Ok - I'm still having this issue. I've gathered more information so maybe someone can help out. The information quoted (below) is still accurate and relevant to my issue.
This is what's happening (which is quoted below, too) > The error, after 21 lines of what looks similar to this: > > [21] 0 0 0x800003c0 - 0x800803df (0x20) IS[B](OprU) > > I get the following error: > > "Fatal server error: xf86MapVidMem: Could not mmap framebuffer > (0x4400000,0x400000) (Cannot allocate memory)" The Compaq iPaq workstations are using Intel 82815 graphics controllers. >From Win2k, here are some specifics (which may or may not aid in troubleshooting): BUS: 0,2,0 MemRange: 44000000-47FFFFFF MemRange: 40300000-4037FFFF I/O Range: 03B0-03BB I/O Range: 03B0-03BB MemRange: 000A0000-000BFFFF The Win2k driver specifics are as follows (for reference): Driver Provider: Intel Corp. Driver Date: 7/14/2000 Driver Version: 5.12.1.2636 (note that the Win2k info is for hardware reference, as the chipset works fine in Win2k). Monitors are all HP L1506 LCDs. My lts.conf has the following options: X_COLOR_DEPTH = 24 and I've tried the following (without any luck): X_RAMPERC = 99 X_RAMPERC = 01 X_RAMPERC = 80 Any ideas, insight or help would be greatly appreciated. The labs can function without X_RAMPERC in place, but the students use OpenOffice and other apps (which apparently have this pixmap bug) and it's quite frustrating for them and the teachers. Thanks all! - Jordan --- Jordan Erickson wrote: > OK - after a few hours of testing in a (thankfully) empty lab this > afternoon, I have the following to report (My network is Ubuntu 7.10 > i386 server & clients): > > 1) After an 'apt-get install ltsp-server' and 'ltsp-build-client', my > /opt/ltsp/arch/usr/lib/ltsp/screen.d/ldm was there, including the > X_RAMPERC code! I was surprised, I didn't know it had been ported to LTSP5. > > However, the 'ulimit ' line in the file read with a -m switch. According > to Scott's earlier post, > > "If your shell is dash (posix shell) on Ubuntu, then it's -m. If it's > bash, then it's -v" > > The script itself calls for #!/bin/bash - and I was having issues with > the lts.conf X_RAMPERC option working with OpenOffice crashing the X > session (although it seemed as though Firefox would crash (expectedly) > after loading that school art webpage Jim so humbly offered as a > troubleshooting aid. It was strange, but I had to put in the X_RAMPERC = > 80 switch (again, note the _ in it - some earlier posts omit it, which I > saw before realizing, and felt dumb) for OpenOffice to crash > (expectedly) and not lock up when I inserted some high-volume clipart > from the 'openclipart-*' packages (look in "People", the first two will > knock X11 out in about 4 drags/drops from the gallery). > > What sucks (for me, anyway) *still* is the fact that when X_RAMPERC is > enabled, my Compaq iPaq workstations (with the intel 810 module) won't > boot to ldm (no matter what you put the value at). I get the same thing > as I posted before... > > > --- > The error, after 21 lines of what looks similar to this: > > [21] 0 0 0x800003c0 - 0x800803df (0x20) IS[B](OprU) > > I get the following error: > > "Fatal server error: xf86MapVidMem: Could not mmap framebuffer > (0x4400000,0x400000) (Cannot allocate memory)" > --- > > Any help would be appreciated on my issue here - most of my labs are > affected because they are (mostly) all the iPaqs that are throwing the > error (different hardware, most likely with more RAM, don't have this > issue). > > Come to think of it, I should look in the BIOS to see how much video > memory is allocated. Maybe that'll help. But any insight would be great. > > > Thanks! > > - Jordan > > > Jordan Erickson wrote: >> Ahh, thank you Gavin. I was trying to figure out the issue tracker for >> OOo...seems very complex (as in, you need to know sql to even perform a >> search/query in the db for an open bug). >> >> Thanks again. I have a bit of information regarding the X_RAMPERC that >> might help in my particular situation - I will test hopefully today or >> tomorrow and report back. >> >> >> Sincerely, >> Jordan >> >> Gavin McCullagh wrote: >>> Hi, >>> >>> On Thu, 17 Jan 2008, Gavin McCullagh wrote: >>> >>>> I haven't had time to try this out myself, but if you have time and >>>> inclination, please do and let us know how it goes. If firefox v3 fixes >>>> it, I guess that fix will extend to thunderbird which has also had this >>>> problem. Konqueror and Opera apparently don't show the issue. OpenOffice >>>> seems to be the next application that needs fixing. >>> I've filed a bug against OpenOffice. Please feel free to contribute extra >>> info that seems pertinent. >>> >>> http://www.openoffice.org/issues/show_bug.cgi?id=85321 >>> >>> Gavin >>> >>> >>> ------------------------------------------------------------------------- >>> This SF.net email is sponsored by: Microsoft >>> Defy all challenges. Microsoft(R) Visual Studio 2008. >>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >>> _____________________________________________________________________ >>> Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto: >>> https://lists.sourceforge.net/lists/listinfo/ltsp-discuss >>> For additional LTSP help, try #ltsp channel on irc.freenode.net > -- Jordan Erickson Owner, Logical Networking Solutions http://www.logicalnetworking.net 707-636-5678 Latest LNS Blogs - http://blogs.logicalnetworking.net RFID Passports coming in 2008 Peru orders 260,000 OLPC laptops for kids, Mexico 50,000 CNet to Microsoft: Ditch Vista ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _____________________________________________________________________ Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto: https://lists.sourceforge.net/lists/listinfo/ltsp-discuss For additional LTSP help, try #ltsp channel on irc.freenode.net
