :) >>> >>> I will try to setup a mirror myself - I have like 1tb unused >>> internet on my new aslo server (digitalocean in Singapore). >>> >>> On Aug 7, 2014 7:50 AM, "Ignacio Rodríguez" >> <mailto:igna...@sugarlabs.org>> wrote: >>> >>> Sebastian, did you remember in sugar camp? >>> >>> We got the same problem.. >>> >>> You fixed it? >>> >>> Or that was automatic? >>> >>> Ignacio Rodríguez >>> >>> >>> 2014-08-06 18:44 GMT-03:00 Sebastian Silva >>> mailto:sebast...@fuentelibre.org>>: >>> >>> Hi, >>> I have experienced the same thing. Not sure it was the same >>> mirror. >>> >>> Regards, >>> Sebastian >>> >>> El mié, 6 de ago 2014 a las 2:47 PM, Martin Abente >>> >> <mailto:martin.abente.lah...@gmail.com>> escribió: >>> >>>> Hello everyone, >>>> >>>> Has anybody else noticed that some ASLO mirrors are broken? >>>> >>>> In example: >>>> >>>> The >>>> URL >>>> http://download.sugarlabs.org/activities/4040/finance-11.xo, >>>> sometimes redirects >>>> to >>>> http://mirror.aarnet.edu.au/pub/sugarlabs/activities/4040/finance-11.xo >>>> which is broken. >>>> >>>> Ideas? >>>> >>>> tch. >>> >>> ___ >>> Systems mailing list >>> syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> >>> http://lists.sugarlabs.org/listinfo/systems >>> >>> >>> >>> ___ >>> Sugar-devel mailing list >>> Sugar-devel@lists.sugarlabs.org >>> <mailto:Sugar-devel@lists.sugarlabs.org> >>> http://lists.sugarlabs.org/listinfo/sugar-devel >>> >>> >> >> > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Broken ASLO mirrors
(dropping a bunch of redundant addresses from the cc list) I noticed a strange pattern: all the rsync mirrors appear to be broken, but in fact they're all responsive and up to date (the timestamp.txt file is fresh). Looks like Mirmon's health check over rsync is broken in some way. On 08/09/2014 06:08 PM, Bernie Innocenti wrote: > Our CDN works like this: > > 1. mirmon checks periodically if mirrors are responding and up to date > > 2. clients send queries to download.sugarlabs.org (sunjammer) > > 3. mod_mirrorbrain in apache uses geoip to find the closest *WORKING* mirror > > 4. apache sends back a 302 redirect to the chosen mirror > > 5. user downloads the file from the mirror > > > So, no matter how many mirrors are broken, you should *never* get > redirected to an error page. Can anyone show me a "wget --save-headers" > showing the problem? > > > On 08/07/2014 08:32 AM, Martin Abente wrote: >> >> >> On Wed, Aug 6, 2014 at 5:54 PM, Sam Parkinson > <mailto:sam.parkins...@gmail.com>> wrote: >> >> We really should perge the mirrors list of dead mirrors. >> http://mirrors.sugarlabs.org >> >> >> +1, I was thinking more like disabling temporarily those who haven't >> been working since a few days. >> >> Maybe we should contact arrnet (the one in question) - they might >> fix it if you ask nicely :) >> >> I will try to setup a mirror myself - I have like 1tb unused >> internet on my new aslo server (digitalocean in Singapore). >> >> On Aug 7, 2014 7:50 AM, "Ignacio Rodríguez" > <mailto:igna...@sugarlabs.org>> wrote: >> >> Sebastian, did you remember in sugar camp? >> >> We got the same problem.. >> >> You fixed it? >> >> Or that was automatic? >> >> Ignacio Rodríguez >> >> >> 2014-08-06 18:44 GMT-03:00 Sebastian Silva >> mailto:sebast...@fuentelibre.org>>: >> >> Hi, >> I have experienced the same thing. Not sure it was the same >> mirror. >> >> Regards, >> Sebastian >> >> El mié, 6 de ago 2014 a las 2:47 PM, Martin Abente >> > <mailto:martin.abente.lah...@gmail.com>> escribió: >> >>> Hello everyone, >>> >>> Has anybody else noticed that some ASLO mirrors are broken? >>> >>> In example: >>> >>> The >>> URL >>> http://download.sugarlabs.org/activities/4040/finance-11.xo, >>> sometimes redirects >>> to >>> http://mirror.aarnet.edu.au/pub/sugarlabs/activities/4040/finance-11.xo >>> which is broken. >>> >>> Ideas? >>> >>> tch. >> >> ___ >> Systems mailing list >> syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> >> http://lists.sugarlabs.org/listinfo/systems >> >> >> >> ___ >> Sugar-devel mailing list >> Sugar-devel@lists.sugarlabs.org >> <mailto:Sugar-devel@lists.sugarlabs.org> >> http://lists.sugarlabs.org/listinfo/sugar-devel >> >> > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Broken ASLO mirrors
Our CDN works like this: 1. mirmon checks periodically if mirrors are responding and up to date 2. clients send queries to download.sugarlabs.org (sunjammer) 3. mod_mirrorbrain in apache uses geoip to find the closest *WORKING* mirror 4. apache sends back a 302 redirect to the chosen mirror 5. user downloads the file from the mirror So, no matter how many mirrors are broken, you should *never* get redirected to an error page. Can anyone show me a "wget --save-headers" showing the problem? On 08/07/2014 08:32 AM, Martin Abente wrote: > > > On Wed, Aug 6, 2014 at 5:54 PM, Sam Parkinson <mailto:sam.parkins...@gmail.com>> wrote: > > We really should perge the mirrors list of dead mirrors. > http://mirrors.sugarlabs.org > > > +1, I was thinking more like disabling temporarily those who haven't > been working since a few days. > > Maybe we should contact arrnet (the one in question) - they might > fix it if you ask nicely :) > > I will try to setup a mirror myself - I have like 1tb unused > internet on my new aslo server (digitalocean in Singapore). > > On Aug 7, 2014 7:50 AM, "Ignacio Rodríguez" <mailto:igna...@sugarlabs.org>> wrote: > > Sebastian, did you remember in sugar camp? > > We got the same problem.. > > You fixed it? > > Or that was automatic? > > Ignacio Rodríguez > > > 2014-08-06 18:44 GMT-03:00 Sebastian Silva > mailto:sebast...@fuentelibre.org>>: > > Hi, > I have experienced the same thing. Not sure it was the same > mirror. > > Regards, > Sebastian > > El mié, 6 de ago 2014 a las 2:47 PM, Martin Abente > <mailto:martin.abente.lah...@gmail.com>> escribió: > >> Hello everyone, >> >> Has anybody else noticed that some ASLO mirrors are broken? >> >> In example: >> >> The >> URL >> http://download.sugarlabs.org/activities/4040/finance-11.xo, >> sometimes redirects >> to >> http://mirror.aarnet.edu.au/pub/sugarlabs/activities/4040/finance-11.xo >> which is broken. >> >> Ideas? >> >> tch. > > ___ > Systems mailing list > syst...@lists.sugarlabs.org <mailto:syst...@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/systems > > > > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Pootle SOS
How about 22:00 UTC? It's 18:00 EDT and 8:00 AEST. http://www.timeanddate.com/worldclock/converted.html?iso=20140806T18&p1=43&p2=152&p3=0&p4=137 On 08/04/2014 07:15 PM, Matthew Ciao wrote: > Hi all, > > this meeting translates to 1am for me so I will hardly be able to join. > > Will follow any email update anyway! > > /Matthew/ > > > On 5 August 2014 03:55, Martin Abente <mailto:martin.abente.lah...@gmail.com>> wrote: > > > > On Mon, Aug 4, 2014 at 1:24 PM, Bernie Innocenti <mailto:ber...@codewiz.org>> wrote: > > On 07/31/2014 01:48 PM, Martin Abente wrote: > > Hey guys, > > > > A few updates: > > > > * I just finished upgrading the exiting instance to Pootle > 2.5.1.1 > > [1]. I don't know exactly what can I test to make sure the > upgrade > > is working properly, @cjl suggestions? > > * Next step would be a fix version control integration. > > > > A few concerns: > > > > * This update process feels like walking in the dark, the > > documentation does not help much to understand what is > expected to > > work or what are the uses cases for our instance. > > Last time someone worked on Pootle, the documentation was clearly > incorrect. Somewhere, the documentation recommended deploying > Pootle as > a stand-alone process, directly listening on port 80 (i.e. without > Apache in front of it). > > So, yeah, take the documentation with a grain of salt. > > > > * I honestly worry that maintaining the current instance > (upgraded or > > not) will not be possible in the long term. > > > That was also my impression: Pootle looks like an unmaintainable > mess, > and it's not getting better. In production, it crashes and OOMs for > mysterious reasons, and that's why it's now running on a > separate VM. > > Pootle can certainly be debugged and ultimately fixed, but it would > require long-term involvement with upstream to make their > codebase more > robust. > > > > How about a meeting this Wednesday, August 6? To discuss about > these > > concerns and possible alternatives. > > Good idea. Can we do it at 11:00 EDT (15:00 UTC)? > > > 15:00 UTC (Wednesday, August 6) sounds good to me :) > > > > > > 1. http://newpootle.sugarlabs.org/about/ > > > > > > > > > > On Thu, Jul 31, 2014 at 2:21 AM, Matthew Ciao > mailto:matt...@laptop.org.au> > > <mailto:matt...@laptop.org.au <mailto:matt...@laptop.org.au>>> > wrote: > > > > Let me know if you need any help up there Martin... Django > can be a > > serious pain in the deck. :) > > > > Cheers > > > > > > > > On 31 July 2014 00:25, Martin Abente > <mailto:martin.abente.lah...@gmail.com> > > <mailto:martin.abente.lah...@gmail.com > <mailto:martin.abente.lah...@gmail.com>>> wrote: > > > > Hey Matthew, > > > > Thanks! Right now I am trying to figure out how to upgrade > > pootle packages, but after I am done we look at that :) > > > > > > On Wed, Jul 30, 2014 at 10:20 AM, Matthew Ciao > > mailto:matt...@laptop.org.au> > <mailto:matt...@laptop.org.au <mailto:matt...@laptop.org.au>>> > wrote: > > > > Hi Martin, > > > > maybe if you have time you / we can also work out > a minimal > > secure setup? > > > > Looks like newpootle needs a bit of firewalling to > start with: > > > > sudo /sbin/iptables -L -nv > > Chain INPUT (policy ACCEPT 324 packets, 29890 bytes) > > pkts bytes target prot opt in out source > > destination > > > > Chain FORWARD (policy ACCEPT 0 packets, 0 bytes
Re: [Sugar-devel] Pootle SOS
On 07/31/2014 01:48 PM, Martin Abente wrote: > Hey guys, > > A few updates: > > * I just finished upgrading the exiting instance to Pootle 2.5.1.1 > [1]. I don't know exactly what can I test to make sure the upgrade > is working properly, @cjl suggestions? > * Next step would be a fix version control integration. > > A few concerns: > > * This update process feels like walking in the dark, the > documentation does not help much to understand what is expected to > work or what are the uses cases for our instance. Last time someone worked on Pootle, the documentation was clearly incorrect. Somewhere, the documentation recommended deploying Pootle as a stand-alone process, directly listening on port 80 (i.e. without Apache in front of it). So, yeah, take the documentation with a grain of salt. > * I honestly worry that maintaining the current instance (upgraded or > not) will not be possible in the long term. That was also my impression: Pootle looks like an unmaintainable mess, and it's not getting better. In production, it crashes and OOMs for mysterious reasons, and that's why it's now running on a separate VM. Pootle can certainly be debugged and ultimately fixed, but it would require long-term involvement with upstream to make their codebase more robust. > How about a meeting this Wednesday, August 6? To discuss about these > concerns and possible alternatives. Good idea. Can we do it at 11:00 EDT (15:00 UTC)? > 1. http://newpootle.sugarlabs.org/about/ > > > > > On Thu, Jul 31, 2014 at 2:21 AM, Matthew Ciao <mailto:matt...@laptop.org.au>> wrote: > > Let me know if you need any help up there Martin... Django can be a > serious pain in the deck. :) > > Cheers > > > > On 31 July 2014 00:25, Martin Abente <mailto:martin.abente.lah...@gmail.com>> wrote: > > Hey Matthew, > > Thanks! Right now I am trying to figure out how to upgrade > pootle packages, but after I am done we look at that :) > > > On Wed, Jul 30, 2014 at 10:20 AM, Matthew Ciao > mailto:matt...@laptop.org.au>> wrote: > > Hi Martin, > > maybe if you have time you / we can also work out a minimal > secure setup? > > Looks like newpootle needs a bit of firewalling to start with: > > sudo /sbin/iptables -L -nv > Chain INPUT (policy ACCEPT 324 packets, 29890 bytes) > pkts bytes target prot opt in out source > destination > > Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) > pkts bytes target prot opt in out source > destination > > Chain OUTPUT (policy ACCEPT 186 packets, 20370 bytes) > pkts bytes target prot opt in out source > destination > > Cheers > > > On 30 July 2014 17:46, Matthew Ciao <mailto:matt...@laptop.org.au>> wrote: > > Thanks Bernie! > > Martin I enclose an apache vhost .conf file that should > work out of the box based on my reading/editing of the > one in production at the moment. > > There are a few edits so if you intend to use it please > remove my comments! :D > > Cheers > > > On 30 July 2014 12:28, Bernie Innocenti > mailto:ber...@codewiz.org>> wrote: > > Try logging in as bashint...@newpootle.sugarlabs.org > <mailto:bashint...@newpootle.sugarlabs.org> > > You have root on newpootle, but please coordinate > with Martin to avoid > stepping on each other's toes :-) > > On 07/29/2014 06:42 PM, Matthew Ciao wrote: > > Hi Martin, > > > > do you have any .htaccess files in your sites > directories? > > > > If yes you need to review the AllowOverride > directive in your config > > files and depending on what is in .htaccess you > may need to enable > > modules such as mod_rewrite. > > > > Anyway the first deny directive you have there > works for the main Apache >
Re: [Sugar-devel] Pootle SOS
