[00:02] I sure hope the defaults are good cause I still can't read this console :) [00:06] Serial console? [00:06] yeah [00:07] I wonder if you have a bad connection or possibly a bad port on the panda. [00:08] it IS going to a usb hub first [00:08] Shouldn't matter. I have 8 usb serial cables on a hub for all my systems. [00:09] And my serial port monitoring system is an older Atom based system in a mini itx case that I remote to. [00:09] is the primary network interface during install the wireless or the lan? [00:11] On server, lan. [00:12] I don't think we have all the packages installed for wifi on the server image. Will check. [00:12] I need the lan one I just couldn't read it [00:12] top default option. [00:13] GrueMaster: The server image will do wifi in the clear or with WEP, but I suspect WPA will fail. Haven't tested. [00:14] (I always used to have a WEP fallback just for d-i testing, but maybe it does WPA properly now) [00:14] will bad things happen if I reset in the middle of this install? [00:14] infinity: The firmware used to be missing from the image. Haven't tested it recently. [00:14] brandini: Is it at a prompt or doing something? [00:14] prompting for network configuration [00:15] You can restart. Main thing is to not reboot during a write process. [00:16] * infinity takes off for the evening. [00:23] anyone use a macbook to do the serial connection? [00:23] I've not been able to figure it out on the mac :) === Jack87|Away is now known as Jack87 === Jack87 is now known as Jack87|Away === Jack87|Away is now known as Jack87 === Jack87 is now known as Jack87|Away [00:32] Not me. never owned one. [00:33] you're not missing anything [00:33] ;) [00:46] * GrueMaster wanders off the grid for a while until the kubuntu images finish downloading. [01:12] sweet, I finally got the serial cable working on my mac and the display is much better :) === Jack87|Away is now known as Jack87 [01:13] are oem-config-firstboot errors trying to write to /var/log/oem-config.log because of a read-only filesystem normal? [01:28] No. You should not have a read-only filesystem [01:29] You may need to reflash and restart. [01:29] done and done [01:35] GrueMaster: nope, same thing happened... "Running post-installation trigger man-db" 100% /usr/sbin/oem-config-firstboot: line 58: /var/log/oem-config.log: Read-only file system [01:36] Hmm. I would suggest trying a smaller SD card. [01:36] ext4_mb_generate_buddy:736: group 160, 4045 blocks in bitmap, 4096 in gd [01:36] Could that be coming from errors=remount-ro ? [01:36] looks like it is [01:36] let me paste it all... [01:37] In that case get a card that isn't buggered [01:38] http://pastebin.com/S8e7CXGH [01:41] brandini: Need to see why journal aborted. Is there any messages before the first line? [01:41] not that I can see [01:41] Im' trying another card [01:48] Everything worked fine here with the latest image. But as I have not tested on anything bigger than 16G, I suspect your card (or the system may not like your card). [01:49] I've never had any luck with CF [01:56] CF? [01:57] These are SD. Not nearly as fast. [01:57] Class 20 SD cards will be mainstream in a few months [01:57] * GrueMaster uses CF cards in Sata adapters for rootfs on 2 servers and a firewall. [01:58] Oddly, I have found that the lower the class, the more reliable and better overall performance. [01:58] with the 16GB card it's been at Resizing, pass: 1 [100/100] for a few minutes... [01:58] The higher class are great for sustained sequential reads/writes (like video cameras). [01:59] brandini: any activity on the LEDs? [01:59] brandini, you need fsck the partition, bro... [01:59] the one closest to the SD card is solid and the other is blinken lots [02:00] the journal is in status (ABRT) [02:00] power off and fsck the card? [02:00] do sudo fsck /dev/mmcblk0p2 -c -y [02:00] yes [02:01] in another machine right? [02:01] yes [02:02] also, I would suggest not using a journalling file