August 2006 - Grml - lists.mur.at

Grml2hd Xorg.conf screen blanks
by Mark 30 Aug '06

30 Aug '06

grml live CD 0.8 final plus "apt-get update; apt-get install grml2hd" At the xorg.conf step, if I say "yes," part of the screen goes black. The dialogs become impossible, although staying active. (If only they were movable with the mouse.) If I say "no" everything works fine. The blackness is top to bottom across ~40% of the horizontal, sort of middle of screen. nVidia GeForce card, but without nvidia drivers, just the live CD with no special boot codes (except noswap). When I ctrl-c out of the script, the screen is still bad, so I have to reboot. M

2 3

afio i/o- Error
by Eugen Romas 30 Aug '06

30 Aug '06

Hallo liebe Freunde, ich bin neu hier und habe gleich ein Problem. :'( Vielleicht kann mir jemand helfen? Ich betreibe ein Debian-Sarge-Crypto-Software-Raid1-System mit zwei Festplatten auf einem Scaleo 600 AXP 21 1 1500MB (AMD Athlon(tm) XP2100+). Bootloader: GRUB, Kernel: 2.6.8-3-k7, Partitionierung: 1. /boot unverschlüsselt /dev/md0 2. / verschlüsselt /dev/mapper/root 3. /usr „ /dev/mapper/usr 4. /var „ /dev/mapper/var 5. /swap „ /dev/mapper/swap 6. /tmp „ /dev/mapper/tmp 7. /home „ /dev/mapper/home 8. unverschlüsselt /dev/md7 Alle sind ext3fs Filesysteme. Von diesem PC will ich eine verschlüsselte Komplettsicherung erstellen. Nach dem Booten von *grml 0.8 - Codename Funkenzutzler *finde ich alle Raid- Partitionen richtig zugeordnet wieder. Nun habe ich die "mapped crypt devices" mit cryptsetup angelegt, genau so wie sie normalerweise erstellt werden. Mit mkdir neue mountpoints unter /mnt angelegt und alle Partitionen gemountet. Lesezugriffe funktionieren einwandfrei. afio bringt aber leider i/o- Error bei einigen wenigen Files. Das bash- Skript, das ich verwende, funktioniert mit dem Originalsystem ohne Schwierigkeiten, Plattenplatz ist also genug da. ... find $DIRECTORY -mount -depth | afio -ovZ -P gpg -Q --symmetric -Q --passphrase-fd=3 -Q --no-verbose -Q--batch -Q --no-options -3 3 $ARCHIVNAME 3<$BACKUPCFG ... ... Die Ausgabe erfolgt auf /mnt/md7. Was kann die Ursache sein? Welche Lösung gibt es? Viele Grüße aus dem schönen Thüringen von Eugen **

