[10:48] <\sh> benc: please fix wlan connection for atheros for feisty pls ;) 
[10:50] <\sh> 2.6.20-9 worked and -15 doesn't work :(
[11:10] <Nafallo> \sh: a bit late to say that now. isn't it? :-)
[11:11] <\sh> Nafallo, there was a bug report...and I didn't update my r200 since 2.6.20-9 because of no time :(
[11:11] <\sh> actually there _is_ a bug report :)
[11:12] <Nafallo> :-)
[11:13] <Nafallo> -ENOTIME seems to be a common issue :-/
[11:13] <\sh> https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/96754
[11:13] <Nafallo> with no obvious fix or work-around.
[11:13] <\sh> reported   2007-03-27  
[11:22] <\sh> Nafallo, well, it's a regression from -9 to >-9 
[11:23] <\sh> I ran into it yesterday evening...:( good that I had -9 still in my bootable kernel list ;)
[11:38] <abogani> ..."The gutsy build system was rewritten from the ground up, to mainly exclude kernel-package, since it was getting too heavy and restrictive for the kernels builds."... I'm very happy to hear this! :-)
[11:38] <abogani> BenC: Could i disturb you with a couple of questions?
[11:47] <abogani> Are there other Ubuntu Kernel Team members that can reply to a couple of question about my specification (https://wiki.ubuntu.com/RealTime) ?
[11:57] <maks_> abogani: lots of 2.6.20 floating around :-P
[12:02] <abogani> maks_: I understand, sorry.
[12:04] <AnAnt> Hello, I think there is problem in new kernel
[12:04] <AnAnt> I can't set the DMA to be ON for cd writer
[12:04] <AnAnt> I did sudo hdparm -d1 /dev/scd0
[12:05] <AnAnt> so I got: HDIO_SET_DMA failed: Inappropriate ioctl for device
[01:11] <zdzichuBG> anant: why do you think DMA is off?
[01:32] <Amaranth> zdzichuBG: He left :)
[01:32] <Amaranth> zdzichuBG: Is it on by default for PATA devices?
[03:05] <roh> morning
[03:16] <roh> i seem to hit a strange ext3 problem once more latelz
[03:17] <roh> eh y
[03:17] <roh> this time on a thinkpad T60 which after clean shutdown had a nonclean ext3 afterwards which i was unable to mount anymore
[03:25] <roh> fsck is running > 12 hours now (the disk says its good, no io errors at all, smart passes, no moved blocks)
[05:00] <enseven> Hi! I've got a question about redhat cluster in ubuntu kernel.
[05:03] <enseven> In linux-source-2.6.15 there is a subdir "cluster" and in .config there is CLUSTER, CLUSTER_DLM, CLUSTER_DLM_PROCLOCKS. Where have they gone in linux-source-2.6.20 ?
[05:07] <zul> ubuntu/ directory I think
[05:14] <enseven> No, there is no "cluster" in ubuntu/.
[05:15] <enseven> Is there a patch to apply redhat cluster to ubuntu kernel 2.6.20?
[05:15] <fabbione> enseven: no... but the cluster is all there
[05:15] <fabbione> enseven: most of the modules have kernel modules are now userland
[05:16] <fabbione> enseven: only gfs2/gfs1 and dlm kernel modules are required
[05:16] <fabbione> dlm/gfs2 is from upstream.. so in fs/
[05:16] <fabbione> gfs1 is in ubuntu/fs/
[05:16] <fabbione> enseven: also note that gfs2 is still eXPERIMENTAL and not ready for production
[05:18] <enseven> fabbione: Is there cman, too? Or is it obsolete?
[05:20] <fabbione> cman -> userland
[05:26] <enseven> If I change from 2.6.15 -> 2.6.20, do I need new redhat-cluster-suite? Newer than 1.20060222-0ubuntu6?
[05:32] <zul> *sigh* http://xenbits.xensource.com/kernels/
[05:33] <fabbione> enseven: yes
[05:34] <fabbione> enseven: you will need the very latest from feisty
[05:36] <enseven> Ok, I'm going to try that. Thanx.
[05:36] <fabbione> enseven: you will need heaps of Depends backported
[05:36] <fabbione> i am not sure it's worth the trouble
[05:36] <fabbione> some of them might not even build in dapper
[05:36] <fabbione> and i won't be able to support you
[05:37] <fabbione> enseven: what's the reason why you want .20 anyway?