system on an SD card [02:02] EXT2 is best [02:03] I didn't format it before flashing this image [02:03] Actually, I have benchmarked and found ext4 to be very fast. But we are just now (like Monday) in the process of switching on the SD preinstalled images. [02:04] I didn't know I was supposed to format the preinstalled image? [02:04] Shouldn't matter. The dd process will overwrite the first 1.6G. Resize will clear the rest. [02:04] its is very fast, but is designed for magnetic media, not FLASH based media, such as the NOR flash in SD cards [02:04] ok, it fixed things [02:04] now lets hope it boots :) [02:04] the journal is constantly rewritten, even if just for a read, making it burn the Flash media in the journals blocks [02:05] Actually, ext4 is designed for SSD, which is based on the same tech as SD. [02:05] GrueMaster, you are mistaken [02:05] btrfs is supposed to be even better, but it isn't stable yet. [02:05] SSD's have wear leveling, SD cards have bit shifting for wear leveling (poor for repeated writes) [02:05] it's resizing again :) [02:06] the wear leveling in SSD's is much more robust than the wear leveling in SD cards [02:06] I can destroy an SD card in a matter of hours, with sparsewriting routines, from EXT4, or ReiserFS [02:07] 92/100 [02:07] EXT2 is much more suited to older lessstable media, with SD fits into [02:07] sorry, my spacebar is failing [02:07] too much pepsi ;) [02:08] brandini, you getting jitters? [02:08] * GrueMaster is being called to dinner. Biab. [02:08] I don't think this is going to work [02:09] brandini, did you do what I asked you to? [02:09] yup, only sudo fsck /dev/sdc2 -c y [02:09] sudo fsck /dev/mmcblk0p2 -c -y? [02:09] ok [02:09] did it write a new error block? [02:09] bad block inode [02:10] yes [02:10] yup [02:10] did it write a new bad block inode? [02:10] you have a failing SD card, then... [02:10] oh fun! [02:10] it's new! [02:11] brand name? [02:11] /dev/sdc2: Updating bad block inode. [02:11] It could also be something like he is writing the partition smaller than the filesystem, right? [02:11] it's a kingston class 10 [02:11] but I'm seeing the same error on my transcend [02:11] I would zero the card [02:12] ugh, ok [02:12] brandini: hey, do you know what kind of FTLs (micro) SD cards have? I couldn't find *any* documentation in wikipedia or google [02:12] Sorry, s/brandini/brokencodes/ [02:12] dd if=/dev/zero of=/dev/sdb bs=8192 [02:12] FTL? [02:12] flash translation layer [02:12] brokencodes: I did that this morning but I flashed the previous daily and then todays daily [02:12] you mean Mean Write before failure? [02:12] The thing that makes it look like a /dev/sda instead of /dev/mmc [02:13] its usually in the SD card interface [02:13] As in the reader? [02:13] si, the reader [02:13] Thanks, I kinda suspected that. [02:14] crappy reader == crappy sd card performance / reliability [02:14] So what I should do is the equivalent of lspci -nn for whatever bus I have instead of PCI, and then look up what kind of FTL I have [02:14] I would say yes [02:14] most emulate ata / atapi [02:15] Also, are micro SD to SD adapters passive (i.e. just changing the physical shape), or do they have active components? [02:15] no, just wires [02:16] passive [02:16] if your SD card reader is plugged into usb, you would do lsusb instead, but you prolly knew that already [02:17] It'd be nice if SD readers had a ioctl you could issue to bypass their FTLs, and just use jffs2/logfs or something :-P [02:17] you can [02:17] its just a block device [02:17] yoyu can make a link, and write as normal NOR flash [02:18] Not if the kernel tells me "hey, here's /dev/sda" [02:18] surely [02:18] ln /dev/sda /dev/mmcblk0 [02:18] wow, really [02:19] ln /dev/sdb2 /dev/mmcblk0p2 [02:19] yep [02:19] No, that can't be right [02:19] * infinity does a whois and suddenly wonders if brokencodes == thebroken, or if it's just odd coincidence. [02:19] The kernel doesn't care what the name