Try logging in as bashint...@newpootle.sugarlabs.org You have root on newpootle, but please coordinate with Martin to avoid stepping on each other's toes :-) On 07/29/2014 06:42 PM, Matthew Ciao wrote: > Hi Martin, > > do you have any .htaccess files in your sites directories? > > If yes you need to review the AllowOverride directive in your config > files and depending on what is in .htaccess you may need to enable > modules such as mod_rewrite. > > Anyway the first deny directive you have there works for the main Apache > document-root (/) which I assume is /srv/www/ itself: check your > httpd.conf (or main config file under /etc/apache2/) and look for the > default document-root. > > Creating an exception as you did to allow access for /srv/www/ works but > that way you slow down Apache a bit because first it evaluates a deny > then an allow directive for the same document-root. > > If you post all the config files we can understand even better.. :) > > Cheers > > On Wednesday, 30 July 2014, Martin Abente > > wrote: > > Hello all, > > The instance is back, I had to made a few changes: > > 1. added ".conf" extension to the site configuration file > 2. temporarily removed caching directives, I still need to figure > what can be use to replace them. > 3. changed some permissions [1]. > > > @matthew can you re-check the configuration file? I am not expert in > apache so my fixes could be more like hacks. > > Next step is trying to upgrade pootle version... > > Refs: > 1. > http://dabase.com/blog/AH01630:_client_denied_by_server_configuration/ > > > On Mon, Jul 28, 2014 at 9:41 PM, Matthew Ciao > wrote: > > Done! > > Sent it from my private email bashintosh at gmail dot com > > On 29 July 2014 10:43, Bernie Innocenti wrote: > > By the way, you don't even have a shell account yet: > > http://wiki.sugarlabs.org/go/Service/shell > > Please apply now, you can use me as your sponsor. > > On 07/28/2014 07:52 PM, Matthew Ciao wrote: > > As a quick suggestion, if you haven't done it already I > recommend to > > start debugging by looking at the Apache logs > (/var/log/apache2 on most > > Debian-based systems). > > > > A general health-check on Apache itself could also include: > > > > * verify the server is running (process) > > * verify the server is listening on the desired ports > (netstat, lsof) > > * check firewall rules (iptables being a good place to > start) > > * inspect vhosts configurations > > > > Lemme know if you need any help Martin, > > > > /Matthew/ > > > > > > On 29 July 2014 01:27, Martin Abente > > <mailto:martin.abente.lah...@gmail.com>> wrote: > > > > Ok! > > > > > > On Mon, Jul 28, 2014 at 11:25 AM, Bernie Innocenti > > mailto:ber...@codewiz.org>> wrote: > > > > Yes, I did it yesterday. The machine boots, but > the apache > > virtual host > > doesn't seem to work any more. > > > > I'll leave the rest to you! > > > > > > On 07/28/2014 10:53 AM, Martin Abente wrote: > > > Hey Bernie! > > > > > > Did you get the time to update the machine? > > > > > > > > > On Thu, Jul 24, 2014 at 7:56 PM, Matthew Ciao > > mailto:matt...@laptop.org.au> > > > <mailto:matt...@laptop.org.au > <mailto:matt...@laptop.org.au>>> > > wrote: > > > > > > Guys please let me know if there's anything > I can do to help! > > > > > > > > > On 25 July 2014 04:01, Martin Abente > > > <mailto:m
Re: [Sugar-devel] Pootle SOS
By the way, you don't even have a shell account yet: http://wiki.sugarlabs.org/go/Service/shell Please apply now, you can use me as your sponsor. On 07/28/2014 07:52 PM, Matthew Ciao wrote: > As a quick suggestion, if you haven't done it already I recommend to > start debugging by looking at the Apache logs (/var/log/apache2 on most > Debian-based systems). > > A general health-check on Apache itself could also include: > > * verify the server is running (process) > * verify the server is listening on the desired ports (netstat, lsof) > * check firewall rules (iptables being a good place to start) > * inspect vhosts configurations > > Lemme know if you need any help Martin, > > /Matthew/ > > > On 29 July 2014 01:27, Martin Abente <mailto:martin.abente.lah...@gmail.com>> wrote: > > Ok! > > > On Mon, Jul 28, 2014 at 11:25 AM, Bernie Innocenti > mailto:ber...@codewiz.org>> wrote: > > Yes, I did it yesterday. The machine boots, but the apache > virtual host > doesn't seem to work any more. > > I'll leave the rest to you! > > > On 07/28/2014 10:53 AM, Martin Abente wrote: > > Hey Bernie! > > > > Did you get the time to update the machine? > > > > > > On Thu, Jul 24, 2014 at 7:56 PM, Matthew Ciao > mailto:matt...@laptop.org.au> > > <mailto:matt...@laptop.org.au <mailto:matt...@laptop.org.au>>> > wrote: > > > > Guys please let me know if there's anything I can do to help! > > > > > > On 25 July 2014 04:01, Martin Abente > <mailto:martin.abente.lah...@gmail.com> > > <mailto:martin.abente.lah...@gmail.com > <mailto:martin.abente.lah...@gmail.com>>> wrote: > > > > Hey Bernie! > > > > I (literally) just finished my first round of tests > setting up a > > pootle 2.5.1.1 instance from scratch. The results are > promising: > > > > 1. Setting up a pootle instance from scratch was very > > straightforward. > > 2. Creating a new project and integrating it with git > > versioning was also surprisingly straightforward. > In fact, > > no custom bits were require to make pootle import > Sugar > > project and start making commits to the remote > repository. > > > > Other premature comments: > > > > * pootle imported _almost_ all the languages, but I > had to > > manually add languages like "guarani". I still > don't know > > why, ideas? > > * I still need to check the SL wiki to have some > ideas of how > > it would be better to organize the projects. IIRC, our > > previous pootle instance followed the same sugar > modules > > like sucrose, fructose, etc. Suggestions? > > > > > > I don't think I will have the time to work on this > during the > > weeking (because of the wedding), but if you could > make sure the > > server is ready, I can continue over the week. > > > > > > Bernie if you don't have time to work on what Martin needs > I can > > step in, no problems at all. > > > > Cheers! > > > > > > > > > > On Thu, Jul 24, 2014 at 1:03 PM, Bernie Innocenti > > mailto:ber...@codewiz.org> > <mailto:ber...@codewiz.org <mailto:ber...@codewiz.org>>> wrote: > > > > Martin is taking Pootle, but if you have time to > help with > > it, I think > > it would be appreciated. > > > > I think I'll start the work on Sunday. Martin, > would you > > have time? > > > > > > On 07/21/2014 07:08 AM, Matthew Ciao wrote: >
Re: [Sugar-devel] Pootle SOS
Yes, I did it yesterday. The machine boots, but the apache virtual host doesn't seem to work any more. I'll leave the rest to you! On 07/28/2014 10:53 AM, Martin Abente wrote: > Hey Bernie! > > Did you get the time to update the machine? > > > On Thu, Jul 24, 2014 at 7:56 PM, Matthew Ciao <mailto:matt...@laptop.org.au>> wrote: > > Guys please let me know if there's anything I can do to help! > > > On 25 July 2014 04:01, Martin Abente <mailto:martin.abente.lah...@gmail.com>> wrote: > > Hey Bernie! > > I (literally) just finished my first round of tests setting up a > pootle 2.5.1.1 instance from scratch. The results are promising: > > 1. Setting up a pootle instance from scratch was very > straightforward. > 2. Creating a new project and integrating it with git > versioning was also surprisingly straightforward. In fact, > no custom bits were require to make pootle import Sugar > project and start making commits to the remote repository. > > Other premature comments: > > * pootle imported _almost_ all the languages, but I had to > manually add languages like "guarani". I still don't know > why, ideas? > * I still need to check the SL wiki to have some ideas of how > it would be better to organize the projects. IIRC, our > previous pootle instance followed the same sugar modules > like sucrose, fructose, etc. Suggestions? > > > I don't think I will have the time to work on this during the > weeking (because of the wedding), but if you could make sure the > server is ready, I can continue over the week. > > > Bernie if you don't have time to work on what Martin needs I can > step in, no problems at all. > > Cheers! > > > > > On Thu, Jul 24, 2014 at 1:03 PM, Bernie Innocenti > mailto:ber...@codewiz.org>> wrote: > > Martin is taking Pootle, but if you have time to help with > it, I think > it would be appreciated. > > I think I'll start the work on Sunday. Martin, would you > have time? > > > On 07/21/2014 07:08 AM, Matthew Ciao wrote: > > Hi folks, > > > > it looks like you're already in a good position in terms > of help from > > rgs and/or icarito so I don't think I would be very useful > at the meeting. > > > > This said, if all goes bad please involve me in the Pootle > mission as I > > am more than keen to lend a hand! :) > > > > Best, > > > > Matthew > > > > On Jul 21, 2014 4:15 AM, "Bernie Innocenti" > mailto:ber...@codewiz.org> > > <mailto:ber...@codewiz.org <mailto:ber...@codewiz.org>>> > wrote: > > > > +rgs, +icarito (both of which expressed interest in > supporting Pootle) > > > > > > On 07/20/2014 01:36 PM, Martin Abente wrote: > > > Hello everyone, > > > > > > What about investing some minutes this week to talk > about Pootle's > > > status, what is missing and how can we get it back > to life? > > > > > > I suggest this Wednesday 23 July, 13:00 UTC. > > > > It's a bit early for me, and definitely too early for rgs. > > Can we do 15:00 UTC? > > > > > > In case I miss the meeting, here's a full dump of my mind: > > > > - you, rgs and icarito already have root to > newpootle.sugarlabs.org <http://newpootle.sugarlabs.org> > > <http://newpootle.sugarlabs.org> > > > > - full backups of the old pootle machine are on > freedom, and all of you > > have root access there too > > > > - Sysadmin documentation is in the usual place (the > wiki). There are > > se
Re: [Sugar-devel] Pootle SOS
Martin is taking Pootle, but if you have time to help with it, I think it would be appreciated. I think I'll start the work on Sunday. Martin, would you have time? On 07/21/2014 07:08 AM, Matthew Ciao wrote: > Hi folks, > > it looks like you're already in a good position in terms of help from > rgs and/or icarito so I don't think I would be very useful at the meeting. > > This said, if all goes bad please involve me in the Pootle mission as I > am more than keen to lend a hand! :) > > Best, > > Matthew > > On Jul 21, 2014 4:15 AM, "Bernie Innocenti" <mailto:ber...@codewiz.org>> wrote: > > +rgs, +icarito (both of which expressed interest in supporting Pootle) > > > On 07/20/2014 01:36 PM, Martin Abente wrote: > > Hello everyone, > > > > What about investing some minutes this week to talk about Pootle's > > status, what is missing and how can we get it back to life? > > > > I suggest this Wednesday 23 July, 13:00 UTC. > > It's a bit early for me, and definitely too early for rgs. > Can we do 15:00 UTC? > > > In case I miss the meeting, here's a full dump of my mind: > > - you, rgs and icarito already have root to newpootle.sugarlabs.org > <http://newpootle.sugarlabs.org> > > - full backups of the old pootle machine are on freedom, and all of you > have root access there too > > - Sysadmin documentation is in the usual place (the wiki). There are > several pages written by the previous Pootle maintainers: > > http://wiki.sugarlabs.org/go/Service/translate > http://wiki.sugarlabs.org/go/Service/Pootle > http://wiki.sugarlabs.org/go/Service/Pootle2.5 > http://wiki.sugarlabs.org/go/Machine/pootle > > The documentation needs to be consolidated into a single page and > updated. *PLEASE*, don't neglect this step; it will save a lot of time > to future maintainers. > > - The newpootle machine is still running Ubuntu 12.04 LTS. I strongly > recommend upgrading it to 14.04 (trusty) before proceeding. It won't > take more than 1h of work and I can do this. > > - There's an outdated instance of Pootle already running on newpootle. > It was setup by a previous volunteer, but never transitioned into > production due to difficulties with the git integration (iirc, cjl might > have more on this). > > - I strongly recommend upgrading to the latest release of Pootle before > doing anything else. > > - Then, the database and data from the backups needs to be restored and > probably converted to the Pootle 2.5 format (good luck with Django db > migration!) > > - Finally, the part where all past maintainers seem to get stuck: make > the Pootle <-> git integration work. I don't know what the actual > problem is; pushing to git should be a basic feature of a translation > system and I'm surprised this can't be made to work out of the box. > > - When everything works (and please test it well, Pootle likes to crash > and burn under load), we can change the DNS to point the CNAME > translate.sugarlabs.org <http://translate.sugarlabs.org> to > newpootle. Follow the Service/nameserver > documentation page to do this, or I can do it in 2 minutes. > > -- > _ // Bernie Innocenti > \X/ http://codewiz.org > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Apache down on sunjammer
+systems@ On 07/21/2014 08:26 AM, Martin Abente wrote: > Hello, > > Any idea why the services at sunjammer were stopped? The server is up an > running, so I wonder if anyone is working there (deploying or updating > something)?. Someone must have fixed it overnight while I was sleeping. Who's the good samaritan? In the future, if something important is down, do not hesitate to ring my phone, even at night. Downtime is bad for users. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Pootle SOS
