/srv/irclogs.ubuntu.com/2008/08/20/#ubuntu-x.txt

bdmurraybryce_: I had an X crash and reconnected to my screen sessions and now can't start any gui apps00:12
bdmurrayfrom those screen session00:12
bryce_weird00:12
bdmurrayusing 'xhost +' worked around but that's not right00:13
bryce_ahh, so they probably had an old XAuthorization or something00:13
bryce_that's probably standard operating procedure00:13
bryce_there's probably a way to re-get the auths00:13
bryce_(if you care)00:14
bdmurrayyeah kind of00:14
bdmurrayI don't recall this happening on previous releases00:14
bryce_start with man xauth00:14
bryce_also doublecheck that X didn't start with a different $DISPLAY or something obvious like that00:15
bdmurraybryce_: is 'AUDIT: Tue Aug 19 16:27:13 2008: 2667 X: client 29 rejected from local host ( uid=1000 gid=100 pid=31094 )' informative?00:32
bryce_I think that confirms my earlier guess00:35
bdmurraybryce_: setting the environment variable for XAUTHORITY fixes it too and my ~/.Xauthority is empty00:54
bryce_cool00:55
bdmurraymy .Xauthority should be empty though should it?00:55
bdmurrayer, shouldn't00:56
bryce_not if you launched X properly00:58
bryce_(afaik)00:58
bdmurraywell, it restarted after a crash00:59
bryce_hmm00:59
bdmurrayso its not my fault00:59
bryce_you could try restarting X and see if it comes back.  If not,maybe try turning off gdm and launch X via startx and see if something useful appears on the console output00:59
bdmurrayno luck with restarting01:06
bdmurrayand I've no virtual consoles to kill gdm from01:10
bdmurray:(01:10
bryce_can you ssh into it and /etc/init.d/gdm stop ?01:11
bdmurrayerror in locking authority file then01:15
bdmurrayuser not authorized to run the X server01:15
tjaaltonx11perf -aa10text results on 965 w. compiz got 50% better with the exa upgrade patch from fedora09:03
tjaalton~53000 -> 8500009:03
tjaaltonstill nothing stellar though09:04
Ngwrt http://www.phoronix.com/scan.php?page=article&item=intel_x4500hd&num=1 - do we know what the support will look like for x4500s in intrepid?10:09
Ng(the second page kinda suggests that intrepid will have some of the stuff required)10:10
Ngmy 945 desktop is getting a bit tired :D10:10
tjaaltonNg: the intel driver already supports it10:41
Ngtjaalton: ok, well I'm probably going to buy a motherboard with such a chip in the next few days, so I guess that box will be getting an Intrepid upgrade and I'll report back how it goes :)10:50
tjaaltonNg: cool :)10:50
Ng(mainly I want a mobo with a S/PDIF output, but since the 945 is very much failing to play HD content, I might as well step up to a better graphics chip too)10:51
tseliottjaalton: I would like to upload 1 new version of the NVIDIA (177) driver and some fixes for the other flavours. Will you be available later? (I'm uploading the source right now)11:41
tjaaltontseliot: I'm here11:42
tseliottjaalton: ok, thanks, I'll give you the links to the files as soon as the upload is complete11:43
tjaaltontseliot: so you need a sponsor? sure, I can upload them11:43
tseliottjaalton: yes, I don't have any kind of upload privilege and, even if I were a MOTU I couldn't upload the drivers since they are in main.11:45
tjaaltonfirst motu, then core-dev ;)11:46
tseliottjalton: yep, we'll see ;)11:46
tseliottjaalton: here's the list of links: http://albertomilone.com/ubuntu/newlrm/pitti/lista_ia.txt12:11
tjaaltontseliot: thanks, downloading12:15
tjaaltontseliot: the builds need tarballs too, since you've used -sa13:07
tjaalton173 & 177 uploaded13:08
tjaaltontseliot: hmm I'll apt-get them13:08
tjaaltonuploaded13:14
tseliottjaalton: thanks. Sorry about the -sa thing, it's just that I use a script to create the source for all the drivers13:27
tjaaltontseliot: np14:12
tseliotbryce: I uploaded the package (screen-resolution-extra) to REVU14:30
tseliotbryce: dholbach will review it too since we need 2 MOTUs14:31
tseliotbryce: I have sent you an email with the link to my upload to REVU. Daniel told me that he would review it tomorrow18:30
brycetseliot: ok cool18:39