is, only the major/minor [02:19] yes [02:19] I'm thebroken [02:20] some utilities wont allow certain functions on certain device names [02:20] They're wrong, then. But I mean stuff like dd and mkfs [02:21] https://secure.wikimedia.org/wikipedia/en/wiki/Memory_Technology_Device <-- what I'm thinking of, where you bypass the FTL [02:21] *bypass, or there isn't an FTL to begin with === Jack87 is now known as Jack87|Away [02:22] http://paste.debian.net/131503/ <-- they show up as character devices, not block devices [02:22] Or a mix. Or something. I guess I don't know what I'm talking about [02:23] I guess a better question would be: if I have a micro SD card that presents as /dev/sda or /dev/mmcblk0, would it be better to use LogFS rather than ext[234] ? [02:24] you could also mknod on the device, and chage to the type of device you want, could work, might not [02:25] I think both sda and mmc are going via the FTL [02:26] Whereas MTD bypasses the hardware FTL and implements on inside the Linux kernel -- mtdblock (b) is mtd (c) + kernel FTL [02:26] Thats what I was referring to also [02:28] I sure hope zeroing makes it work :) [02:31] twb: yup that's it [02:31] well mtd doesn't bypass ftl, it's more that there is no ftl :p [02:32] phh: right [02:32] twb: logfs is theorically better than ext* [02:32] i don't need how it goes in real world [02:32] nilfs2 is indeed faster than ext* [02:33] phh: I'm saying "gee, it'd be nice if I could bypass the crappy hardware FTL when there *is* one, and use jffs2 or whatever on the mtd character device" -- brokencodes says it might work, but I think he's wrong. [02:33] but hum logfs is more for mtd than for FTLed device [02:33] Unless it's some developer board-type SD card reader [02:33] twb: he is wrong [02:33] Thanks, I feel better now :-) [02:34] the thing is that the MMC """api""" doesn't make it easy to make a FTL-less (or FTL bypassed) device [02:34] well afaik [02:34] And since I can't bypass the hardware FTL, I am screwed by it regardless and I might as well use a fs designed for block devices [02:35] twb: na [02:35] twb: on MMCs, hardware FTL sucks [02:35] so a (real) logged FS is way better [02:36] Even on top of the crappy hardware FTL? [02:37] yes [02:37] OK, so I have an 32GB eMMC /dev/mmcblk0p1 that will have the root filesystem, and a 16GB micro SD /dev/sda that will have /home on it. What's the best filesystem to use? [02:37] (ogra_ is going to kill me.) [02:37] Oh, and this is with 2.6.38 [02:38] twb: basically, nilfs2 is the best FS i've found for MMC and friends, now the devs are saying it's unstable, so you might not want to rely on it for your data [02:38] Well I'm using btrfs on my SSD in my old netbook :-P [02:38] Incidentally, stupid SD card reader doesn't handle ext4 -o discard :-/ [02:39] and sdcard does ? :p [02:39] I don't know [02:39] I magical imagination land where everything DWIMs, yes :-) [02:39] DWIMs ? [02:39] Does What I Mean [02:40] :) [02:40] phh: that's where all my hardware has OpenFirmware too :P [02:40] twb: i've been through some specs about it, and trim/erase/whatever it is called is only in some eMMC docs [02:41] (yes eMMC/SD/uSD/MMC/whatever are using the same protocol, and yes it seems there are different specs for them, why ? :D) [02:41] Because as long as it boots windows they will sell it :P [02:46] I really hope this zero'ing pays off... I'm doubtful [02:47] brandini: it depends on how clever FTLs are [02:47] so yes i'm kind of doubtful too :D [02:47] :) === dash__ is now known as dash [02:47] About as clever as a five year old on PCP cut with draino [02:52] well seeing as I've got two brand new SD cards that were both rated to work well and they completely don't work I don't know where to turn [02:52] can the pandaboard boot off usb? [02:58] It's using u-boot? Can you get into the u-boot serial console? [03:00] hrmmm, maybe [03:26] infinity: I need to file a bug on this lack of swap so that it can be added to the