Re: [olpc-nz] Testing summary - 12 December 2009 Wellington, NZ
Hi Munro, This isn't an ordinary laptop, and some of our deployment areas have very expensive external DC power; from solar panels, cows, or a grid that needs maintenance. When we make a million laptops suspend like that even though they are on external power, we save megawatts. An ordinary laptop passes that burden to the consumer. We're not really wanting to do that here. We have to think of the cost per child. -- James Cameron http://quozl.linux.org.au/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [olpc-nz] Testing summary - 12 December 2009 Wellington, NZ
On 14 Dec 2009, at 19:54, Alastair Munro wrote: > Hi Cameron, > > I checked a couple of the laptops after seeing your email about the > suspending, the wifi disappearing was after a suspend. I was thinking of > suspend in the terms of a normal laptop where suspending effectively turns > off the computer and saves state to the hard disk. I find it odd that the > suspend is still allowed to occur even when running off AC power though. I think the idea is that power is precious in the majority of deployments (if not all of us). Burning cycles on a diesel generator, wind turbine, or solar is detrimental to every other device plugged into that limited power source. Regards, --Gary > -Alastair > > On Mon, Dec 14, 2009 at 1:13 PM, James Cameron wrote: > G'day Alastair, > > Thanks for responding. I'd like to drill down further into the facts > just to make sure we're not missing anything. > > On Mon, Dec 14, 2009 at 09:19:19AM +1300, Alastair Munro wrote: > > We did not disable suspend before testing, all these observations were > > made on a fresh boot while plugged into AC power with no suspending. > > You did not disable suspend, yet there was no suspending at all? That's > surprising. The design since about build os49 is that dim screen with > begin at about 15 seconds and suspend will begin at about 25 seconds of > idle. The suspend doesn't turn off the screen, so you might not notice > it. The power LED begins to blink at about a five second rate. > > Can you confirm that you keep the keyboard and touchpad active, and > never idle for more than 30 seconds? It is very hard to do that, > especially with other people interrupting. > > > The wireless just didn't seem to activate at all. When checking > > 'ifconfig -a' the only device shown was the local loopback device > > 'lo'. > > This is quite consistent with not knowing that an automatic idle suspend > has happened. > > -- > James Cameron > http://quozl.linux.org.au/ > > ___ > Devel mailing list > Devel@lists.laptop.org > http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [olpc-nz] Testing summary - 12 December 2009 Wellington, NZ
Hi Cameron, I checked a couple of the laptops after seeing your email about the suspending, the wifi disappearing was after a suspend. I was thinking of suspend in the terms of a normal laptop where suspending effectively turns off the computer and saves state to the hard disk. I find it odd that the suspend is still allowed to occur even when running off AC power though. -Alastair On Mon, Dec 14, 2009 at 1:13 PM, James Cameron wrote: > G'day Alastair, > > Thanks for responding. I'd like to drill down further into the facts > just to make sure we're not missing anything. > > On Mon, Dec 14, 2009 at 09:19:19AM +1300, Alastair Munro wrote: > > We did not disable suspend before testing, all these observations were > > made on a fresh boot while plugged into AC power with no suspending. > > You did not disable suspend, yet there was no suspending at all? That's > surprising. The design since about build os49 is that dim screen with > begin at about 15 seconds and suspend will begin at about 25 seconds of > idle. The suspend doesn't turn off the screen, so you might not notice > it. The power LED begins to blink at about a five second rate. > > Can you confirm that you keep the keyboard and touchpad active, and > never idle for more than 30 seconds? It is very hard to do that, > especially with other people interrupting. > > > The wireless just didn't seem to activate at all. When checking > > 'ifconfig -a' the only device shown was the local loopback device > > 'lo'. > > This is quite consistent with not knowing that an automatic idle suspend > has happened. > > -- > James Cameron > http://quozl.linux.org.au/ > ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [olpc-nz] Testing summary - 12 December 2009 Wellington, NZ
G'day Alastair, Thanks for responding. I'd like to drill down further into the facts just to make sure we're not missing anything. On Mon, Dec 14, 2009 at 09:19:19AM +1300, Alastair Munro wrote: > We did not disable suspend before testing, all these observations were > made on a fresh boot while plugged into AC power with no suspending. You did not disable suspend, yet there was no suspending at all? That's surprising. The design since about build os49 is that dim screen with begin at about 15 seconds and suspend will begin at about 25 seconds of idle. The suspend doesn't turn off the screen, so you might not notice it. The power LED begins to blink at about a five second rate. Can you confirm that you keep the keyboard and touchpad active, and never idle for more than 30 seconds? It is very hard to do that, especially with other people interrupting. > The wireless just didn't seem to activate at all. When checking > 'ifconfig -a' the only device shown was the local loopback device > 'lo'. This is quite consistent with not knowing that an automatic idle suspend has happened. -- James Cameron http://quozl.linux.org.au/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [olpc-nz] Testing summary - 12 December 2009 Wellington, NZ
We did not disable suspend before testing, all these observations were made on a fresh boot while plugged into AC power with no suspending. The wireless just didn't seem to activate at all. When checking 'ifconfig -a' the only device shown was the local loopback device 'lo'. -Alastair On Sat, Dec 12, 2009 at 2:11 PM, James Cameron wrote: > On Sat, Dec 12, 2009 at 01:40:04PM +1300, Tabitha Roder wrote: > > On all XO 1.5s we updated to OS54 wifi does not work (except on 1 of 5 > > machines showed wireless connections for about 20 seconds then gone) > > although we now realise there is OS56 so maybe we need to just update > > again. > > You probably hit the known problem described in the release candidate > announcement "If you have a B2 without hardware mods, you must disable > suspend using Control Panel->Power->Automatic power management, else > your wireless device will disconnect on suspend and not come back until > the next reboot." > > Can you confirm that you did not disable suspend before testing > wireless? > > -- > James Cameron > http://quozl.linux.org.au/ > ___ > olpc-nz mailing list > olpc...@lists.laptop.org > http://lists.laptop.org/listinfo/olpc-nz > ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel