Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On Wed, Jul 15, 2009 at 04:14, Martin Langhoffmartin.langh...@gmail.com wrote: On Wed, Jul 15, 2009 at 8:35 AM, Greg Smithgregsmit...@gmail.com wrote: As I got there I noticed a kid had unplugged their USB before full shut down and they were frantically trying to plug it back in while errors scrolled on the screen. Still not sure if they lost data (will check tomorrow on kid who made the same mistake last week). Is there anything we can do to ensure that all data is written/saved ASAP on clicking Shut down from the menu? This is a very valid concern, and very tricky. I've seen lots of problems with LiveUSB sticks and unclean shutdown. The short of it is that the scheme behind LiveUSB, specifically the persistent storage is very brittle, to the point that it can hose the entire disk badly. The USB stick will only be usable again after re-formatting and re-installing the SoaS software (losing user data, etc). It is a bad combination of - Using rw overlay mountpoints that only sync every 5s. If the overlay partition is corrupted, it will prevent the SoaS from booting. - Writing on a FAT partition which is notoriously fragile... and mounted async. Mounting it sync slows things tremendously and wears the USB stick out much faster. IIRC, recent kernels have a mount flag for vfat (and maybe other FSs?) that makes the kernel more eager to flush things to disk, so you can keep the mountpoint async but reduce the 5s window significantly. Unfortunately, I can't find it now. It was well reported on LWN and kernelnewbies a few months ago. I agree with the conclusion that Linux support for live usb sticks still has a long way to go. See https://bugzilla.redhat.com/show_bug.cgi?id=465725 for more details about the overlay issue, though note that in Fedora the whole stick is not thrashed, you can discard any changed you did to / by passing the kernel the option reset_overlay. If deployers of SoaS keep in mind this limitation and users mostly write inside /home, this limitation is much less important. This really has very little to do with Sugar itself, like hardware support. I think we need to find the resources to make the needed work in the kernel and in the distro level before we can do anything in SoaS itself. Regards, Tomeu cheers, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
To minimize this risk, for the first run of Sugar Labs branded sticks I had chosen an LED-equipped model, everybody including kids can figure out what to do / not do depending on light status. Unfortunately, SD Cards which represent a cheaper and more robust form factor in particular for netbooks don't have LEDs :-( Sean On Wed, Jul 15, 2009 at 4:14 AM, Martin Langhoffmartin.langh...@gmail.com wrote: On Wed, Jul 15, 2009 at 8:35 AM, Greg Smithgregsmit...@gmail.com wrote: As I got there I noticed a kid had unplugged their USB before full shut down and they were frantically trying to plug it back in while errors scrolled on the screen. Still not sure if they lost data (will check tomorrow on kid who made the same mistake last week). Is there anything we can do to ensure that all data is written/saved ASAP on clicking Shut down from the menu? This is a very valid concern, and very tricky. I've seen lots of problems with LiveUSB sticks and unclean shutdown. The short of it is that the scheme behind LiveUSB, specifically the persistent storage is very brittle, to the point that it can hose the entire disk badly. The USB stick will only be usable again after re-formatting and re-installing the SoaS software (losing user data, etc). It is a bad combination of - Using rw overlay mountpoints that only sync every 5s. If the overlay partition is corrupted, it will prevent the SoaS from booting. - Writing on a FAT partition which is notoriously fragile... and mounted async. Mounting it sync slows things tremendously and wears the USB stick out much faster. IIRC, recent kernels have a mount flag for vfat (and maybe other FSs?) that makes the kernel more eager to flush things to disk, so you can keep the mountpoint async but reduce the 5s window significantly. Unfortunately, I can't find it now. It was well reported on LWN and kernelnewbies a few months ago. cheers, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On Thu, Jul 16, 2009 at 3:30 AM, Sean DALYsdaly...@gmail.com wrote: To minimize this risk, for the first run of Sugar Labs branded sticks I had chosen an LED-equipped model, everybody including kids can figure out what to do / not do depending on light status. That's intuitive but... not quite true right now. It is reasonable _only if we can get the lazy sync mode going_. With async (which is what we have), the kernel may have dirty pages but won't write them immediately even if the disk is idle. You may have up to 5s of no LED activity, which can be very misleading. cheers, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