release notes and because I may need to fail kubuntu desktop as a result. What should I file against? [06:42] GrueMaster: livecd-rootfs, I suppose. We can wiggle it around if we have to. === Jack87|Away is now known as Jack87 === Jack87 is now known as Jack87|Away [09:14] * ogra_ runs an ac100 install [09:14] bbl [10:20] ogra_: lool: ping [10:21] i think we found a bug in dkms, how do you recommend that we report it? [10:22] i sure can report in launchpad, but isn't that better if i report upstream directly? [10:27] lool: hi. do linaro have access to the secret java tck tests that checks if the arm jvm follows the java specification? [10:30] ndec: pong [10:31] ndec: it's best if you report it upstream directly if it affects upstream, yep [10:31] ndec: you can file it on launchpad as well as a meta-bug tracker to track it upstream, in ubuntu, in your project etc. [10:31] lool: do you want a LP as well once (if) it's confirmed [10:31] xranby: no [10:31] argh.. [10:31] xranby: at least not that I know of ;-) [10:32] lool: ok, it would be good if someone with access to it could test the current armel and armhf packages [10:32] ndec: I don't strongly care about having a LP bug for DKMS personally, up to you :-) [10:32] lool: jamvm work on armhf.. so its simply up to ubuntu and debian to enable and package it [10:32] xranby: armel and armhf packages of what? [10:32] lool: openjdk6 and openjdk7 [10:32] lool: the problem is that dkms status does not work if the module name has 'kernel' in the name. it was working with 10.10. it's broken in oneric. dkms add, build and install work fine though. but since CDBS will do add, build check with status if build was ok and then install. in 11.10 the module is built but not installed. [10:33] xranby: so currently openjdk 6 uses jamvm, but it wouldn't work with openjdk 7; this is going to make it a real pain :-/ [10:33] lool: jamvm can bootstrap itself using openjdk7 [10:35] xranby: apparently some new opcode and API are missing [10:35] lool: so everything that work using openjdk6 + jamvm work using openjdk7 + jamvm robert are working on adding the new bytecode that are used by some dynamic languages so we will get support for those as well [10:36] xranby: Ok; I actually tried contacting Robert a couple of times on this topic, but I didn't manage to reach him === Quintasan_ is now known as Quintasan === txwikinger2 is now known as txwikinger [14:36] hey guys, I've been trying to get ubuntu on a arm board for the last 2 days and had no luck with anything. there is a precompiled version of angstrom that comes with it. Is there anyway of installing / compiling ubuntu on that ? [14:44] define "a arm board" [14:44] :) [14:44] a Technexion Thunderpack [14:45] what cup is on there ? we only support certain cpu types in ubuntu [14:45] s/cup/cpu/ [14:46] OMAP 3530 [14:46] well, theoretically we have omap3 images [14:46] but they are built and tested for beagleboard only [14:46] you could try one of them and see if it works [14:47] will i have to change the uboot settings ? [15:15] ogra_: the standard beagle image doesn't work [15:30] sniperjo_: Try the beagle image with the MLO & u-boot from your system. See if that works. [15:30] GrueMaster: how do i use my u-boot ? just simply copy it across? [15:30] I've just tried the standard beagle image and got http://pastebin.com/xGSa9rL4 , couldn't read uImage [15:31] What you should do is backup the first partition of the SD, reformat (mkfs.vfat -F 32 ), then copy IN ORDER: Your MLO, your u-boot, uImage, uInitrd, boot.scr. [15:32] That's odd. [15:32] I'm in a meeting atm. Hang tight and I should be able to help more in ~15-20 minutes. [15:33] GrueMaster: amazing, thanks ! [15:52] hi everyone. i want to do some research into querying hardware on arm devices in the absence of lspci [15:52] is udev still usable? [15:53] ls /sys [16:09] sniperjo_: Any progress? [16:10] GrueMaster: i was sort of waiting your return, but someone in the