On 07/20/2014 02:15 PM, Bernie Innocenti wrote: > +rgs, +icarito (both of which expressed interest in supporting Pootle) > > > On 07/20/2014 01:36 PM, Martin Abente wrote: >> Hello everyone, >> >> What about investing some minutes this week to talk about Pootle's >> status, what is missing and how can we get it back to life? >> >> I suggest this Wednesday 23 July, 13:00 UTC. > > It's a bit early for me, and definitely too early for rgs. > Can we do 15:00 UTC? Just noticed: I have a dentist on Wed 23, and it's also going to be a long procedure :-( How about Tue 22 or Thu 24, still 15:00 UTC? > In case I miss the meeting, here's a full dump of my mind: > > - you, rgs and icarito already have root to newpootle.sugarlabs.org > > - full backups of the old pootle machine are on freedom, and all of you > have root access there too > > - Sysadmin documentation is in the usual place (the wiki). There are > several pages written by the previous Pootle maintainers: > > http://wiki.sugarlabs.org/go/Service/translate > http://wiki.sugarlabs.org/go/Service/Pootle > http://wiki.sugarlabs.org/go/Service/Pootle2.5 > http://wiki.sugarlabs.org/go/Machine/pootle > > The documentation needs to be consolidated into a single page and > updated. *PLEASE*, don't neglect this step; it will save a lot of time > to future maintainers. > > - The newpootle machine is still running Ubuntu 12.04 LTS. I strongly > recommend upgrading it to 14.04 (trusty) before proceeding. It won't > take more than 1h of work and I can do this. > > - There's an outdated instance of Pootle already running on newpootle. > It was setup by a previous volunteer, but never transitioned into > production due to difficulties with the git integration (iirc, cjl might > have more on this). > > - I strongly recommend upgrading to the latest release of Pootle before > doing anything else. > > - Then, the database and data from the backups needs to be restored and > probably converted to the Pootle 2.5 format (good luck with Django db > migration!) > > - Finally, the part where all past maintainers seem to get stuck: make > the Pootle <-> git integration work. I don't know what the actual > problem is; pushing to git should be a basic feature of a translation > system and I'm surprised this can't be made to work out of the box. > > - When everything works (and please test it well, Pootle likes to crash > and burn under load), we can change the DNS to point the CNAME > translate.sugarlabs.org to newpootle. Follow the Service/nameserver > documentation page to do this, or I can do it in 2 minutes. > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Pootle SOS
+rgs, +icarito (both of which expressed interest in supporting Pootle) On 07/20/2014 01:36 PM, Martin Abente wrote: > Hello everyone, > > What about investing some minutes this week to talk about Pootle's > status, what is missing and how can we get it back to life? > > I suggest this Wednesday 23 July, 13:00 UTC. It's a bit early for me, and definitely too early for rgs. Can we do 15:00 UTC? In case I miss the meeting, here's a full dump of my mind: - you, rgs and icarito already have root to newpootle.sugarlabs.org - full backups of the old pootle machine are on freedom, and all of you have root access there too - Sysadmin documentation is in the usual place (the wiki). There are several pages written by the previous Pootle maintainers: http://wiki.sugarlabs.org/go/Service/translate http://wiki.sugarlabs.org/go/Service/Pootle http://wiki.sugarlabs.org/go/Service/Pootle2.5 http://wiki.sugarlabs.org/go/Machine/pootle The documentation needs to be consolidated into a single page and updated. *PLEASE*, don't neglect this step; it will save a lot of time to future maintainers. - The newpootle machine is still running Ubuntu 12.04 LTS. I strongly recommend upgrading it to 14.04 (trusty) before proceeding. It won't take more than 1h of work and I can do this. - There's an outdated instance of Pootle already running on newpootle. It was setup by a previous volunteer, but never transitioned into production due to difficulties with the git integration (iirc, cjl might have more on this). - I strongly recommend upgrading to the latest release of Pootle before doing anything else. - Then, the database and data from the backups needs to be restored and probably converted to the Pootle 2.5 format (good luck with Django db migration!) - Finally, the part where all past maintainers seem to get stuck: make the Pootle <-> git integration work. I don't know what the actual problem is; pushing to git should be a basic feature of a translation system and I'm surprised this can't be made to work out of the box. - When everything works (and please test it well, Pootle likes to crash and burn under load), we can change the DNS to point the CNAME translate.sugarlabs.org to newpootle. Follow the Service/nameserver documentation page to do this, or I can do it in 2 minutes. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [SUMMARY] development meeting 18/07/14
On 07/18/2014 12:46 PM, Martin Abente wrote: > Hello everyone, > > This is a summary of the topics we discussed in our first development > meeting. Logs can be found here [1]. > > *Topic #1: What should we do for 0.104?* > > * Focus on core ideas: > o Fix or re-work collaboration? Cover web activities integration. > o Make Sugar more customizable? Get rid of control panel by > converting sections into applications. Make sugar itself > modifiable like activities. > o Make student assessment part of the sugar experience? Think of > ways in which all these statistics gathering efforts can become > part of this. > > * Performance enhancements: > o Make Sugar more responsive? Starting by the journal. > o Make activities launch time faster? Moving common bits to > sugar-toolkit, ie., text-to-speech. > > * Cover other classroom needs: > o Printing in Sugar? > o Integrate screen sharing in Sugar? > > These are initial suggestions, we all agreed that collaboration and > performance should be prioritized. > > > *Topic #2: How do we do it?* > > * Keep the current structure for phases and deadlines, ie., [2]. > * Keep the 6 months release schedules. > * Find a way to re-sync with Gnome [3] and Fedora [4], so we can > achieve the previous item. > * Have regular development meetings. Each 2 weeks? Each month? > * Have at least one code sprint for each release, ie., SF meeting in > Oct/14, and UY youth summit in Sept/14. > * Request for former Sugar developers guidance. > * Share design team responsibilities among all of us. > * Document maintainers and responsibilities. > > *Topic #3: What are the Infrastructure needs?* > > * _Bring Pootle back to life_! Call for a special meeting regarding this. Let me know when this is happening. I can't volunteer to maintain Pootle, but I can support whoever steps up to do it. > * Move activities repositories to github? > * Set up a discourse instance (GSoC project)? > > > Many thanks to gonzalo, walterbender, llaske, bashintosh and satellit > for joining and contributing to the meeting! > > Saludos, > tch. > > _Refs:_ > > 1. > http://meeting.sugarlabs.org/sugar-meeting/meetings/2014-07-18T13:29:08.html > 2. http://wiki.sugarlabs.org/go/0.100/Roadmap > 3. https://wiki.gnome.org/ThreePointThirteen > 4. http://fedoraproject.org/wiki/Releases/21/Schedule > > -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Pootle and sucrose 0.102.0
+rgs Pootle was still running on housetree when it died. There was a new VM ready to host it, but the migration was never completed. We have full backups, but someone needs to step forward and setup a new instance. Raul also expressed interest a couple of weeks ago and I gave him access to newpootle. Would you like to work together on this? On 07/01/2014 08:37 AM, Martin Abente wrote: Hello Bernie and Chris, We are preparing a new release of sucrose, but noticed no Pootle commits were added recently [1], then noticed that Pootle itself [2] is not responding. Bernie, think you could check whats going on? Regards, Martin. Refs: 1. https://github.com/sugarlabs/sugar/commits?author=pootle-sugarlabs 2. http://translate.sugarlabs.org/ -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Happy / Sad laptop system load monitor
Look, even Chrome copied our design to show users that they're opening too many tabs! :-) http://imgur.com/Ep6VhC0 On 05/23/14 09:50, Walter Bender wrote: On Fri, May 23, 2014 at 9:32 AM, Sebastian Silva mailto:sebast...@fuentelibre.org>> wrote: Thans for the archeology Bernie! I agree that perhaps happy/sad isn't the proper metaphor. I remember having a similar WindowMaker dockapp, that would show the face from Doom's guy for the same purpose. Not very informative, but fun. Now, about this implementation, as i remember, once the palette is open, proper cpu/mem graphs appear. Perhaps we can come up with a less emotionally loaded metaphor. I would disagree that user's shouldn't need to know this data, in fact with such a constrained machine as XO it is pretty useful feedback and as I remember it was part of the original Sugar design (it was not in the frame, but in the home view's relative area of use for each active icon). The original Sugar design used a different approach (a circle of activities filling up in the center of the home screen). My objection to the Smiley implementation was not that users shouldn't need to know these data. Rather, not enough information was available and it was over-simplifying something complex, masking the inherent complexity. Regards, Sebastian El vie, 23 de may 2014 a las 6:08 AM, Walter Bender mailto:walter.ben...@gmail.com>> escribió: +1 to something like conky, which is real information instead of pretending to give the kids information. if you want smiley faces, they should have real data behind them. regards. -walter On Fri, May 23, 2014 at 5:16 AM, Iain Brown Douglas mailto:i...@browndouglas.plus.com>> wrote: On Thu, 2014-05-22 at 22:34 -0005, Sebastian Silva wrote: > Hi, > I remember Sugar as deployed in Dextrose 1 had a signature happy/sad > laptop icon and both memory / cpu load indicators. > > > I am dogfooding Sugar again on a memory constrained laptop and am in > need of a memory monitor, and this would be lovely. > > > Is that code alive? Maybe it was a good idea to have by default. > I remember testing that on a low resource machine and the avatar lagged the events, and was not educational. I felt that the ability to embed a couple of elements from conky [1] in the frame would be superior. I have found conky to be understood by very young testers. Regards, Iain [1] http://conky.sourceforge.net/ > > Regards, > Sebastian > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> http://lists.sugarlabs.org/listinfo/sugar-devel -- Walter Bender Sugar Labs http://www.sugarlabs.org -- Walter Bender Sugar Labs http://www.sugarlabs.org -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Happy / Sad laptop system load monitor
Concky looks great, but I thought we could only embed Python code in the frame. If a rewrite in Python is required, then we could as well start from Tincho's original work and improve it. While the happy/sad laptop might be an over-simplification, opening the palette did reveal real memory and cpu usage data: http://wiki.sugarlabs.org/go/File:Dextrose_resource_monitoring.png The lag observed by Iain in updating the data also bothered me. The stats were being updated every 5 seconds because a reviewer was concerned about increasing idle power usage. I'd be in favor of increasing the update frequency and maybe skip the UI updates when the frame is hidden. The same optimization would benefit all the existing frame indicators and background updates to the Network view, which take a ton of CPU time. On 05/23/2014 04:08 AM, Walter Bender wrote: > +1 to something like conky, which is real information instead of > pretending to give the kids information. if you want smiley faces, they > should have real data behind them. > > regards. > > -walter > > > On Fri, May 23, 2014 at 5:16 AM, Iain Brown Douglas > mailto:i...@browndouglas.plus.com>> wrote: > > On Thu, 2014-05-22 at 22:34 -0005, Sebastian Silva wrote: > > Hi, > > I remember Sugar as deployed in Dextrose 1 had a signature happy/sad > > laptop icon and both memory / cpu load indicators. > > > > > > I am dogfooding Sugar again on a memory constrained laptop and am in > > need of a memory monitor, and this would be lovely. > > > > > > Is that code alive? Maybe it was a good idea to have by default. > > > I remember testing that on a low resource machine and the avatar lagged > the events, and was not educational. > > I felt that the ability to embed a couple of elements from conky [1] in > the frame would be superior. I have found conky to be understood by very > young testers. > > Regards, > Iain > > [1] http://conky.sourceforge.net/ > > > > Regards, > > Sebastian > > ___ > > Sugar-devel mailing list > > Sugar-devel@lists.sugarlabs.org > <mailto:Sugar-devel@lists.sugarlabs.org> > > http://lists.sugarlabs.org/listinfo/sugar-devel > > > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel > > > > > -- > Walter Bender > Sugar Labs > http://www.sugarlabs.org -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Happy / Sad laptop system load monitor
On 05/22/2014 03:39 PM, Sebastian Silva wrote: > Hi, > I remember Sugar as deployed in Dextrose 1 had a signature happy/sad > laptop icon and both memory / cpu load indicators. > > I am dogfooding Sugar again on a memory constrained laptop and am in > need of a memory monitor, and this would be lovely. > > Is that code alive? Maybe it was a good idea to have by default. The "laptop avatar" feature was controversial: the main objection I remember hearing was that the UI shouldn't be cluttered with low-level information that users shouldn't be concerned about. More interestingly, field testing in Uruguay revealed that users were sometimes using the feature in an unintended way: they'd purposely open as many activities as possible to "make the laptop cry"! Such cruelty... :-) Anyway, the smarter fix for system overload is automatically limiting the number of activities open at once, which recent versions of Sugar do. If you're still interested, the original patch was pretty simple and might still apply cleanly to the current Sugar: http://download.sugarlabs.org/dextrose/patches/0.88/sugar/cpu-and-memory-resource-indicator.patch http://download.sugarlabs.org/dextrose/patches/0.88/sugar-artwork/cpu-and-memory-indicator-icons.patch -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] git.sugarlabs.org down for unplanned maintenance