[05:37] <enseven> Ohh, not so easy as I thougt?!
[05:37] <fabbione> no it's not easy
[05:38] <fabbione> if you already have a cluster running, you will need to fix the configuration too
[05:38] <enseven> I have got a problem with GFS and Samba. So I wanted to find out if it is still there in .20.
[05:39] <fabbione> enseven: understood
[05:39] <enseven> The problem is not real bad, but it is uncomfortable.
[05:39] <fabbione> what kind of problem are you seeing?
[05:40] <fabbione> how many nodes are in the cluster?
[05:40] <fabbione> (also.. #linux-cluster here might be better help if it's an upstream issue)
[05:42] <enseven> I have a samba share with acls on gfs. When I change attributes of a directory from a windows client, I get am error message: Access denied! But then the changes are applied when I cancel.
[05:42] <fabbione> hmm interesting
[05:43] <enseven> This does not happen with files and it does only happen with gfs, not with ext3.
[05:45] <fabbione> i can't see anything in the cvs changelog related to samba
[05:45] <fabbione> if the cluster is only 2 nodes, you can perhaps stop the cluster, make one node feisty and see
[05:45] <fabbione> make it feisty with a spare disk or something along the line
[05:45] <fabbione> just for testing
[05:45] <fabbione> it's probably faster than backporting the * for testing
[05:49] <enseven> fabbione: It is a 3 node cluster, but I could stop the cluster, make one node feisty and then say cman_tool expect -e 1 to run the services with one node only
[05:50] <enseven> btw: it is a virtual machine so spare disks are no problem.
[06:11] <enseven> thanx
[06:26] <bdmurray> BenC: Have you seen any PS2 mouse bugs like 108350?
[06:29] <BenC> bdmurray: I hadn't heard anything, but I can honestly say I don't have a real ps2 mouse to test with
[06:30] <bdmurray> BenC: I tested on one system here and didn't have an issue
[06:30] <BenC> ps2 is so basic, I can't understand it having problems
[06:42] <roh> hm... are there any known ext3-issues with recent feisty kernels or anything that would break a fs on blocklevel? i'm on a TP60p here. 100gb sata on ahci at ICH7
[06:52] <johanbr> roh: There were a few bugs like that, but I think they were fixed in the final Feisty kernel.
[06:56] <roh> johanbr i installed with the beta 2 days before the release and everythin was quite fine. yesterday midday i shut down the note cleanly and stashed it away. in the evening the X11 didn't come up because of something not writeable and the kernel trew some kind of hci/usb errors. then i rebootet cleanly and it came up without its rootfs ... it sayd "#fragments of group too big" and mount failed with invalid argument
[06:57] <roh> currently i'm typing this with an edgy bootcd since the initrd of feisty didn't bring me fsck, and the hw is working fine (fresh lenovo T60p (8741-J3G))
[07:00] <johanbr> roh: Probably this bug: https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/82314
[07:00] <roh> fsck couln't find the superblock/meant it is corrupted. fsck -n -v on the fs ran through in about 100minutes so i started it without -n... since then i'm waiting for it to finish (> 15h now)
[07:01] <johanbr> Ouch.
[07:02] <roh> i gave it the second superblock which it accepted. just now it takes for ages and spits out zillions of multiply-claimed blocks in multiple files
[07:02] <roh> johanbr the disk i'm working on with that fs is sda via ahci. no pata. native sata
[07:05] <roh> also ich7 chipsel, no via in thinkpads
[07:05] <johanbr> roh: I'm not sure, but it's possible the problem is with drives with hpa in general, not just pata.
[07:06] <roh> what does hpa exactly do? i deactivated all that rescue-foo in the bios and used the whole disc.
[07:07] <johanbr> http://en.wikipedia.org/wiki/Host_Protected_Area
[07:10] <roh> uh.. shouln't the bios switch that off completely as soon as i disable the rescue-foo and make it visible?
[07:11] <johanbr> Don't know.
[07:12] <roh> any idea how to check if that is indeed the cause of my problem?
[07:13] <johanbr> Try the release kernel and see if you get any corruption. :)
[07:16] <roh> good joke. i do not have any other (free) 2.5" sata disks left
[07:23] <roh> *sigh* .. as more i read about hpa and "rescue" s_it as more i feel the sudden urge to burn some companys to the ground...