linux channel had what seemed to me a pretty good idea, put my original linux back onto NAND and then chroot into the sd card to try it [16:10] janimo: Do you have a comprehensive list of timeout builds for me to shove on to pandas? [16:11] That works too. If you want to do that, the easiest thing to do would be to pull a copy of ubuntu-core for armel from http://cdimage.ubuntu.com [16:12] Use that as a base for your chroot. After chrooting to it, run apt-get update and apt-get install tasksel to pull in major package seeds (desktop, etc). [16:13] ok, i was just following the manual trying to get it back on, i flashed my u boot off a well, how do i get that back on ? at the moment i am just cp -rfa /* /mnt/nand [16:13] I have no idea. Never used that particular device. [16:16] thats what the ins ructions say but I'm getting "cp: recursion detected, omitting directory " http://pastebin.com/2mQAEJDX [16:22] infinity, I have an old list which may be not up to date but I can get you a new one [16:26] janimo: That would be shiny. [16:26] the old list in the meantime https://bugs.launchpad.net/ubuntu/+bugs?field.tag=arm-build-timeout [16:26] washngo and ants definitely are still there [16:27] the haskell one is building ATM [16:28] infinity, yep, all on that list are still valid timeouts [16:28] just checked [16:31] since the haskell-src-ext package was given back on gourd which is an old builder it too will fail in about 10 h [16:33] GrueMaster, does the beagle work with the same power cord as the panda? I do not remember having separate cables for it, but I am being extra careful now in light of recent....events [16:33] it sure fits in the jack :D [16:34] yes. Both work off the same power, as does the mx5 (conveniently). [16:35] janimo: Also, is there a bug for the mx5 kernel issue? [16:35] infinity, yes, jcrigby is on it [16:36] janimo: (I was looking for a bug number) [16:36] https://bugs.launchpad.net/ubuntu/+source/linux-linaro-lt-mx5/+bug/856432 [16:36] Ubuntu bug 856432 in linux-linaro-lt-mx5 "FFE: Enable CONFIG_LBDAF so mx5 image boots fine on EXT4 rootfs" [Undecided,New] [16:36] janimo: Danke. [16:36] jcrigby, infinity says thanks [16:37] :) [16:37] janimo, all TI boards use the same plug and 5v (different amps for different models though) [16:37] 5 v == 5 v [16:37] * janimo betrays his deep understanding of all things electronic [16:37] * ogra_ remembers that with babbage they even changed within one production line [16:37] jcrigby / janimo: You people need nicks with different first characters. [16:38] and its fun if you plug your 19v into a 12v board [16:38] infinity, at least with different colors [16:38] ogra_, I had my share of such fun for this dev cycle [16:38] yeah, i saw [16:38] :( [16:38] I saw _and_ smelled [16:38] last time we had such a disaster it was jamie bennett [16:39] beginning with a j too [16:39] who did exactly what i wrote above [16:39] (19V inot a 12V board) [16:39] the bad thing was that the PSUs looked all the same you could only tell them apart by reading the fineprint on the sticker [16:40] smart boards would use 12 of that for powering and 7 v for fancy effects on the leds [16:40] Thankfully, my ac100 and S10-3 seem to not mind me accidentall confusing their 19V and 20V PSUs (which makes sense, I'm sure their tolerance is wider than that), cause I do that constantly. [16:40] heh [16:41] ogra_: hey, you got a url link to ubuntu console image for panda? [16:41] prpplague, natty headless you mean ? [16:41] ogra_: yea [16:41] Down with headless, up with server! [16:41] http://cdimage.ubuntu.com/releases/natty/release/ [16:42] prpplague, ^^ [16:42] infinity, ++ [16:42] thanks [16:42] infinity, though for panda having minimal headless might make sense once we have real server HW and dont need to abuse pandas [16:43] ogra_: Other than the pool, they're pretty much the same thing anyway. So, it's just a bit of wasted disk space. That doesn't hurt my feelings. [16:43] ogra_: just need something to do a quick test with [16:43] if we get extra headcount or so [16:44] ogra_: But once we have "real server hardware", we could just blank the pool on the cell phone subarches, and a -server build on, say, omap4 or mx5 would end up being "headless". [16:44] yeah [16:45] well, what people asked for (which resulted in headless) was a tiny dev image you could download quickly [16:45] I'm sort of dreaming of a day when we stop shipping for cell phone dev boards, but what are the odds that will ever happen? [16:45] so dropping the pool on the non server arches sounds like a good idea if we actually want to keep that image type at all for these boards [16:46] ogra_: With any luck, a combination of ubuntu-core and a quick tutorial on applying hwpacks should satisfy the need for "tiny". [16:46] ogra_: I'm hoping. [16:46] well, we have one netbook arch too :) [16:46] is there a posted schedule for 11.10 ? [16:46] * ogra_ hasnt seen tegra cellphones yet [16:46] ogra_: My phone is a Tegra2. [16:46] prpplague, http://wiki.ubuntu.com/ReleaseSchedule [16:46] pick your release :) [16:47] infinity, oh, really ? [16:47] ogra_: http://www.lg.com/us/mobile-phones/LG-P999.jsp [16:47] you got one of these hybrid motorolas ? [16:47] ah [16:47] that quite new [16:47] * prpplague is totally frustrated with trying to keep up with android and ubuntu for panda [16:47] Motorola Atrix has been out for quite a while now with the Tegra 2. [16:47] i guess i'm behind :) [16:47] prpplague, drop android, nobody uses it anyway :P [16:47] Droid 3 has the omap4 processor. [16:48] GrueMaster, well, thats a pocket-desktop ... not a phone [16:48] now I'm getting a kernel panic "Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)" [16:48] It's a smart phone with a clamshell docing station. [16:48] *docking [16:48] well, you said the same just differently phrased :) [16:48] sniperjo_: Is this with your stock image? [16:49] * prpplague wonders why this schedule info is not posted on pandaboard.org and omapedia.org [16:50] GrueMaster: yes, http://pastebin.com/LsQe2D85 if you have a look near the end it i asking me to correct the boot option [16:50] GrueMaster: and I'm confused! [16:50] prpplague, yeah, at least until 2014 ... [16:50] then we need new schedules :) [16:51] ogra_: yea, well as you are aware, i normally do board bringup and validation, lately i've been getting hammered on the lists and channels about ubuntu and android support [16:51] well, point them here or to the ubuntu-devel ML [16:51] if you cant handle the load [16:53] sniperjo_: I am not sure how much I can help as I don't have that hw or image. If you followed all the steps to get it back to stock, it should work. [16:54] GrueMaster: ill give it all another go, you think this is the best way to get ubuntu running on it ? instead of changing the beagle settings [16:55] ogra_: not only can i not handle the load, it isn't my area of expertise [16:55] I honestly don't know. I haven't had much of a chance to work on ubuntu on non-supported systems. It should work though. [16:56] prpplague: By all means, pass all ubuntu related questions to us. I regularly monitor the pandaboard mailing list, and respond where I can. [16:57] GrueMaster: is there anyone that can update the omapedia.org wiki pages for ubuntu related stuff? [16:57] prpplague, someone from omappedia i guess :) [16:57] ogra_: hehe which means no one [16:57] I suppose I can help (as time allows), but I don't know if I have write access. [16:57] GrueMaster: public wiki [16:58] http://omapedia.org/wiki/OMAP_Ubuntu_Main [16:58] Ok. Will add to my never-ending todo list. [16:59] GrueMaster: main issues are people are confused about which versions are supported, which ones have PPA's , which ones have wifi support, and what the future releases are going to have [17:00] GrueMaster: i'd be willing to spring for a pandaboard to someone who would keep the ubuntu stuff uptodate on omapedia.org [17:00] usually all the same versions as all other arches modulo LTS are supported [17:00] thats why we are ubuntu-arm and not linaro ;) [17:00] Right, ok. I can help clarify that. Part of the problem we have is making releases work with new board changes, but we have been proactively trying to stay on top of that. [17:01] GrueMaster: yea it is more of a communication issue [17:01] So having new boards that change stuff would be nice. [17:01] I just got an A3 last Friday for example. No issues with Oneiric, but haven't tested the other releases yet. [17:01] GrueMaster: yea, all of the A1 to A3 should be pretty much the same, the A4's will have a few tweaks [17:02] GrueMaster: ahh [17:02] a4 has the next gen SoC ? [17:02] 4460 or so ? [17:02] A4 tweaks? We need to know how that affects us asap for Oneiric release support. [17:02] ogra_: ES2.3 of the 4430 [17:02] ah [17:02] GrueMaster, didnt we just discuss that in the meeting ? [17:03] in the linaro topic [17:03] only major item is the pullup resistors on the edid signals [17:03] not sure if prpplague was in the meeting. [17:03] no [17:03] can the beagle board with the extension board (RS232, eth) output uboot/kernel boot messages on said serial console [17:03] ? [17:03] GrueMaster, he surely wasnt ... only people paid to get up that early come [17:03] :) [17:04] janimo: You mean the zippy board? [17:04] janimo: the standard uart on the beagleboard cna be used for kernel output [17:04] prpplague, and on the zippy ? [17:04] prpplague, standard uart needs some type of connector I don't have. I have null modem cable only [17:04] GrueMaster, BeagleBuddy Zippy rev-A [17:05] janimo: the zippy can not be used as the console without modifying the code, you need to get the correct cable [17:05] I have one, but haven't tested it yet. Need to add a header to my beagle for it to mount. [17:05] prpplague, so there's no kernel cmdline option to switch to that? [17:05] hahaha [17:05] janimo: no [17:05] this is how I got my beagle [17:06] that is sad. How is that port used then? [17:06] janimo: http://tincantools.com/product.php?productid=16144&cat=0&page=1&featured [17:06] I hope the hdmi output works at least [17:07] prpplague, I am not buying more cables :) [17:07] hdmi? on a zippy? [17:07] I have enough cruft already that confuses me :) [17:07] GrueMaster, on the beagle [17:07] DVI on a HDMI slot [17:07] On the beagle it works fine. [17:07] I can bring you a serial cable for your beagle to UDS. [17:08] janimo: well without the cable you have very little chance of actually getting things working proplery [17:08] at least if you dont have a zippy connected :) [17:08] ok, thanks. bye bye people, I am plugging this monitor into the beagle to test daily :) [17:09] * janimo hopes desktop image starts up on the monitoe fine [17:09] janimo: daily desktop will fail misserably due to no swap file. [17:09] Lots of oom errors (showing up as pager errors). [17:09] yeah, we had that before [17:10] * ogra_ remembers fiddling with compcache like crazy to get plars some working images back in the days [17:10] that was before we dicided for a swap file [17:16] GrueMaster, hmm thanks. I'll try something else the now [17:17] I'll fire mine up shortly. I can swap it in place of the mx5 for now. [17:17] FYI: for anyone interested, i am willing to provide a free pandaboard to anyone who will spend some time updating the omapedia.org wiki pages for ubuntu support === lopi is now known as Lopi [17:31] prpplague: sweet :-) [17:32] I think for ease of documenting, it would be better to have omappedia point to wiki.ubuntu.com as we have to keep our own set of documentation and I try to keep it uptodate with current release info. [17:33] +1 [17:33] Otherwise there are two sets of docs that can get out of sync really fast. [17:34] what good is a wiki if it isnt 90% falsehood and lies :-) [17:34] but that does basically mean a free pandaboard for some deletes and a link :-) [17:35]