On 04/12/2014 02:07 AM, Sebastian Silva wrote: > Here I just got home. Sorry for the inconvenience I might have caused. > > Bernie, do you know which log was/is growing out of hand? Both access.log and node.sugarlabs.org.log. I discarded the first and compressed the second (it compresses very well). You can still examine it by doing: xzless access.log-20140411.xz | tail You'll see lines like this one: node.sugarlabs.org:80 181.65.159.107 - - [11/Apr/2014:19:51:36 -0400] "GET /?cmd=subscribe HTTP/1.1" 200 232 "-" "python-requests/1.2.1 CPython/2.7.0 Linux/2.6.35.13_xo1.5-20120508.1139.olpc.eb0c7a8" The problem seems to be that laptops retry the connection to /context.atom and /feedback.atom quickly. It's probably near the end of the file though. Don't try to uncompress the whole file because it's over 2GB. > Here's a report on everything I know about the issue. > We've been experiencing some performance degradation and also some > downtime in Sugar Network services (this is documented at > http://tareas.somosazucar.org/hxp/issue71 ). > We've seen a burst in users since deployment OS images with Sugar > Network features ( http://network.sugarlabs.org/stats-viewer/ growing > pretty fast user_total). > There is a notification feature that is polling the sugar network node > service. > This was causing the allocation and exhaustion of resources (open > files). Crashes got to a frequency of every hour or so. > It's code I don't understand really well, but I went ahead and patched > the Sugar Network with: > http://tareas.somosazucar.org/hxp/file66/sn_disable_notifications.patch > This made the SN much snappier and it stopped crashing. However logs > were saving a traceback several times per second. I thought I had > contained the log issue but apparently I missed some other logs (I guess > apache logs but they seem clean now). > > I took a glance at jita and could not find the growing log. > > Let me know where I can help mitigation. > > Regards > Sebastian > > > El vie, 11 de abr 2014 a las 7:51 PM, Bernie Innocenti > escribió: >> I was notified that git.sugarlabs.org was showing errors. After some >> head scraping I realized that the root filesystem on jita was full. I >> looked around and found giant request logs containing millions of >> requests apparently originating from XOs located in Peru. We've been >> DDOSed by our own creature :-) Anyway, the machine also had a giant, >> very fragmented mysql database that I'm currently cleaning up. >> Gitorious will be back online in less than 1 hour. Contact me on IRC >> if this is blocking your work, I can postpone the maintenance. >> -- >> Bernie Innocenti Sugar Labs Infrastructure Team >> http://wiki.sugarlabs.org/go/Infrastructure_Team >> ___ Sugar-devel mailing >> list Sugar-devel@lists.sugarlabs.org >> http://lists.sugarlabs.org/listinfo/sugar-devel -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] git.sugarlabs.org down for unplanned maintenance
Ok, we're back in business, with a snappier database too. Today I felt lucky so I also replaced the aging MySQL 5.1 with a shiny new MariaDB 10. Don't worry, MariaDB should be 100% backwards compatible, and we do daily dumps in case anything goes wrong. I also deleted several *millions* of records from the database for old login sessions and logs of clone actions. You may have to login again, but Gitorious feels lot faster now. We still need to keep an eye at those evil XOs that keep reconnecting to network.sugarlabs.org. Icarito, can you look at implementing some form of exponential backoff? If the fix can't be deployed within a few days, we should defend ourselves with iptables rules or at least stop logging every connection. As always, please notify us if anything malfunctions. Note that alsroot said he would be offline until Apr 13. On 04/11/2014 08:51 PM, Bernie Innocenti wrote: > I was notified that git.sugarlabs.org was showing errors. > > After some head scraping I realized that the root filesystem on jita was > full. I looked around and found giant request logs containing millions > of requests apparently originating from XOs located in Peru. > > We've been DDOSed by our own creature :-) > > Anyway, the machine also had a giant, very fragmented mysql database > that I'm currently cleaning up. Gitorious will be back online in less > than 1 hour. Contact me on IRC if this is blocking your work, I can > postpone the maintenance. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] git.sugarlabs.org down for unplanned maintenance
I was notified that git.sugarlabs.org was showing errors. After some head scraping I realized that the root filesystem on jita was full. I looked around and found giant request logs containing millions of requests apparently originating from XOs located in Peru. We've been DDOSed by our own creature :-) Anyway, the machine also had a giant, very fragmented mysql database that I'm currently cleaning up. Gitorious will be back online in less than 1 hour. Contact me on IRC if this is blocking your work, I can postpone the maintenance. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] wiki.sugarlabs.org upgraded to MW 1.22.2
I upgraded the main wiki to MediaWiki 1.22.2, which brings some new user-visible features: http://www.mediawiki.org/wiki/Release_notes/1.22#MediaWiki_1.22.2 I also patched our OpenID plugin to allow creating new users, so if you couldn't create an account in the past, please try again now and let us know if you still have any issue. Please accept my apologies on behalf of the Infrastructure Team for the inconvenience. As you may know, some persistent spammers have been vandalizing our wiki and even ReCaptcha doesn't seem to stop them. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] New year, new testing image
Sorry for the long delay, I started a draft response on a tablet that I didn't pick up for some time. On 01/05/2014 08:33 PM, Matthew Ciao wrote: > Hi Bernie, how are you going there? > Lng time no see, hope you're doing well mate! :) So I'm still here in Cambridge MA, enjoying the freezing weather of New England. And you're enjoying the nice, mild weather of Sydney, I suppose? > Can I ask to what degree you're planning to use AU images? > Would it be for testing only (yourself?) or actual major deployments? I've installed the oz images on all my XOs. I've setup a mini-museum of OLPC at Google's Cambridge office. People passing by sometimes stop and play with the laptops, so I want them to see the best version of Sugar available. Sorry I couldn't find the time to test properly and file bugs. > Very briefly, the AU software now includes a statistics-collection > software that sends data to our servers matching the serial number of > the XO against our local serial-numbers database. > > This means that if you're going to deploy the image on, say 100 laptops, > those will then sync data to our db which results in serial numbers not > matching. > > I am worried about the scale of this issue, which might fill our db with > incoherent data so perhaps (Martin, Gonzalo?) we should think about a > way to prevent any confusion in case the AU image is used somewhere else > around the world. Have you thought of asking for permission to send statistics as part of first-time setup? Then I'd just answer no and be done. Moreover, it's standard industry practice to obtain explicit user consent for data collection, and if you don't do that soon or later someone might get upset. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] New year, new testing image
On 01/01/2014 03:28 PM, Gonzalo Odiard wrote: > Happy new year! > > I have prepared a new testing image with the usual bug fixes, > and features needed by the AU deployment, for xo-4. > > You can download from: > http://dev.laptop.org/~gonzalo/AU1B/33026xx4.zd > md5file: > http://dev.laptop.org/~gonzalo/AU1B/33026xx4.zd.md5 > > A new feature included is the integration of the Help activity and the > help shortcut. > Based on the work of Kalpa Welivitigoda [1], now you can press Ctrl+Shift+H, > and get a help related with the place where you are. > NOTE: By a known bug, you need use the help activity at least one time > before using > the shortcut, to see the images. > > Testing is welcomed! Could you please release images built for all XO models? It's ok if you can't offer to support them. Also, would you mind creating a wiki page with release notes, bug report contacts, etc? -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Gambiarra game
+1 from me as well. For the record: I drafted the current policy a couple of years ago in the attempt to give activity developers a clearly documented process that they can just follow without getting stuck into a long policy discussion. As Daniel noted, the current process seems a bit too laborious and I would support shortening it in the future. To me, "the Activity Team coordinator decides on a case-by-case basis" would work too. The important point is documenting the process in advance so everyone knows how to handle future cases. On 12/24/2013 04:22 PM, Daniel Narvaez wrote: > +1 > > On Tuesday, 24 December 2013, Walter Bender wrote: > > It is current and we should be following it, IMHO. > > -walter > > On Tue, Dec 24, 2013 at 3:57 PM, Daniel Narvaez > wrote: > > (Assuming the policy is not obsolete or something, I think we > should move it > > to developer.sugarlabs.org <http://developer.sugarlabs.org>). > > > > > > On 24 December 2013 21:51, Daniel Narvaez > wrote: > >> > >> By the way, we seem to have a non responsive maintainer policy > already. > >> > >> > >> > > http://wiki.sugarlabs.org/go/Activity_Team/Policy_for_nonresponsive_maintainers > >> > >> Any reason we are not following it? > >> > >> > >> On 24 December 2013 21:49, Daniel Narvaez > wrote: > >>> > >>> On 24 December 2013 15:10, Aleksey Lim > wrote: > >>>> > >>>> On Tue, Dec 24, 2013 at 08:49:02AM -0500, Walter Bender wrote: > >>>> > IMHO, the git rep is less the issue than the ownership on > ASLO. git is > >>>> > set up for forks, ASLO less obvious. I can give Alan joint > ownership > >>>> > on ASLO. (The versions available from Luiz will still be > available > >>>> > even after Alan uploads new ones.) > >>>> > >>>> I don't see how ASLO is critically different in comparison with > git.sl.o > >>>> in this case (passing ownership). > >>> > >>> > >>> ASLO is more similar to a distribution than to gitorious > repositories and > >>> distributions usually have non-responsive maintainer policies > >>> > >>> > >>> > > https://fedoraproject.org/wiki/Policy_for_nonresponsive_package_maintainers > >>> > >>> > >>> > > http://www.debian.org/doc/manuals/developers-reference/beyond-pkging.html#mia-qa > >>> > >>>> > >>>> At the end, the important thing is that > >>>> both versions should be available for users (the original > version, and > >>>> the one which was improved by new developers). It is hardly > possible to > >>>> have only one download entity [on ASLO]. > >>> > >>> > >>> If we have to choose, I think it's more important to make an > improved > >>> activity available then old versions provided by the original > maintainer. > >>> > >>>> > >>>> In any case, that might be a topic for SN (as an ASLO's superset) > >>>> to handle this kind of issues in the future. > >>> > >>> > >>> We can reevaluate when SN takes over, but given the current > >>> infrastructure I think giving Alan joint ownership is the most > pragmatic > >>> approach. > >> > >> > >> > >> > >> -- > >> Daniel Narvaez > > > > > > > > > > -- > > Daniel Narvaez > > > > -- > Walter Bender > Sugar Labs > http://www.sugarlabs.org > > > > -- > Daniel Narvaez > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Gambiarra game
Aleksey, could you please do this? If you're busy, I'll do it over the week-end. On 12/19/13 05:11, Gonzalo Odiard wrote: > Bernie, Aleksey, > Alan proposed work in this activity a month ago. > The owner was notified, but didn't replied, > can you give him access to the git repository? > Thanks > > Gonzalo > > > On Mon, Nov 18, 2013 at 4:12 PM, Alan Jhonn Aguiar Schwyn > mailto:alan...@hotmail.com>> wrote: > > Hi, > > I want to continue working with this game. It's interesting. > Now, I have some changes: > > -add translations for other languages > -add compatibility with new sugar versions > -replace OLPCGames library for SugarGames > > Wich is the way to get permissions to original GIT: > > https://git.sugarlabs.org/gambiarra > > Regards! > > Alan > > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] git sugarlabs is down
Aleksey notified me and I got the machine back online. We couldn't determine the root cause of the crash, and justice has been running for about two years without a glitch. If it ever happens again, we'll transfer the hard drives to our hot spare freedom. I'll be on the west coast until Monday, but Dogi, Walter Bender and Luke Faraone should have access to the server room at the Media Lab so we should be covered. On 12/20/13 04:10, Gonzalo Odiard wrote: > Looks down. > cc: Bernie & Aleksey > > Gonzalo > > > On Fri, Dec 20, 2013 at 9:04 AM, Gustavo Duarte > mailto:gdua...@activitycentral.com>> wrote: > > Hi guys, > > Is git.sugarlabs.org <http://git.sugarlabs.org> service down ? > > Somebody know something ? > > Gustavo. > > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] [ANNOUNCE] Sugar Labs servers statistics
Thanks for this amazing work, Aleksey! On 11/30/13 15:26, Aleksey Lim wrote: > Hi all, > > There is a new Sugar Labs site, http://stats.sugarlabs.org/ which was > created to visually present Web analytics for sites Sugar Labs supports. > It is based on AWStats[1], an open source Web analytics reporting tool > (Piwik was considered as well but was found much slower for collecting > Web reports). > > For the majority of sites, reporting history is about 3 years (not the > whole history, though). Besides, not all Web sites, that Sugar Labs > servers host, are represented. If you think that there are missed sites, > reply to this thread or use administrative contact [2]. > > > [1] http://en.wikipedia.org/wiki/AWStats > [2] http://wiki.sugarlabs.org/go/Service/stats#Administrative_contact > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] IRC chanserv command