Analyze activity is on http://wiki.laptop.org/go/Activities direct download http://dev.laptop.org/~edsiper/bundles/Analyze-8.xo Dave On Tue, Jul 14, 2009 at 4:35 PM, Greg Smithgregsmit...@gmail.com wrote: Hi All, I'm sending this to the devel instead of the IAEP list as it contains a number of technical/UI design comments that I think need developer follow up. Feel free to forward to other lists, blogs etc. as appropriate. Notes: I joined the team (Anurag, Caroline, and Bill) as the class of 22 x 2nd graders was leaving. This was their third class with Sugar/Turtle Art. They do it twice a week. As I got there I noticed a kid had unplugged their USB before full shut down and they were frantically trying to plug it back in while errors scrolled on the screen. Still not sure if they lost data (will check tomorrow on kid who made the same mistake last week). Is there anything we can do to ensure that all data is written/saved ASAP on clicking Shut down from the menu? The team told me that they worked with Turtle Art to load a map of the Charles River and to add tags to it showing where the GPA school is located. They said that some kids got it and some did not. Here's the warm/cold post class coments: Caroline Warm: More than half the glass got the idea of negative numbers and the gird (aka cartesian coordinates for placing turtle). A few got the idea of putting together a program. Cold: There was a large variance in kids who go it and those who didn't. Those who didn't understand got frustrated and they lost participation of about 1/3 of the class as a result. Bill Warm: A group of kids zero'd in quickly on the ideas of positive and negative. Cold: One USB was bad and that kid lost the flow and didn't catch up. Neutral: One girl finished right away and then didn't know what to do. Caroline said a few of the kids really looked like they could be programmers. Anurag Warm: All kids seem to be retaining what they learned in previous classes. Like how to put the turtle in the corner with Set (x,y). Lots of improvement from last week as a result like using show text, color, size. Lots of kids understood that 0,0 is in the center and that helped them find positions. Cold: Some kids lost focus because they didn;t understanbd the required order of operation in Turtle art (more below). Some built things then built other unattached programs which therefore didn't work together (e.g. show on the map then an un atached show text program somewhere else on the palette). Not enough time in the class. Other comments in debrief: The reality of teaching is time constrained. Next class we could just do the work for the kids who didn't get the GPA school noted on the Charles River map but they may lose a learning chance that way. Another option is to slow the class down, let kids who were done do something else and give the kids still working on the map more time. Kids seemed to think more this time rather than just clicking around a lot. Next class is Thursday. Goal is to put their pictures on the web. Comments that some kids don't understand the Journal/Keep paradigm which was introduced this time. An example: Sometimes they were in Turtle art and wanted to re-load an image from a previous instance. Many kids used the Turtla Art show tile (my turtle art is rusty so exact naming may be off). This allows clicking on it to open a Journal picker then clicking on it again (executing it?) shows an image of the last Turtle Art screen as seend in the Journal screen shot. What they really wanted was to just re-load the map in to the new blank Turtle Art screen. The suggestion was to rename the Turtle Art activity, then click Keep (I believe this step is not intended by Eben et al), then click stop, then open Turtle Art again, load the Kept instance from the journal. That's what my notes say but maybe Anurag and/or Caroline can clarify. Another tough work flow was getting the map in Turtle Art, scaling it then placing the turtle. The right work flow in terms of order execution of tiles in Turtle art is: - Move turtle to right place on screen - Set scale - Show map Many kids did this: - Show map - Set x,y of turtle - scale map Again, I'm not sure why the kids method is wrong but believe its a nuance of Turtle Art. Clarification appreciated. Two notes about the computer lab class: - Its also the library! There weren't a huge number of books and they were behind book cases kind of pushed out of the way to make room for computers. That made me a little sad as I prefer spend time with hard copy books than any other inanimate object and quite a number of animate ones too. I saw a Dr Seuss book I have never read (The Cat's Quizzer) and to be honest I would rather my kids read Dr. Seuss than spend time on the computer any day! :-) Fortunately they can do both. - The wall had these instructions (algorithm?), each on a separate
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