[07:37] <johanbr> roh: If there's nothing important on the disk, it's probably easiest just to reformat, reinstall the final Feisty release and play with it for a while.
[07:39] <roh> i have to check first what i moved and what i copied from the old disk (had a dapper on t60 before) but i have to give that machine back soon (old bosses(
[08:13] <roh> uh.. fsck is still continuing and making progress... 77% it sayd... if that runs through and i do not loose this install i will propably use ext based fs till i die %-] 
[09:02] <zul> BenC: ping 
[09:02] <BenC> zul: yo
[09:03] <zul> BenC: have you heard of X-windows crashing with nvidia and the mouse not able to be accessible with the scroll-wheel?
[09:03] <BenC> zul: not specifically those things, no
[09:03] <zul> okie dokie
[09:04] <zul> good luck on your thingy today
[09:09] <roh> ieks "XXX should never happen!!!"
[09:36] <johanbr> roh: fsck finished yet?
[10:54] <h4wk0> On the subject of my  question in #ubuntu-classroom - Ive been pushed her
[10:55] <JanC> h4wk0: wait until after the classroom session  :)
[10:55] <h4wk0> How come my ATI X1300 require an extra little fix
[10:55] <h4wk0> Ok
[11:01] <joebaker> Hi BenC
[11:01] <j1mc> hi BenC:  QUESTION: What new features are you anticipating being integrated with the newer versions of the kernel, and which are you most excited about?  =)
[11:01] <BenC> I'm taking 10 minutes for a break...brb to answer your questions
[11:01] <joebaker> Thanks!  It was neat to hear about the challenges you face.
[11:01] <j1mc> Thanks, BenC
[11:02] <alienSkull> Thanks Ben
[11:06] <joebaker> I have a laptop which only has the max RAM capacity of 192mb.  I'd like it to be able to access RAM in a server as swap over ethernet?
[11:08] <joebaker> I know the LTSP project has been swapping over Ethernet to server disk space for years and that they've recently changed to something called ltspfs.  Yet I see this as beneficial to more than just thin clients.
[11:08] <el_isma> Is it really that fast over disk swap?
[11:09] <h4wk0> From my opinion - I dont think it will!
[11:11] <alienSkull> well you would need to compare gigabit ethernet with harddrive IO speeds right?
[11:11] <h4wk0> extensive benchmarking will be needed
[11:12] <BenC> joebaker: I've heard that nbd (network block device) can be used for swap
[11:12] <alienSkull> if the speeds do turn out in its favor I think it would have great applications in thin clients
[11:12] <el_isma> what's the typical harddrive latency?
[11:12] <joebaker> Benc  Thanks.  Yes, that's what LTSP used to use.
[11:13] <BenC> h4wk0: can you paste your URL again please?
[11:13] <h4wk0> Sure
[11:13] <h4wk0> http://ubuntuforums.org/showthread.php?t=399913
[11:13] <h4wk0> I am not the creator of the thread, however have the same problem (I am using ATI x1300)
[11:14] <h4wk0> And it worked perfectly on Herd 2
[11:14] <BenC> h4wk0: I've have your exact model of laptop here and i see the same problem. The main issue is that the ati xorg driver doesn't support the chipset, and the ATI vesa BIOS is broken. But we are doing some tests on work arounds
[11:14] <BenC> h4wk0: in the mean time, using fglrx is the best way (using an alternate CD install)
[11:15] <BenC> livecd wont work at the moment
[11:15] <h4wk0> Well if you need anything testing
[11:15] <h4wk0> <--
[11:15] <BenC> thanks, will do
[11:15] <h4wk0> Havent tried alertnate cd yet :P (Had to pop out and know ive got side tracked with tv)
[11:18] <joebaker> Swapping against remote ram:  I think if it is built that applications will come to it.  Well there now the idea has been planted.  It's time for me to go.  Thanks again for the wonderful presentation Benc.
[11:18] <roh> johanbr it exitet with the above error... 
[11:19] <BenC> joebaker: np, thank you for participating
[11:24] <roh> ah.. and "Directory inode 8093697 has an unallocated block #3.  Allocate<y>?" makes it do that
[11:25] <roh> if i say no there it continues till the end
[11:41] <johanbr> roh: Hmm. Are most of your files intact?
[11:54] <roh> johanbr will see.. i want to dd it to an image before doing anything else (just got a big usbdisk)
[12:01] <h4wk> Right time to install on laptop