2 2

fglrx victory
by Martin Yazdzik 30 Aug '06

30 Aug '06

Dear mika and friends, As y ou can see from the attached, I have 3d working. I cheated - since the xorg log error was "unable to find screen" I took my working file from grml 7 and simple copied it. The differences are: Working: Section "Monitor" Identifier "Monitor0" # ModelName "Old Monitor (no DDC)" Option "DPMS" "true" # HorizSync 28.0 - 78.0 # Warning: This may fry very old Monitors # HorizSync 28.0 - 96.0 # Warning: This may fry old Monitors HorizSync 28.0 - 96.0 # VertRefresh 50.0 - 76.0 # Very conservative. May flicker. # VertRefresh 50.0 - 60.0 # Extreme conservative. Will flicker. TFT default. VertRefresh 50.0 - 60.0 # Need more information? # http://xtiming.sourceforge.net/cgi-bin/xtiming.pl # http://en.tldp.org/HOWTO/XFree86-Video-Timings-HOWTO/ EndSection Section "Device" ### Available Driver options are: ## sw_cursor is needed for some ati and radeon cards # Option "sw_cursor" # Option "hw_cursor" # Option "NoAccel" # Option "ShowCache" # Option "ShadowFB" # Option "UseFBDev" # Option "Rotate" ## xorg + nvidia: # Option "RenderAccel" "true" # Option "AllowGLXWithComposite" "true" Identifier "Card0" # The following line is auto-generated by grml-x Driver "fglrx" VendorName "All" BoardName "All" ## Workaround for drivers which send output to wrong device: # Option "MonitorLayout" "LVDS, AUTO" # Option "MonitorLayout" "LVDS, CRT" # Option "MonitorLayout" "NONE, STV" # Option "MonitorLayout" "LVDS" ## Specify BusID: # BusID "PCI:1:0:0" EndSection Section "Screen" Identifier "Screen0" Device "Card0" Monitor "Monitor0" DefaultColorDepth 24 SubSection "Display" Depth 1 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 4 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 8 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 15 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 16 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 24 # Modes "1920X1200" "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection SubSection "Display" Depth 32 # Modes "1024x768" "800x600" "640x480" "1600x1200" "1280x1024" "1280x960" EndSubSection EndSection Non-wokring: Section "Monitor" Identifier "Monitor0" # ModelName "Old Monitor (no DDC)" Option "DPMS" "true" # HorizSync 28.0 - 78.0 # Warning: This may fry very old Monitors # HorizSync 28.0 - 96.0 # Warning: This may fry old Monitors HorizSync 28.0 - 96.0 # VertRefresh 50.0 - 76.0 # Very conservative. May flicker. # VertRefresh 50.0 - 60.0 # Extreme conservative. Will flicker. TFT default. VertRefresh 50.0 - 60.0 # Need more information? # http://xtiming.sourceforge.net/cgi-bin/xtiming.pl # http://en.tldp.org/HOWTO/XFree86-Video-Timings-HOWTO/ EndSection Section "Device" ### Available Driver options are: ## sw_cursor is needed for some ati and radeon cards # Option "sw_cursor" # Option "hw_cursor" # Option "NoAccel" # Option "ShowCache" # Option "ShadowFB" # Option "UseFBDev" # Option "Rotate" ## xorg + nvidia: # Option "RenderAccel" "true" # Option "AllowGLXWithComposite" "true" Identifier "Card0" # The following line is auto-generated by grml-x Driver "fglrx" VendorName "All" BoardName "All" ## Workaround for drivers which send output to wrong device: # Option "MonitorLayout" "LVDS, AUTO" # Option "MonitorLayout" "LVDS, CRT" # Option "MonitorLayout" "NONE, STV" # Option "MonitorLayout" "LVDS" ## Specify BusID: # BusID "PCI:1:0:0" EndSection Section "Screen" Identifier "Screen0" Device "Card0" Monitor "Monitor0" DefaultColorDepth 16 SubSection "Display" Depth 1 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 4 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 8 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 15 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 16 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 24 # Modes "1920x1200" EndSubSection SubSection "Display" Depth 32 # Modes "1920x1200" EndSubSection EndSection Since the screen is correctly identified in the non-working file as 1920x1200, which is in fact the current and functioning resolution, and the olny other difference appears to the the default bit depth, I am perplexed. However, there is definitely a speed improvement over 7. Also, I fixed the time zone issue with brute force, by deleting every reference in every /etc file I could find - works fine now.... Thus, we have linux for sysadmins and one rock ape - martin... note also, I am using the ati 28 drivers and kernel-module. Now, if the hsf driver works, I am in business. Thanks for all the help, hard work, and sheer brilliance. Gruesse aus NY, M

2 1

RE: JMicron 20360/20363 AHCI Controller ### fw to grml@mur.at
by Felix Resch 30 Aug '06

30 Aug '06

>>* Felix <f_resch(a)rosenberger.de> [20060821 09:15]: >> >> i just bought new Hardware (asus p5b deluxe and a core2) >> and ran into probs when booting grml 0.8 (and gentoo 2006.0) >> from a (parallel attached) cdrom drive. >> >> when booting the kernel it doesent seem to find any install sources >> (namely it cant detect the cdrom drive) and so drops me to the busybox shell. [...] >Do you have any further news which could help me in investigating >the problem? > >regards, >-mika- Hi, I worked around and installed gentoo via pxe :) The installation worked with 2.6.17-r5 and the kernels sata ahci driver recognizing all sata hds. Afterwards i?ve tried booting gentoo 2.6.17-r5 and -r6 and vanilla 2.6.18-rc4 ; neither could get the controller to function properly. The JMicron controllers seem to be detected by the system, but not the attached drives... In the Bios you may set IDE or AHCI mode to either the SATA and the "PATA" controller but for me this makes no difference. root@wood ~ # lshw -C storage *-ide description: IDE interface product: JMicron 20360/20363 AHCI Controller vendor: JMicron Technologies, Inc. physical id: 0 bus info: pci@03:00.0 version: 02 width: 32 bits clock: 33MHz capabilities: ide bus_master cap_list configuration: driver=ahci resources: ioport:bc00-bc07 ioport:b880-b883 ioport:b800-b807 ioport:b480-b483 ioport:b400-b40f iomemory:fe9fe000-fe9fffff irq:16 *-storage description: SATA controller product: 82801HB (ICH8) SATA AHCI Controller vendor: Intel Corporation physical id: 1f.2 bus info: pci@00:1f.2 version: 02 width: 32 bits clock: 66MHz capabilities: storage ahci_1.0 bus_master cap_list configuration: driver=ahci resources: ioport:ec00-ec07 ioport:e880-e883 ioport:e800-e807 ioport:e480-e483 ioport:e400-e41f iomemory:febff800-febfffff irq:18 I found some posts concerning the problem with the ide controller. Afik it boils down to the "stupid error" alan cox made, and that the patches will be in 2.6.18 final. see http://www.knoppix.net/forum/viewtopic.php?t=25196&sid=dd496b66a9b801cd397e3 36e44dea772 and http://fcp.surfsite.org/modules/newbb/viewtopic.php?topic_id=24885&forum=12& post_id=105517 and http://www.uwsg.iu.edu/hypermail/linux/kernel/0608.2/0688.html some technical information http://www.gatago.com/linux/kernel/14665031.html @mika: I suppose the "internal server error" has been fixed, since the page looks to work fine from here. I hope you excuse me writing in english towards the benefit for people that dont speak german. thanks for your interest Felix