As Greg said we finally got collaboration working! The problem was that we had cloned the owner.key and owner.key.pub. The fix was to delete those before cloning and on existing sticks delete them and restart. This gives us solid collaboration inside the computer lab if we go local (take out the network setting). It gives us the seen frequently before, not so solid jabber collaboration. Most people can see someone some of the time but not everyone see's everyone. We will try to work with the local collaboration tomorrow. Sugar will be very hard for schools to scale until we have an easier way for teachers to create sticks for an entire class and we have a ticket in for this but I don't know if anyone is working on it. Caroline On Tue, Jul 14, 2009 at 4:35 PM, Greg Smith gregsmit...@gmail.com wrote: Hi All, I'm sending this to the devel instead of the IAEP list as it contains a number of technical/UI design comments that I think need developer follow up. Feel free to forward to other lists, blogs etc. as appropriate. Notes: I joined the team (Anurag, Caroline, and Bill) as the class of 22 x 2nd graders was leaving. This was their third class with Sugar/Turtle Art. They do it twice a week. As I got there I noticed a kid had unplugged their USB before full shut down and they were frantically trying to plug it back in while errors scrolled on the screen. Still not sure if they lost data (will check tomorrow on kid who made the same mistake last week). Is there anything we can do to ensure that all data is written/saved ASAP on clicking Shut down from the menu? The team told me that they worked with Turtle Art to load a map of the Charles River and to add tags to it showing where the GPA school is located. They said that some kids got it and some did not. Here's the warm/cold post class coments: Caroline Warm: More than half the glass got the idea of negative numbers and the gird (aka cartesian coordinates for placing turtle). A few got the idea of putting together a program. Cold: There was a large variance in kids who go it and those who didn't. Those who didn't understand got frustrated and they lost participation of about 1/3 of the class as a result. Bill Warm: A group of kids zero'd in quickly on the ideas of positive and negative. Cold: One USB was bad and that kid lost the flow and didn't catch up. Neutral: One girl finished right away and then didn't know what to do. Caroline said a few of the kids really looked like they could be programmers. Anurag Warm: All kids seem to be retaining what they learned in previous classes. Like how to put the turtle in the corner with Set (x,y). Lots of improvement from last week as a result like using show text, color, size. Lots of kids understood that 0,0 is in the center and that helped them find positions. Cold: Some kids lost focus because they didn;t understanbd the required order of operation in Turtle art (more below). Some built things then built other unattached programs which therefore didn't work together (e.g. show on the map then an un atached show text program somewhere else on the palette). Not enough time in the class. Other comments in debrief: The reality of teaching is time constrained. Next class we could just do the work for the kids who didn't get the GPA school noted on the Charles River map but they may lose a learning chance that way. Another option is to slow the class down, let kids who were done do something else and give the kids still working on the map more time. Kids seemed to think more this time rather than just clicking around a lot. Next class is Thursday. Goal is to put their pictures on the web. Comments that some kids don't understand the Journal/Keep paradigm which was introduced this time. An example: Sometimes they were in Turtle art and wanted to re-load an image from a previous instance. Many kids used the Turtla Art show tile (my turtle art is rusty so exact naming may be off). This allows clicking on it to open a Journal picker then clicking on it again (executing it?) shows an image of the last Turtle Art screen as seend in the Journal screen shot. What they really wanted was to just re-load the map in to the new blank Turtle Art screen. The suggestion was to rename the Turtle Art activity, then click Keep (I believe this step is not intended by Eben et al), then click stop, then open Turtle Art again, load the Kept instance from the journal. That's what my notes say but maybe Anurag and/or Caroline can clarify. Another tough work flow was getting the map in Turtle Art, scaling it then placing the turtle. The right work flow in terms of order execution of tiles in Turtle art is: - Move turtle to right place on screen - Set scale - Show map Many kids did this: - Show map - Set x,y of turtle - scale map Again, I'm not sure why the kids method is wrong but believe its a nuance of Turtle Art. Clarification