On 06/30/13 23:33, Samuel Greenfeld wrote: > The +n (no external messages from users not on the channel) is an > anti-spam measure and a default for all newly-created Freenode channels. > > So you might not want to remove it, even if the IRC server will let you. Thank you for the clarification. Daniel, could we make the github bot just join the channel instead of setting -n? This is how we were doing it with I've given ACLs to dnarvaez on #sugar and #sugar-meeting and left +n set. Switch the flag if there's no better way, but be prepared to switch it back in case it gets abused by spammers. > --- > SJG > > > On Sun, Jun 30, 2013 at 6:26 AM, Daniel Narvaez <mailto:dwnarv...@gmail.com>> wrote: > > The channelname being sugar of course. > > > On Sunday, 30 June 2013, Daniel Narvaez wrote: > > Hey, > > could someone with chanserv access give this command > > |/msg chanserv set #channelname mlock -n > > | > |Or well, give me access and I can do it. > > | > |This is for github commits notification witout the bot joining > every time. > | > > -- > Daniel Narvaez > > > > -- > Daniel Narvaez > > > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org <mailto:Sugar-devel@lists.sugarlabs.org> > http://lists.sugarlabs.org/listinfo/sugar-devel > > > > > _______ > Systems mailing list > syst...@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/systems > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Trac update
On 06/29/2013 01:23 PM, Daniel Narvaez wrote: > sudo doesn't seem to work, or am I supposed to get root in another way? Try now, I added you to adm and admin. For historic reasons sunjammer does not use group 'sudo', like other Ubuntu systems do. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Trac update
On 06/29/2013 09:07 AM, Daniel Narvaez wrote: > Tried the update locally and it seems pretty painless. Awesome! For posterity, could you document the upgrade procedure in the service page? Also, if you'd like to support bugs.sugarlabs.org, set yourself as the primary admin contact in the wiki. You can keep me as a secondary although I'm trying to forget Trac as much as possible. > There are broken > permissions in the eggs and attachments directories though, the group > cannot write on some of the files. Can you either fix those up or give > me root? Fixed the permissions _and_ gave you root. It may be that Trac has the wrong umask when running under apache. > On 29 June 2013 14:13, Daniel Narvaez <mailto:dwnarv...@gmail.com>> wrote: > > Actually maybe on -testing, -devel seems for trunk. Ok. > On 29 June 2013 14:12, Daniel Narvaez <mailto:dwnarv...@gmail.com>> wrote: > > Could you add me to sugartrac? I'll give it a try on -devel and > see if there are blockers. > > On 29 June 2013 13:36, Bernie Innocenti <mailto:ber...@codewiz.org>> wrote: > > On 06/28/2013 06:57 PM, Daniel Narvaez wrote: > > Hello, > > > > we seem to have a pretty old trac version. I'm trying to > improve sugar > > core bug tracking state and practices and as part of that > effort I think > > it might make sense to upgrade to 1.0. > > That would be great. > > > > Is the only reason we didn't upgrade so far lack of time? > Or do we know > > about blockers? > > Trac probably needs a new maintainer. Would you like to take > it over or > replace it with something else? > > > > Is our trac configuration somewhere? I didn't find it in git. > > The only documentation we have on Trac is this page: > > http://wiki.sugarlabs.org/go/Service/bugs > > The production instance is located on sunjammer, in > /srv/www-sugarlabs/bugs. There's also a development > environment in > bugs-devel which is accessible as > http://bugs-devel.sugarlabs.org/ > > Before touching the production instance, you might want to > try upgrading > the -devel environment to see how it goes. > > Ping me on IRC if you need any help. I have vague memories > of upgrading > Trac several years ago (hint: it's quite painful). > > -- > _ // Bernie Innocenti > \X/ http://codewiz.org > > > > > -- > Daniel Narvaez > > > > > -- > Daniel Narvaez > > > > > -- > Daniel Narvaez > > > ___ > Systems mailing list > syst...@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/systems > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Trac update
On 06/28/2013 06:57 PM, Daniel Narvaez wrote: > Hello, > > we seem to have a pretty old trac version. I'm trying to improve sugar > core bug tracking state and practices and as part of that effort I think > it might make sense to upgrade to 1.0. That would be great. > Is the only reason we didn't upgrade so far lack of time? Or do we know > about blockers? Trac probably needs a new maintainer. Would you like to take it over or replace it with something else? > Is our trac configuration somewhere? I didn't find it in git. The only documentation we have on Trac is this page: http://wiki.sugarlabs.org/go/Service/bugs The production instance is located on sunjammer, in /srv/www-sugarlabs/bugs. There's also a development environment in bugs-devel which is accessible as http://bugs-devel.sugarlabs.org/ Before touching the production instance, you might want to try upgrading the -devel environment to see how it goes. Ping me on IRC if you need any help. I have vague memories of upgrading Trac several years ago (hint: it's quite painful). -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Moving pootle to github
I'm very happy to hear that. Kudos! On 06/09/13 06:13, Daniel Narvaez wrote: > Something I would like to stress, aside from philosophical > considerations, is that the new process is working. There is not a > single patch stuck in the queue (couple of old ones from Walter but they > are just blocked on another set he submitted, which is going to land soon). > > > On 9 June 2013 11:51, Daniel Narvaez <mailto:dwnarv...@gmail.com>> wrote: > > On 9 June 2013 01:38, Bernie Innocenti <mailto:ber...@sugarlabs.org>> wrote: > > On 06/07/2013 09:10 AM, Daniel Narvaez wrote: > > No, just glucose. You can see the exact list of modules on > > https://github.com/sugarlabs/ > > By the way, what is "sugarlabs", a shared account? > > > It's an organization > > https://github.com/blog/674-introducing-organizations > > > Wouldn't this subvert GitHub's philosophy that all forks are created > equal, by creating one that looks more official than the others? > > > In my experience the large majority of github repository has an > official repo, very visibly linked from the project official > website. For example > > http://nodejs.org/ > > People fork the official repo and send patches through pull requests. > > Which is exactly what we are doing. > > > If it seems that this approach wouldn't be feasible for a > project with > many collaborators, check out http://git.kernel.org . Most of > the repos > under kernel/git/ are clones of the kernel tree with various patches > applied. The most "official" tree that I can think of is > kernel/git/torvalds/linux.git, the one maintained by Linus. > There are of > course many other public forks of the Linux kernel hosted on > other sites. > > > I can't think of a single github repository that follows the kernel > development model. I'm sure there but I'm also pretty sure it's not > the normal development model for github repositories. > > > I'm making the assumption that switching to GitHub was motivated > in part > by the desire to adopt the bazaar development style. If it's not the > case, then GitHub may not be a very good fit for a central > repository > shared by multiple committers. > > > If with bazaar development model you mean kernel like, I don't think > that was one of the reasons. But as I said I don't think github > pushes that model either. It's pretty similar to gitorious really, > just a better implementation of it :) > > In general I don't think kernel development practices are a good > model for our community, as proved by the attempt to push their > patch review practices and badly failing. We are a very different > kind of communities. > > > > > -- > Daniel Narvaez -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Moving pootle to github
On 06/07/2013 09:10 AM, Daniel Narvaez wrote: > No, just glucose. You can see the exact list of modules on > https://github.com/sugarlabs/ By the way, what is "sugarlabs", a shared account? Wouldn't this subvert GitHub's philosophy that all forks are created equal, by creating one that looks more official than the others? If it seems that this approach wouldn't be feasible for a project with many collaborators, check out http://git.kernel.org . Most of the repos under kernel/git/ are clones of the kernel tree with various patches applied. The most "official" tree that I can think of is kernel/git/torvalds/linux.git, the one maintained by Linus. There are of course many other public forks of the Linux kernel hosted on other sites. I'm making the assumption that switching to GitHub was motivated in part by the desire to adopt the bazaar development style. If it's not the case, then GitHub may not be a very good fit for a central repository shared by multiple committers. Just my $0.02, of course. Advance apologies if it's something that has been previously discussed on sugar-devel@ while I wasn't looking. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Moving pootle to github
+rralcala,alsroot Roberto, this might be a good task to get you started on Pootle. Chris and Aleksey can probably assist you. On 06/02/2013 10:09 AM, Daniel Narvaez wrote: > Ping? > > On 23 May 2013 17:32, Daniel Narvaez <mailto:dwnarv...@gmail.com>> wrote: > > Hello, > > we need to move pootle to push on github for a few modules (see > https://github.com/sugarlabs/). We will also need to turn it off for > a bit while we resync the repositories. Who has access/knowledge to > help with this? > > -- > Daniel Narvaez > > > > > -- > Daniel Narvaez > > > ___ > Systems mailing list > syst...@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/systems > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Update docs
BTW, sorry it took so long. We need someone who will take care of these things when I'm on vacation or busy. If you or Simon would like to help co-maintain our DNS infrastructure, read the documentation (*) then ping me on irc to get started. (*) http://wiki.sugarlabs.org/go/Service/Nameservers On 05/21/13 11:21, Bernie Innocenti wrote: > On 05/20/13 08:07, Daniel Narvaez wrote: >> If you could make that >> >> developer.sugarlabs.org <http://developer.sugarlabs.org> -> >> http://sugarlabs.org/~buildbot/docs/index.md.html (notice the md) >> >> that would make my life a little easier. > > Done. > > What about doc.sugarlabs.org? Is it still current? Is someone still > maintaining it? > > >> On 20 May 2013 12:18, Simon Schampijer > <mailto:si...@schampijer.de>> wrote: >> >> On 05/20/2013 12:13 PM, Daniel Narvaez wrote: >> >> On 20 May 2013 12:07, Simon Schampijer > <mailto:si...@schampijer.de>> wrote: >> >> On 05/20/2013 10:40 AM, Daniel Narvaez wrote: >> >> I think the buildbot slave which uploads the docs hasn't >> yet had a >> successfull build since the doc was changed, so no >> updates yet. I'm on it, >> I should have probably made all these changes a bit more >> gradually. >> >> >> Great! >> >> Did the infra team handed out the new docs url already? >> >> >> Nope :/ >> >> >> Hey Bernie, >> >> did you get to reserve us the sugar-doc url already? >> >> Cheers, >>Simon >> >> >> >> >> -- >> Daniel Narvaez > > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Update docs
On 05/20/13 08:07, Daniel Narvaez wrote: > If you could make that > > developer.sugarlabs.org <http://developer.sugarlabs.org> -> > http://sugarlabs.org/~buildbot/docs/index.md.html (notice the md) > > that would make my life a little easier. Done. What about doc.sugarlabs.org? Is it still current? Is someone still maintaining it? > On 20 May 2013 12:18, Simon Schampijer <mailto:si...@schampijer.de>> wrote: > > On 05/20/2013 12:13 PM, Daniel Narvaez wrote: > > On 20 May 2013 12:07, Simon Schampijer <mailto:si...@schampijer.de>> wrote: > > On 05/20/2013 10:40 AM, Daniel Narvaez wrote: > > I think the buildbot slave which uploads the docs hasn't > yet had a > successfull build since the doc was changed, so no > updates yet. I'm on it, > I should have probably made all these changes a bit more > gradually. > > > Great! > > Did the infra team handed out the new docs url already? > > > Nope :/ > > > Hey Bernie, > > did you get to reserve us the sugar-doc url already? > > Cheers, >Simon > > > > > -- > Daniel Narvaez -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] developer.sugarlabs.org
I'm traveling, can someone please do it? If not, I'll try to get it done this week-end. On 05/10/13 14:03, Daniel Narvaez wrote: > Hello, > > can we please setup developer.sugarlabs.org > <http://developer.sugarlabs.org> to point to > > http://sugarlabs.org/~buildbot/docs/ > > That's just static html, generated from sugar-docs. > > -- > Daniel Narvaez -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] ASLO is really slow
On 05/03/2013 02:34 PM, Gonzalo Odiard wrote: > Hi, > I don't know if all people suffer the same problem, > but for me ASLO is really, really slow. > I have tried with Firefox and with Chrome, and always while wait, > show "Waiting for www.google-analytics.com" I heard that there was some sort of global analytics outage. > Is googl analytics slowing the site? Is really needed? I normally use awstats to analyze our logs, so it should be safe to remove. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Local Labs Wikis was: Re: [IAEP] wiki.sugarlabs.org ongoing maintenance
On 04/21/13 10:40, Ignacio Rodríguez wrote: > "I'm leaning to think the best option is to copy the pages that are good > from all {ar, co, cl, pe} wikis > into one single Sugar Labs / Laboratorios Azúcar América Latina instance. > " > +1 for the idea. > Maybe integrate Uruguay in SL pages {uy.sl.org <http://uy.sl.org>?} ! +1 from me, if the coordinators of the Local Labs agree to merge. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [IAEP] wiki.sugarlabs.org ongoing maintenance