2 1

time zone again
by Martin Yazdzik 30 Aug '06

30 Aug '06

Dear Mika and Friends, Okay, as you read in the other thread, I now have fglrx with working 3d hw accel. I dare not, however, boot out of old grml into new, because new grml is convinced that my hw clock is set to gmt. Since this is a laptop, I need to boot into new grml and tell it that my clock is in NYC, edt, my laptop is, sadly, in NYC, edt, and there is only one time I need to know today, which is the time in NYC. tz config sets the time to six hours earlier, because it is sure that I must be wrong. I am not. I know where I am. I have disabled all the time stuff in autoconfig. There is not point even trying to set up evolution and so on until every app takes its time settings from the system clock, which should take its settings from the cmos clock, which, because we all have to boot into windows once a year, is set to local time. Please find a simple solution to this. Best, M -- "...children and young people feel instinctively the difference between those who genuinely wish them well and those who regard them merely as raw material for some scheme." --Russell

2 1

Re: [Grml] everthing except fglrx (goodnews/badsnews)
by Martin Yazdzik 29 Aug '06

29 Aug '06

On Tuesday 29 August 2006 08:06, Lars Schimmer wrote: > > I think I´ve got 3d hardware running, because glxinfo ist really fast. > Before installing it right, glxgears was damned slow. > And fgl_glxgears is fast, to. Thanks - according to your attachment' name of display: :0.0 display: :0 screen: 0 direct rendering: No by fast, do you mean more fps, or the gears move faster? vielen dank, m -- "...children and young people feel instinctively the difference between those who genuinely wish them well and those who regard them merely as raw material for some scheme." --Russell

1 0

Re: [Grml] everthing except fglrx (goodnews/badsnews)
by Martin Yazdzik 29 Aug '06

29 Aug '06

Dear Friends, Many thanks to Lars for the files. Lars has the driver working, but not the kernel modules for 3d. I can, with no effort, have the fglrx driver for 2d.(never been an issue - perhaps I was unclear - by "working, I mean a full module load with functional 3d, nothing less, since that is the design objective of the driver. Note that the usual blurb is "if x works but no 3d, kernel module is not loading.) Finally, one can build and load the kernel module from either the grml sources, or the ati site. This is big progress, as only the grml sources worked in .7. The good news on this is that the behaviour of the live cd and the install are exactly the same - whether grml-x -module fglrx or modprobe fglrx, I get the same result. The bad news is that that result is best described as the screen slowly blanking to a white mess of nothing. Notes: same for all versions of xorg, 7 and 7.1. 3d works perfectly in grml 7 on this box, with xorg 7.0, so there is no issue with 7.0, in se. only the grml version works in .7, any attempt to use the ati packages does not allow the build of a loadable module. obviously, the ati installer works fine in sid when sid is installed via the etch installer, although I have not tested it with xorg 7.1. Please read the word "works" as in strict english, to wit, function as per design, i e, 3d is provided by fglrx libs, since the purpose of the driver is to provide both 2d and 3d acceleration. So, why does the driver work in grml 7 but not 8? glxinfo.txt is from the .7 install. Will keep exploring. Best, m -- "...children and young people feel instinctively the difference between those who genuinely wish them well and those who regard them merely as raw material for some scheme." --Russell