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On Wed, Jul 15, 2009 at 10:19 AM, Caroline Meekssolutiongr...@gmail.com wrote: As Greg said we finally got collaboration working! The problem was that we had cloned the owner.key and owner.key.pub. The fix was to delete those before cloning and on existing sticks delete them and restart. Excellent. Great to hear that. cheers, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On 14 Jul 2009, at 21:35, Greg Smith wrote: I also spent 1/2 hour debugging collaboration. Dave and Gary helped on IRC. They pointed me a tool called Analyze (can someone put that on the main activities page ASAP? I didn't write down the URL and will need to fetch it again tomorrow). Done, the current version 8 is up at: http://activities.sugarlabs.org/addon/4200 Regards, --Gary P.S. I've cloned the Analyze git rep over to http://git.sugarlabs.org/projects/analyze , shout if you want commit access. ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
Hi Martin, On 15 Jul 2009, at 02:33, Martin Langhoff wrote: On Wed, Jul 15, 2009 at 10:19 AM, Caroline Meekssolutiongr...@gmail.com wrote: As Greg said we finally got collaboration working! The problem was that we had cloned the owner.key and owner.key.pub. The fix was to delete those before cloning and on existing sticks delete them and restart. Excellent. Great to hear that. What other system keys/ID are used for authentication with an XS? I'd hate to see collaboration solved and then for future server backup or Moodle support to get it's knickers in a twist! I have grave worries about the current 'clone a booted and hacked stick' without folks involved knowing all those first boot processes in detail – I just know enough not to want to try it in the first place :-) Regards, --Gary ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On 7/14/09, Gary C Martin g...@garycmartin.com wrote: Hi Martin, On 15 Jul 2009, at 02:33, Martin Langhoff wrote: On Wed, Jul 15, 2009 at 10:19 AM, Caroline Meekssolutiongr...@gmail.com wrote: As Greg said we finally got collaboration working! The problem was that we had cloned the owner.key and owner.key.pub. The fix was to delete those before cloning and on existing sticks delete them and restart. Excellent. Great to hear that. What other system keys/ID are used for authentication with an XS? I'd hate to see collaboration solved and then for future server backup or Moodle support to get it's knickers in a twist! I have grave worries about the current 'clone a booted and hacked stick' without folks involved knowing all those first boot processes in detail – I just know enough not to want to try it in the first place :-) The current method I'm using of hand crafting a stick, going in and deleting files by hand, then cloning it is a total hack, hard to reproduce and very subject to human error. We need something like Ticket 74 where we have a real clone function that does the right things automatically. If we had that, and especially if it worked with a $15 USB hub to let you clone 4 or 6 sticks at a time, I think it would be far easier to have teachers do pilots. I really think that all of our current stick creation methods are too inefficient, inflexbile and human error prone to have many teachers with 20 to 30 kids in a class use them. Regards, --Gary -- Caroline Meeks Solution Grove carol...@solutiongrove.com 617-500-3488 - Office 505-213-3268 - Fax ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] GPA School Visit report for Tuesday July 14
On Wed, Jul 15, 2009 at 8:35 AM, Greg Smithgregsmit...@gmail.com wrote: As I got there I noticed a kid had unplugged their USB before full shut down and they were frantically trying to plug it back in while errors scrolled on the screen. Still not sure if they lost data (will check tomorrow on kid who made the same mistake last week). Is there anything we can do to ensure that all data is written/saved ASAP on clicking Shut down from the menu? This is a very valid concern, and very tricky. I've seen lots of problems with LiveUSB sticks and unclean shutdown. The short of it is that the scheme behind LiveUSB, specifically the persistent storage is very brittle, to the point that it can hose the entire disk badly. The USB stick will only be usable again after re-formatting and re-installing the SoaS software (losing user data, etc). It is a bad combination of - Using rw overlay mountpoints that only sync every 5s. If the overlay partition is corrupted, it will prevent the SoaS from booting. - Writing on a FAT partition which is notoriously fragile... and mounted async. Mounting it sync slows things tremendously and wears the USB stick out much faster. IIRC, recent kernels have a mount flag for vfat (and maybe other FSs?) that makes the kernel more eager to flush things to disk, so you can keep the mountpoint async but reduce the 5s window significantly. Unfortunately, I can't find it now. It was well reported on LWN and kernelnewbies a few months ago. cheers, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