bdmurraybryce: not every -ati bug uses the same driver right?20:00
bryceno, all -ati bugs use the same driver20:02
brycehowever, to confuse things, the '-radeon' driver is an alias for -ati20:03
BraveSpearanyone know anything about dexconf?20:03
bryceso sometimes you'll get reports about -radeon, but that just means -ati20:03
bdmurraywhat about r300?20:03
bryceBraveSpear: a bit20:03
jcristaubdmurray: r300 is the dri driver20:03
jcristau-ati is a wrapper around radeon, r128 and mach6420:03
BraveSpearI need to set screen resolution and refresh rate on a livecd i'm creating.  From what I understand, dexconf creates the xorg.conf on the fly.. I need to know how to have the livecd boot with a resolution of 1024x768 x 16bit color with 60hx frefresh rate20:03
BraveSpearWhen it boots up, it auto detects the monitor and vid card, and dynamically sets the resolution, refresh and color depth on the fly. If it worked on all pc's, then it wouldn't be an issue (the livecd is one I am creating for our work-at-home users that need to access citrix via a web browser through vpn).20:04
BraveSpearAny ideas how to do this?20:04
bdmurraySo my inability to recreate a -ati bug might not mean it is fixed then?20:07
bryceBraveSpear: if I understand correctly, you're seeing some problems with certain pc's not getting autodetected properly, so are trying to default all pcs to the same minimal settings?20:12
bryceyes you can force that in dexconf (you could also force the driver to be vesa, if you really want lowest common denominator settings).  However, generally we focusing on fixing the corner cases that don't work20:14
BraveSpearbryce: thats what I've done as a workaround - XFORCEVESA as a kernel option, but I can only get 800x600 resolution that way.  Unless you know of a way to get it to default to 1024x768??20:15
BraveSpearand thats what my manager wants -- lowest common denominator - he wants the cd to run on as many pc's as possible.20:17
bryceIt should go to 1024x768 if it can; unfortunately I think only 640x480 and 800x600 are guaranteed20:17
BraveSpearIf I can get this running, it will be a cost savings of millions of dollars for my company.20:17
BraveSpearhmmm - maybe the kvm switch I am using is causing it to default to 800x600..20:18
BraveSpearI'll try directly connected to the monitor.20:19
bryceyeah many kvm switches don't forward the edid from the monitor20:19
bryceso if you're testing monitor detection, don't use a kvm20:19
BraveSpearthanks :P20:20
BraveSpearI've been using linux as a casual user for a couple of years, and my boss seems to think that makes me qualified to create a livecd that will change the way our division does business20:21
BraveSpear;)20:21
bryceto be honest, I suspect forcing to vesa and forcing reduced resolution isn't necessarily going to work better than the stock xorg autodetection, unless you're targeting extremely old or extremely unusual hardware20:21
BraveSpearWell, I guess using gentoo for a couple of years counts for something20:21
brycehehe20:21
bryceah, gentoo's fun, I used that for a number of years myself20:21
BraveSpearrofl - gentoo is fun like an enema is fun :P20:25
BraveSpear:)20:25
brycehehe20:27
BraveSpearwell, directly connected to the monitor, the cd boots up to 1280x1024 @ 76hz20:27
BraveSpearwith the vesa driver20:28
brycebtw, there is a version of dexconf at /etc/gdm/failsafeDexconf you should look at20:28
bryceit does something similar to what you're doing (except without ambitions to get >800x640)20:28
bryceBraveSpear: ah good, so that's fully supported then.  It should do 1024x768 as well if you use the Screen Resolution applet20:29
BraveSpearso you think the vesa setting will work on 95% + of all computers?20:29
BraveSpear(thats my managers requirement - not mine)20:30
brycewell, there had been some problems with vesa on some ATI hardware, but I believe that's been resolved now.  I could be wrong21:26
brycebut assuming that issue's gone, then yeah 95% with vesa at 1024x768 seems reasonable21:26

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!