1 0

everthing except fglrx works
by Martin Yazdzik 28 Aug '06

28 Aug '06

grml2hd was without any problem whatsoever, except for ghe fglrx driver - I shall try two or three solutions and write back, but grml-x should work to produce a valid working 3d system if at all possible. Well, that is a big "if" ;) best, M

2 1

Re: Grml Digest, Vol 16, Issue 18
by Eugen Romas 27 Aug '06

27 Aug '06

grml-request(a)mur.at schrieb: > Send Grml mailing list submissions to > grml(a)mur.at > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.mur.at/mailman/listinfo/grml > or, via email, send a message with subject or body 'help' to > grml-request(a)mur.at > > You can reach the person managing the list at > grml-owner(a)mur.at > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Grml digest..." > > > Today's Topics: > > 1. afio i/o- Error (Eugen Romas) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 17 Aug 2006 11:01:17 +0200 > From: Eugen Romas <eugen.romas(a)romas.de> > Subject: [Grml] afio i/o- Error > To: grml(a)mur.at > Message-ID: <44E4305D.2010201(a)romas.de> > Content-Type: text/plain; charset=UTF-8 > > Hallo liebe Freunde, > ich bin neu hier und habe gleich ein Problem. :'( Vielleicht kann mir > jemand helfen? > > Ich betreibe ein Debian-Sarge-Crypto-Software-Raid1-System mit zwei > Festplatten auf einem Scaleo 600 AXP 21 1 1500MB (AMD Athlon(tm) XP2100+). > Bootloader: GRUB, Kernel: 2.6.8-3-k7, Partitionierung: > 1. /boot unverschlA~1/4sselt /dev/md0 > 2. / verschlA~1/4sselt /dev/mapper/root > 3. /usr a^EURz( /dev/mapper/usr > 4. /var a^EURz( /dev/mapper/var > 5. /swap a^EURz( /dev/mapper/swap > 6. /tmp a^EURz( /dev/mapper/tmp > 7. /home a^EURz( /dev/mapper/home > 8. unverschlA~1/4sselt /dev/md7 > Alle sind ext3fs Filesysteme. > > Von diesem PC will ich eine verschlA~1/4sselte Komplettsicherung erstellen. > Nach dem Booten von *grml 0.8 - Codename Funkenzutzler *finde ich alle Raid- Partitionen richtig zugeordnet wieder. Nun habe ich die "mapped crypt devices" mit cryptsetup angelegt, genau so wie sie normalerweise erstellt werden. Mit mkdir neue mountpoints unter /mnt angelegt und alle Partitionen gemountet. Lesezugriffe funktionieren einwandfrei. afio bringt aber leider i/o- Error bei einigen wenigen Files. > Das bash- Skript, das ich verwende, funktioniert mit dem Originalsystem ohne Schwierigkeiten, Plattenplatz ist also genug da. > ... > find $DIRECTORY -mount -depth | afio -ovZ -P gpg -Q --symmetric -Q --passphrase-fd=3 -Q --no-verbose -Q--batch -Q --no-options -3 3 $ARCHIVNAME 3<$BACKUPCFG > ... > ... > Die Ausgabe erfolgt auf /mnt/md7. > > Was kann die Ursache sein? > Welche LA~?sung gibt es? > > Viele GrA~1/4A~Y"e aus dem schA~?nen ThA~1/4ringen von > Eugen > ** > > > > ------------------------------ > > _______________________________________________ > Grml mailing list - Grml(a)mur.at > http://lists.mur.at/mailman/listinfo/grml > join #grml on irc.freenode.org > grml-devel-blog: http://grml.supersized.org/ > > End of Grml Digest, Vol 16, Issue 18 > ************************************ > Hello all, sorry that I used German. Is anyone here who can help? If more Informations are required please ask for them. THX. Eugen

1 0

error: linux/errno.h not found
by T 26 Aug '06

26 Aug '06

Hi I must have one fundamental step missing from setting up my compiling environment. This is what I got when compiling a c file: In file included from /usr/include/errno.h:36, from archh.c:1637: /usr/include/bits/errno.h:25:26: error: linux/errno.h: No such file or directory make[1]: *** [/home/tong/local/bin/archh] Error 1 I found the file "linux/errno.h" is at /usr/src/linux-headers-2.6.16-grml/include/linux/errno.h $ ls -l /usr/src/linux-headers-2.6.16-grml/include/linux/errno.h -rw-r--r-- 1 root root 1027 03-20 06:53 /usr/src/linux-headers-2.6.16-grml/include/linux/errno.h How should I tell gcc to look for headers in /usr/src/linux-headers-2.6.16-grml/include for all c files, the Debian way? thanks (and sorry for the a bit OT question). tong

1 1