Cristian Paul detected a surge of spam on co.sugarlabs.org... can someone please check? I'll be on a bus to NYC this afternoon. On 04/18/2013 02:27 AM, Sebastian Silva wrote: > > El 16/04/13 00:03, Bernie Innocenti escribió: >> Our wiki has grown quite large and the upgrade from Mediawiki 1.19 to >> 1.20 required several slow steps, so unfortunately I had to leave before >> fixing up the local wikis. >> > > I ran maintenance/update scripts for: > pe.sugarlabs.org > cl.sugarlabs.org > ar.sugarlabs.org > co.sugarlabs.org > > They are back up, spammy as they are. > > Regards, > Sebastian -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] wiki.sugarlabs.org ongoing maintenance
Today a serious security hole was announced for MediaWiki, so I started a long overdue upgrade of our main wiki. Our wiki has grown quite large and the upgrade from Mediawiki 1.19 to 1.20 required several slow steps, so unfortunately I had to leave before fixing up the local wikis. Meanwhile, please test login and editing and report any problems. Tomorrow I'll try to find some time to cleanup any remaining fallout. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Editing sources in GetBooks
Hello everyone, I'm in SF with Sameer and Alex and the question came up about adding an interface to add and edit OPDS feeds to GetBooks. Specifically, they have a Schoolserver running Pathagar, an OPDS-compliant library. An alternative design would be to discover this and automatically add a Schoolserver feed when present. The path would look like: http://schoolserver/by-title.atom?q= What do you guys think? -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Git is down
On Tue, 2013-01-08 at 13:26 +, Aleksey Lim wrote: > On Tue, Jan 08, 2013 at 01:15:31PM +, Aleksey Lim wrote: > > On Tue, Jan 08, 2013 at 11:12:56AM +0100, Daniel Narvaez wrote: > > > I'll see if I can figure out what is going on. > > > > Git daemon works well all this time. The last restart was 2012-12-13. > > Though, I just found that git daemon discarded some connections due to > concurent connections limit. I've increased it from 30 to 128. You probably got it! But do we have enough memory on jita to handle 128 concurrent connections without swapping? Anyway... it would be great if you had time to rebuild jita from scratch on a new VM. Now we do have the hosting capacity to do it, but migrating gitorious is a lot of work. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Git is down
I can't make sense of it. Maybe an ipv6 problem? On Sat, 2013-01-05 at 01:37 +0100, Daniel Narvaez wrote: > And... Now it's back... > > On Saturday, 5 January 2013, Daniel Narvaez wrote: > Now git seems to be down for good. git:// only, http and ssh > still works. > > -- > Daniel Narvaez > > > > -- > Daniel Narvaez > -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Systems] Frequent git errors
On Fri, 2013-01-04 at 02:14 +0100, Daniel Narvaez wrote: > Hello, > I'm getting this error from git.sugarlabs.org pretty frequently today: > fatal: read error: Connection reset by peer > This is with the git protocol, on two different machines (buildbot > master and slave). Next time you see it, please annotate the time and your IP and we'll try to trace the error to our logs. Also, is this over ssh:// or git://? In either case, could you please telnet to the port and see if it responds? -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] UNPLANNED OUTAGE: git.sugarlabs.org and several other services
All the Sugar Labs machines hosted at the Media Lab are currently unreachable and most likely powered off due to a black out affecting several buildings in south Cambridge. As soon as the power comes back I'll check the filesystems for data corruption before bringing everything back up. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH Speak] Use radians instead of degrees
On Mon, 2012-10-29 at 13:28 -0300, Manuel Kaufmann wrote: > Cairo.Context.arc needs the angle in radians instead of degrees. Give > it as 360 (radians) was taking up to 8 seconds to draw the eyes. Ouch, thanks for fixing this! -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Sugar Labs service outage: Thu, Sep 20 9:30-12:30 EDT
We're back in business. Let me know if anything didn't come back online. On Thu, 2012-09-20 at 00:26 -0400, Bernie Innocenti wrote: > Tomorrow, Thursday 20 Sep 2012, between 9:30 and 12:30 eastern time, the > Media Lab sysadmins will reconfigure the rack in room E15-243. > > During the maintenance work, the following services hosted on > treehouse.sugarlabs.org may become temporarily unavailable: > > - git.sugarlabsa.org and all related services > - chat.sugarlabs.org > - jabber.sugarlabs.org > - meeting.sugarlabs.org > - network.sugarlabs.org > - obs.sugarlabs.org > - rt.sugarlabs.org > - schooltool.sugarlabs.org > - ns1.sugarlabs.org (primary nameserver for multiple domains) > - Various services related to ole.org > - Various services related to paraguayeduca.org > - Various services related to treehouse.su > - Others I might have missed > > We'll use this opportunity to rack our two new servers and prepare them > for production. > -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Sugar Labs service outage: Thu, Sep 20 9:30-12:30 EDT
Tomorrow, Thursday 20 Sep 2012, between 9:30 and 12:30 eastern time, the Media Lab sysadmins will reconfigure the rack in room E15-243. During the maintenance work, the following services hosted on treehouse.sugarlabs.org may become temporarily unavailable: - git.sugarlabsa.org and all related services - chat.sugarlabs.org - jabber.sugarlabs.org - meeting.sugarlabs.org - network.sugarlabs.org - obs.sugarlabs.org - rt.sugarlabs.org - schooltool.sugarlabs.org - ns1.sugarlabs.org (primary nameserver for multiple domains) - Various services related to ole.org - Various services related to paraguayeduca.org - Various services related to treehouse.su - Others I might have missed We'll use this opportunity to rack our two new servers and prepare them for production. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Maintenance, reviews, QA & co
p8) > L. Leveraging the community of upstream components > > > == Proposal == > > All of the means listed above have different trade-offs, there's no > silver bullet. Review bandwidth of senior developers is limited, manual > tests take so much effort that only a subset can be tested regularly and > automated tests needs to be maintained alongside the code base. We need > a good balance to arrive at a great product. > > My current proposal would be: > > I. Do private (B) or public (C) review amongst colleagues to address > goal 1 (few obvious mistakes). > II. Do public in-depth reviews by senior developers (D), but make them > optional (i.e. _not_ G) and non-blocking. Addresses goals 3 (few > hard to fix bugs), 4 (maintainability) and 5 (better developers). > III. Require public short reviews by some senior developer > (E). Addresses goals 1 and to a limited amount 2 (few bugs > affecting the user), 3 and 4. > IV. Manual UI/UX (H) testing to the extent feasible. Addresses goal 2. > V. Implement automated UI and system tests (J). Strongly encourage > contributors to run them (successfully) before posting a patch > upstream. Where necessary contributors should extend the test > suite. Routinely run the tests against mainline across a large > number of supported systems (different distros, different hardware > including VMs, XO-* and Olidata JumPC). Addresses goals 1, 2 and to > some degree 3. > VI. Accept patches into mainline that are likely to increase the number > of contributors using Sugar themselves (A) or to increase their > usage of Sugar, even if the patch doesn't directly benefit the XO > target user base. It should not have a negative impact on the XO > target user base, of course. This addresses goals 2, 3, 4 and 5. > VII. Work on making Sugar more modular, using upstream components and > standard protocols or APIs as much as possible (L), allowing users > to mix and match components or simply configure them differently > (A) and reducing the footprint of bugs. This addresses goals 2, 3, > 4 and 5. > > > AFAICT I. is already being done. With Manuel Quiñones' appointment as > Glucose maintainer and his subsequent actions, we're evidently also > doing III. now. OLPC-AU and OLPC are doing IV., no change there > either. I've been working on implementing automated tests that can be > used for V., both on system tests (sugar-datastore test suite) as well > as UI tests (Gnome a11y based, not ready yet). Similarly, I've been > working on moving to standard protocols (e.g. XDG desktop entries for > Activities, ICCCM compliance) and upstream components > (e.g. gnome-session). > > The two major changes are making in-depth reviews by senior developers > optional and non-blocking (II.) as well as accepting "no ceiling" > (VI.) and standards compliance (VII.) patches. > > Sascha > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/sugar-devel -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Introducing sugar-build
On Fri, 2012-06-15 at 14:16 +0200, Daniel Narvaez wrote: > * We are still in the middle of the gobject-introspection/gtk3 > transition. Adding another bunch of invasive changes to the source > repository would likely slow that work down and introduce more > instability. We don't have enough resources to do both at the same > time. Agreed, this is probably not the best time. Still, soon or later we'll have to agree that this is the direction we want to take and make a coordinated effort to make it happen. > * The reduction of the number of modules is happening somewhat > naturally. sugar-base and sugar-presence-service are deprecated and at > some point they will disappear. Merging gradually, while also > refactoring where necessary, is likely to produce an higher quality > outcome than just taking the code as is and dumping it all in a single > repository. Good point, but i feel that someone needs to take ownership of completing the removal of sugar-presence-service and merging sugar-base into sugar. I've observed GCC being engulfed for years by incomplete transitions which leave behind lots of deprecated code that can't be removed because some obscure target still depends on it. > * The port of the shell to gobject-introspection will require > developers to build master of a few GNOME repositories. Even worst, > the Sugar touch work will require to build gtk+ code which has yet to > be written Ouch. I hope any weird dependencies we have to add for touch support will be at least optional. > I also have a longer term concerns about the sugar-core approach. > > * Are we actually going to get to a point where every external > dependency we need for a certain development cycle will be available > in a packaged form for the start of that cycle? I think the only > realistic way to always ensure that's the case is to do developement > that requires bleeding edge dependencies in separate branches and only > merge them as soon those dependencies are packaged in the latest > distributions. Though I think this kind of branching willl be too > costy until the number of people hacking core sugar modules grow > considerably. So we have a bit of a chicken and egg problem here. True. I just hope that we continue to work closely with distributions to get any new dependencies packaged quickly. At the same time, we should resist the temptation of adding new dependencies to esoteric things that nobody else uses. > Anyway I consider sugar-build a pragmatic compromise between > sugar-core and sugar-jhbuild. It's certainly not a wrapper around > jhbuild. Think of jhbuild more like a tool that we use internally to > pull and build modules (it can do that work fine and it's less code to > write and maintain). But we already forked jhbuild and wrapped it with something called sjhbuild. The result isn't particularly pretty, but perhaps you could fork sugar-jhbuild and implement your ideas directly there. What I'm trying to avoid here is a solution that adds yet another layer of indirection between the developer and the code being built. > [...] I've omitted the rest because I basically agree. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Introducing sugar-build
On Thu, 2012-06-14 at 12:06 +0200, Daniel Narvaez wrote: > * I think these tools are used by few people because they don't work > well. I know despite having a lot of experience with GNOME and linux > builds I was highly frustrated by sugar-jhbuild complexity and > unreliability. I'm sure a lot of people have been in that situation > and gave up. This is not to pick on sugar-jhbuild, as I said in > rationale it's a very difficult issue to solve. Mine is another > try.and I agree with who said competition is good here. This is also my experience, and I saw several developers at deployments who were struggling to build Sugar from source in jhbuild. I used to believe jhbuild could be improved in incremental steps, by shaving off dependencies and removing complexity here and there, but after several years it's still very fragile. The hardest issue seems to be finding consensus among the developers on one of the solutions. My preference would be towards merging all the glucose modules into a single repository that could be built with the familiar sequence of commands "configure && make && make install". This is basically what sugar-core [1] already did 2 years ago. It was created by Marco Pesenti Gritti based upon an earlier proposal of Michael Stone [2]. Both of them had had enough of fighting jhbuild and wanted something simple and clean by merging the repositories. Michael's proposal was more radical because it also replaced the autoconf build system with a hand-made configure script and a hand-made Makefile, both very short. I'm not sure I'd go that far, but it was a tempting idea. [1] http://sugar.marcopg.org/intro.html#build-from-source-code [2] http://dev.laptop.org/git/users/mstone/sugar/ > * Maintenance of these tools has an high cost but losing talented > contributors because they are unable to get the thing to build is a > much much higher one. Also, many of the design choices in sugar-build > intends to reduce maintenance cost (which I think in turn will favor > reliability). I couldn't agree more on your motives, but I think that jhbuild simply deserves to die. Wrapping it with something easier to use can be a temporary solution, but it can't eliminate all the underlaying complexity of repeated network operations to fetch code from disparate repositories. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] First impressions of a new XO user
On Thu, 2012-05-24 at 08:51 -0400, Walter Bender wrote: > Ouch. Well, we knew that 12.1 was not going to be recommended for deployments > :) > > You may want to help Diego load something more stable now that this > experiment has been conducted. I sent Diego a link [1] to the latest OLPC release for the XO-1, In case he gets tired of dogfooding development images for us. Though I hope he doesn't give up so easily: parents who are also Linux engineers make very valuable testers. [1] http://wiki.laptop.org/go/Release_notes/11.3.0#XO-1 -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] First impressions of a new XO user
Hello, a coworker got an old XO-1 and asked me to refurbish it for his kids. So I asked him to test OLPC-OS 12.1.0 build 11 and report any problems he ran into. New users can offer a fresh perspective on things that we give for granted. Some of the items below could be turned into trac tickets or wishlist items (provide a way to kill activities, provide a way to wipe the user profile, make more activities work in tablet mode...). If you need more information from the reporter, feel free to contact him directly at diegol...@google.com. Forwarded Message From: Diego Puppin To: Bernie Innocenti Subject: primo bug report Date: Wed, 23 May 2012 15:24:02 -0400 My older kid (7yr) and I had a lot of fun last with the OLPC. We played maze, we took pictures and more. The younger kid (15mo) seemed confused by the lack of a touch screen :-) - I can't find a way to connect to a network with hidden SSID. I tried to add my network using Gnome, but I couldn't find a "Connect" button there. Back to Sugar, the hidden network is still hidden. - The audio/video system seems unreliable. "Record" takes a long time to start. When I try to record (audio or video) the success rate is around 30% - the activity gets stuck. I can leave the activity but it keeps running in the background (green leds for camera and mike still on). How can I kill the activity entirely? - I found a recording in the Journal, but the Play activity failed with something as "resource busy". - While playing maze, with some gesture I changed the borders from the activity-specific border/menus to a different view. The only way out was the "square" button on top right. What's the gesture in and out of the activity-specific border/menus? - Only maze and collapse seem to work in game mode (rotated screen, use of the joypad). The other activities all requires a mouse. That's too bad. - Is there a way to upload an Epub and use the computer as an e-reader? Is there a mode where the backlight is off (I heard rumors like that)? - Reliability seems an issue. For a reason or another, I ended up rebooting every few minutes when I was switching among apps. When I stuck to play maze, everything was smooth and my kid did a million runs through the labyrinth. Just out of curiosity, did you install a beta build, or is it supposed to be stable? I don't mind "dogfooding", but I would be worried if this is the build going out to the public. - Is there a way to wipe it and restore to factory? You mentioned something like that. Diego -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Sugar Labs maintenance outage: TODAY 16:00 EDT (20:00 UTC)
We're replacing a dead drive in housetree, a kvm host running the following services: git.sugarlabs.org cgit.sugarlabs.org chat.sugarlabs.org jabber.sugarlabs.org meeting.sugarlabs.org obs.sugarlabs.org sweets.sugarlabs.org translate.sugarlabs.org ns1.sugarlabs.org rt.sugarlabs.org ...plus a number of services related to OLE, Paraguay Educa and other projects. We hope to complete this work in less than 1h if there are no complications. Apologies for the inconvenience. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Dextrose] activities on SD card, XO 1.0
[cc += sugar-devel] On Mon, 2012-05-14 at 12:47 +0200, Dominik Granada wrote: > > Our problem is that this feature does not work with Dex2. SD card is > not being mounted on startup. Once the system is on it is enough to > take it out/in and it starts to work. Yet for kids at school we would > need that automatic at start up. > > I would appreciate any hints... If all you need the SD for is adding extra activities, there might be an easier way: modify Sugar to add an extra lookup path for activities. The system directory for activities is hard-coded by configure into the variable "activities_path" of src/jarabe/config.py. In the rpm package of Sugar, this ends up in /usr/lib/python2.7/site-packages/jarabe/config.py If you're feeling a little braver, you could change bundleregistry.py to support a colon-separated list of paths: --- a/src/jarabe/model/bundleregistry.py +++ b/src/jarabe/model/bundleregistry.py @@ -61,7 +61,7 @@ class BundleRegistry(gobject.GObject): self._gio_monitors = [] user_path = env.get_user_activities_path() -for activity_dir in [user_path, config.activities_path]: +for activity_dir in [user_path, config.activities_path.split(':')]: self._scan_directory(activity_dir) directory = gio.File(activity_dir) monitor = directory.monitor_directory() Proper support for activities on removable media would involve finding the callback where Sugar handles hotplug events for storage media and, if the filesystem contains a directory called Activities, add it dynamically to BundleRegistry. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Wiki login
On Wed, 2012-04-25 at 12:47 -0400, Walter Bender wrote: > Do you think that something you did is now preventing downloads from > the wiki? I cannot seem access .gz files, e.g., [1] > > [1] http://wiki.sugarlabs.org/go/File:Food-plugin.tar.gz > > ??? Grrr... Somehow the TimedMediaHandler extension interferes with tar.gz files. I'll try to figure out why as soon as I find some time. For the time being, I've disabled TimedMediaHandler in the production wiki. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Wiki login
On Tue, 2012-04-24 at 10:20 -0700, Bernie Innocenti wrote: > On Mon, 2012-04-23 at 22:47 -0400, Frederick Grose wrote: > > > Sorry for offering false hope. It seems that there are problems with > > all the available methods of new account creation, as well as with the > > OpenID authentication. > > > > With a new account name and selecting the sugarlabs domain option, the > > 'Login error, incorrect password..' message appears. > > > > Selecting the local domain option instead just results in a blank > > screen. > > > > We need Bernie's help now. > > Works for my account. Which username and password are you using with the > sugarlabs domain? It should be your ldap password, the one that is > associated with your shell account on Sunjammer. > > As for OpenID, I still couldn't find the time to test. Sorry about this, > I don't even have the time to read my personal email lately :-( Ok, I've finally fixed it. Apologies for taking so much time and thanks for working around the issue by re-enabling user creation in the wiki. TL;DR: if you don't care about the technical details, you can stop reading now. There were two separate issues: - We still had php-openid 2.1.3 installed. I upgraded it to 2.2.2 made the OpenID authentication get through - The OpenID extension now honors the MediaWiki setting for disallowing the creation of new user accounts and gives only the option to bind the OpenID identity with an existing account. This is generally a good policy, but not for our particular case. Lacking a cleaner solution, I simply commented out the check in the code. I tested two OpenID providers: SourceForge and Google. Both worked fine, while my personal id.sugarlabs.org is rejected because the SSL certificate is expired. We should fix that. Luke, could you please take care of this? -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Wiki login
On Mon, 2012-04-23 at 22:47 -0400, Frederick Grose wrote: > Sorry for offering false hope. It seems that there are problems with > all the available methods of new account creation, as well as with the > OpenID authentication. > > With a new account name and selecting the sugarlabs domain option, the > 'Login error, incorrect password..' message appears. > > Selecting the local domain option instead just results in a blank > screen. > > We need Bernie's help now. Works for my account. Which username and password are you using with the sugarlabs domain? It should be your ldap password, the one that is associated with your shell account on Sunjammer. As for OpenID, I still couldn't find the time to test. Sorry about this, I don't even have the time to read my personal email lately :-( -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] doc sprint data save
On Tue, 2012-04-10 at 10:24 +0200, Sascha Silbe wrote: > Excerpts from Bernie Innocenti's message of 2012-04-09 21:36:33 +0200: > > On Mon, 2012-04-09 at 12:06 -0400, Paul Fox wrote: > > > here's the tarball of /home/olpc/sugar/default that i took > > > from the laptop adam handed me this morning: > > > > > > [URL STRIPPED] > > > > > > it's about 100MB. > > Downloaded. Is this confidential? I.e. may I send it to other developers > and / or publish it (after stripping out the owner key)? Mark Battley provided the data, so I guess it's up to him to decide. > > I think we'd better give this corpus to someone who worked on the > > Datastore recently and tell them that it was made with Sugar 0.94 on a > > hard-drive install of SoaS Pineapple (Fedora 16). > > OK, that's some background information, thanks. But what's the actual > problem? Start Sugar with this datastore and you'll see a few corrupted objects in the Journal. Perhaps Mark can describe what he did to get them (if anything). > We should continue the discussion on sugar-devel, BTW. Just strip out > the URL above to make sure it doesn't get published in its current form > (i.e. including owner keys). Ok, adding sugar-devel@ to the cc list. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Dextrose] Local server equivalent of "jabber.sugarlabs.org"
On Thu, 2012-03-29 at 20:17 +0530, Ajay Garg wrote: > Thanks Jerry ... :-) We also have an ejaberd instance running on sunjammer to provide @sugarlabs.org jabber accounts (not for Sugar collaboration). Let me know what your experience with Prosody is, if it's substantially better than ejabberd I'd like to switch. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Imminent maintenance downtime for sunjammer.sugarlabs.org
On Tue, 2012-02-28 at 10:15 -0500, Bernie Innocenti wrote: > As anticipated, today the systems administrators of the Free Software > Foundations are moving the bulk of their infrastructure from GNAPS to a > new colo. Sorry for the late notice: we've been back in business since about 6pm EST. The new IP for sunjammer.sugarlabs.org is 208.118.235.53, the IPv6 is 2002:d076:eb35::1. All the zones under our control have already been updated. Thanks to the FSF sysadmins for the smooth transition (so far, at least). There may be another, short outage within a few days to move the dom0 to its final rack slot. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Imminent maintenance downtime for sunjammer.sugarlabs.org
As anticipated, today the systems administrators of the Free Software Foundations are moving the bulk of their infrastructure from GNAPS to a new colo. Our core server sunjammer will be down for a few hours, affecting the following services: activities.sugarlabs.org activities-devel.sugarlabs.org activities-testing.sugarlabs.org activities-lb0.sugarlabs.org actividades.sugarlabs.org api.sugarlabs.org bugs.sugarlabs.org bugs-testing.sugarlabs.org bugs-devel.sugarlabs.org buildbot.sugarlabs.org cal.sugarlabs.org dev.sugarlabs.org doc.sugarlabs.org download.sugarlabs.org download-testing.sugarlabs.org ftp.sugarlabs.org groups.sugarlabs.org ldap.sugarlabs.org id.sugarlabs.org imap.sugarlabs.org join.sugarlabs.org karma.sugarlabs.org karma-devel.sugarlabs.org karma-testing.sugarlabs.org logcollect.sugarlabs.org mail.sugarlabs.org munin.sugarlabs.org mirrors.sugarlabs.org patchwork.sugarlabs.org patchwork-devel.sugarlabs.org people.sugarlabs.org pydocweb.sugarlabs.org planet.sugarlabs.org planet-testing.sugarlabs.org planet-devel.sugarlabs.org rsync.sugarlabs.org secure.sugarlabs.org services.sugarlabs.org static.sugarlabs.org shell.sugarlabs.org smtp.sugarlabs.org ssl-test.sugarlabs.org stats.sugarlabs.org trac.sugarlabs.org upload.sugarlabs.org vueltaciclista.sugarlabs.org webmail.sugarlabs.org wiki.sugarlabs.org wiki.ipv4.sugarlabs.org wiki.ipv6.sugarlabs.org wiki-devel.sugarlabs.org wiki-testing.sugarlabs.org www.sugarlabs.org www.ipv4.sugarlabs.org www.ipv6.sugarlabs.org www-testing.sugarlabs.org www-devel.sugarlabs.org lists.sugarlabs.org status.sugarlabs.org _jabber._tcp.sugarlabs.org co.sugarlab.org dc.sugarlab.org cl.sugarlabs.org pe.sugarlabs.org planet.py.sugarlabs.org ar.sugarlabs.org Afterwards, sunjammer will switch to a new IP in the 208.118.235.0/24 range (to be defined) and a corresponding 6to4 address. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Dextrose] [PATCH sugar] sl#2890: Changing to more user-friendly message.
On Sun, 2012-02-19 at 13:09 +0530, Ajay Garg wrote: > Earlier, the "Regsitration failed" message seemed more like a threat. > The new message sounds more like an information-conveying one. > > The new message is :: > > Primary (title) : > -- > No School Server found on the network Sounds much more helpful than the prevuos message. > Secondary (details) : > -- > Please ensure that you are connected to the correct network and that the > server is available. Also useful, but I don't see this in the code below. > src/jarabe/desktop/favoritesview.py |2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/src/jarabe/desktop/favoritesview.py > b/src/jarabe/desktop/favoritesview.py > index 81e32c8..e9c87ea 100644 > --- a/src/jarabe/desktop/favoritesview.py > +++ b/src/jarabe/desktop/favoritesview.py > @@ -337,7 +337,7 @@ class FavoritesView(hippo.Canvas): > try: > schoolserver.register_laptop() > except RegisterError, e: > -alert.props.title = _('Registration Failed') > +alert.props.title = _('No School Server found on the network') > alert.props.msg = _('%s') % e > else: > alert.props.title = _('Registration Successful') -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Mystery reboots
Today at 10:57 EST and 11:01, our only remaining kvm host housetree rebooted for no apparent reason, causing a short outage of several services. We are currently investigating the cause. Houstree had over 300 days of uptime, but since as of this week the load is considerably higher due to the VMs we moved from Treehouse. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Migrating to the Media Lab
On Fri, 2012-01-13 at 19:34 -0500, Bernie Innocenti wrote: > == Phase 1: treehouse -> housetree == > > To minimize downtime, we'll temporarily migrate all our VMs to > housetree, a server which is already racked in E15. > > There's question is whether housetree can take the load of 13 > additional virtual machines: > > Id Name State > -- > 54 aslo-web running > 130 zatoichi running > 180 schooltool running > 185 anno running > 227 bookirunning > 228 subuntu running > 233 lightwaverunning > 234 rt running > 240 pootle running > 243 ole running > 247 identity running > 248 idea running > 249 monitoring running Most VMs have already been migrated to housetree with no downtime for user-visible services. The migration of rt, pootle, lightwave and zatoichi is starting now. Affected services include: translate.sugarlabs.org rt.sugarlabs.org keys.sugarlabs.org ns1.sugarlabs.org *.paraguayeduca.org The downtime is projected to last about 2 hours (longer in case of unexpected issues). -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [IAEP] Sugar Labs Website Revamp IRC Mtg 1-8-12 11:00amEST(16:00UTC)
Apologies for the very late response, I fell behind on email processing for a while. On Sun, 2012-01-15 at 09:58 -0500, Steve Thomas wrote: > So one form of content would be project kids/teachers create using > various activities (Turtle Art, Etoys, etc.) TurtleArt's home page in the wiki has plenty of good examples to rip for the website. For the other activities, I collected several representative screenshots for use in talks. My slides are already published in the wiki, but I can't retrieve the link right now because I'm working offline. > I am working on a project hosting site for activities (similar to the > Scratch Website) where Sugar users can post projects they create. > > > I am using Google App Engine, but would also need some other > infrastructure (ex:WebDav server for storing/serving project files) We can do that on Sunjammer, our main server. If you want, the infrastructure team can host the entire application, but you'd have to switch to a free web framework such as LAMP, Rails, Drupal or (I hope not) Django. Another idea could be integrating with the existing ASLO (php) or perhaps rebasing ASLO on the current upstream Remora codebase (python). > Also, Andres Monroy-Hernandez had a good blog post on Supporting an > Online Community for Kids. I'll read it when I get online, thanks. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] Migrating to the Media Lab
Hello everyone, as anticipated long ago, the FSF is loosing its rack at GNAPS, so we'll be relocating sunjammer and treehouse to a new server room at the Media Lab. The absolute deadline to move out is Feb 29, but it would be safest to complete the migration sooner than that, possibly by mid February. Here's the current plan: == Phase 1: treehouse -> housetree == To minimize downtime, we'll temporarily migrate all our VMs to housetree, a server which is already racked in E15. There's question is whether housetree can take the load of 13 additional virtual machines: Id Name State -- 54 aslo-web running 130 zatoichi running 180 schooltool running 185 anno running 227 bookirunning 228 subuntu running 233 lightwaverunning 234 rt running 240 pootle running 243 ole running 247 identity running 248 idea running 249 monitoring running The most critical machines for Sugar Labs are lightwave, also-web and pootle. rt is not essential but it would be nice to keet it running. We can temporarily turn off subuntu and schooltool. I need to know from the PyEdu folks if zatoichi is still in production. All the others (anno, booki, ole, identity, idea and monitoring) belong to dogi. == Housetree preparations == Dogi and I have been working to optimize housetree. Last week, the load was peaking at over 10 with almost nothing running on it. We stopped a couple of unused VMs (openqwaq & template-squeeze) and solved a few issues with munin and jita. There are currently two VMs (ole2 and munin) which are causing an abnormally high load which I suspect might be caused by poor I/O performance to a fragmented qcow2 file. We'll migrate them to LVM partitions over the week-end. Housetree also has a broken drive. We've already bought a spare, but we decided to postpone the replacement until after the migration is over. == IPv4 and IPv6 == Currently we have only 7 IPv4 addresses assigned to housetree, which is not at all sufficient for all our VMs. Dogi noted that DHCP at the Media Lab gives long leases that remains stable for long periods of time, but in the long term we need a subnet with 32 or better 64 IPs dedicated to Sugar Labs, possibly with reverse DNS delegation. Dogi will ask to Michailis. As for IPv6, currently the Media Lab does not provide real IPv6 transit, but we can still use 6to4. == DNS adjustments == The TTL in our DNS is currently 3600 seconds. I'll lower it to 30 seconds before starting the migration. == Sunjammer == Sunjammer will move last, after treehouse is back online in the Media Lab. Currently, it's a XEN domU running on an FSF machine which is more or less as fast as treehouse. We can use the daily backups to speed up the filesystem migration. == activities.sugarlabs.org == An additional complication is that aslo-web requires a low-latency connection to sunjammer for NFS, which won't be available during the transition. Tonight I've removed aslo-web from the load balancer to see if sunjammer survives the extra load. I don't expect any trouble during the week-end, and I'll be on vacation on Monday so I can monitor the situation. == Physical access == We currently have no access to the machine room, so if something happens on a Friday night we're screwed for the whole week-end. Walter has an MIT pass and Michailis should be able to get him access to the machine room as well. It seems that we could also request a special card type called "DLC / Unofficial Members of the MIT Community" to some office in the Media Lab. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Sugar Labs Website Revamp IRC Mtg 1-8-12 11:00amEST(16:00UTC)
On Sun, 2012-01-08 at 11:21 -0800, Christian Marc Schmidt wrote: > Does this coming Saturday 14th at 11:00am EST/16:00 UTC work for > everyone? Or is a time during the week better? Suggestions welcome. Works for me. As for the calendar, for please use the Sugar Labs Meetings calendar: Calendar ID: h9cfuk10894em7a8moemquu...@group.calendar.google.com atom: https://www.google.com/calendar/feeds/h9cfuk10894em7a8moemquusmg%40group.calendar.google.com/public/basic ical: https://www.google.com/calendar/ical/h9cfuk10894em7a8moemquusmg%40group.calendar.google.com/public/basic.ics html: https://www.google.com/calendar/b/0/embed?src=h9cfuk10894em7a8moemquu...@group.calendar.google.com&ctz=America/New_York If you don't have write access, please give me a gmail / google account. BTW: I'm working to setup a calendar server within the Sugar Labs infrastructure, but it's not yet ready. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH sugar-toolkit-gtk3] Add MAINTAINERS file
On Sat, 2011-12-17 at 19:03 -0500, Bernie Innocenti wrote: > On Tue, 2011-12-13 at 19:55 +, Daniel Drake wrote: > > Just point at the wiki where the info is kept. > > Requested by Sascha Silbe. > > --- > > MAINTAINERS |1 + > > 1 files changed, 1 insertions(+), 0 deletions(-) > > create mode 100644 MAINTAINERS > > > > diff --git a/MAINTAINERS b/MAINTAINERS > > new file mode 100644 > > index 000..379f8ba > > --- /dev/null > > +++ b/MAINTAINERS > > @@ -0,0 +1 @@ > > +See http://wiki.sugarlabs.org/go/Development_Team/Release/Modules > > Great. Can we do the same for all modules? I mean: replacing the MAINTAINERS file with a simple link to the wiki page. Looking at the wiki page, I just noticed that sugar-toolkit-gtk3 is maintained by you and Simon, which makes it different from all the other modules. Presently it doesn't really matter, but when the rest of glucose switches to gtk3, this arrangement would bring us back to the time when getting any non-trivial change merged required seeking the combined approval of multiple module maintainers. So I'd propose that we continue to have glucose share the same set maintainers, either by making Silbe co-maintain sugar-toolkit-gtk3, or by adding you to the other glucose modules. I think it's important to keep our review process smooth and avoid frustrating our contributors with patches stuck in a half-approved state. -- _ // Bernie Innocenti \X/ http://codewiz.org ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [PATCH sugar-toolkit-gtk3] Add MAINTAINERS file
On Tue, 2011-12-13 at 19:55 +, Daniel Drake wrote: > Just point at the wiki where the info is kept. > Requested by Sascha Silbe. > --- > MAINTAINERS |1 + > 1 files changed, 1 insertions(+), 0 deletions(-) > create mode 100644 MAINTAINERS > > diff --git a/MAINTAINERS b/MAINTAINERS > new file mode 100644 > index 000..379f8ba > --- /dev/null > +++ b/MAINTAINERS > @@ -0,0 +1 @@ > +See http://wiki.sugarlabs.org/go/Development_Team/Release/Modules Great. Can we do the same for all modules? -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] FW: scratch on aslo
On Sat, 2011-12-17 at 00:11 -0500, Samuel Klein wrote: > I understand why Scratch isn't hosted on aslo at present, but it would > be nice if it were hosted in a fashion that supported discovery of and > review/discussion of it. Scratch is Sugar's most popular programming > activity, after all... A few months ago it sounded like the Media Lab was going to correct the binary license of Scratch. Has anyone heard back from them? Cc'ing Mako and Claudia who might know the right person to ask. -- Bernie Innocenti Sugar Labs Infrastructure Team http://wiki.sugarlabs.org/go/Infrastructure_Team ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel