- Grml - lists.mur.at

Boot by UUID or disk label
by ][ 03 Feb '07

03 Feb '07

Hi, Grml wiki indicates that grml can use disk labels/UUIDs as root= option for booting [1]. What's the correct syntax for that? [1] http://wiki.grml.org/doku.php?id=usb&DokuWiki=88eb0d5110993c6c4d0ca5992abee… I searched hard and came up with the following, trying to boot my USB key: title GRML on 1st HD root (hd0,1) kernel /boot/vmlinuz root=/dev/disk/by-uuid/b53f0a8b-deab-40b3-a84c-0779cb8b7772 ro boot title GRML on 1st HD root (hd0,1) kernel /boot/vmlinuz root=/dev/disk/by-label/grml.boot ro boot title GRML on 2nd HD root (hd1,1) kernel /boot/vmlinuz root=/dev/disk/by-uuid/b53f0a8b-deab-40b3-a84c-0779cb8b7772 ro boot But all gave me root filesystem not found error (grub boot normally, just can't find root fs). I'm sure the uuid and disk label are correct. What else could be the problem? thanks -- Tong (remove underscore(s) to reply) http://xpt.sf.net/techdocs/ http://xpt.sf.net/tools/

2 3

nohwcap
by Erich Minderlein 31 Jan '07

31 Jan '07

Hello I refer to my posting http://www.mail-archive.com/grml@mur.at/msg00781.html in which a whole discussion thread deals with the subject. It had also been relayed to the maintainers of x-servers x.org IMHO the nohwcap issue shall be dealt with only by the kernel and / or any other hardware related software, which are currently the Xserver and the drivers only (in case of insertion / removal of hw). No application should try to find something out about hardware. Kernel shall do it once after booting, insertion or removal of hw only and then hold the data in its virtual filesystem. then such a weird device as /etc/ld.so.nohwcap is at best not existing or second best once read updated during system boot, insertion or removal of hw. best regards a system engineer -- Erich Minderlein <erminderlein(a)locoware.de> MARC Mannheim Rail Commerce GmbH

1 0

sarge and kernel 2.6.18-grml
by Erich Minderlein 30 Jan '07

30 Jan '07

Am Freitag, den 26.01.2007, 12:00 +0100 schrieb grml-request(a)mur.at: > sarge and kernel 2.6.18-grml (Erich Minderlein) Hello I repeat this mail _console and keyboard_ I did not post a question in this mail, so it has been overlooked. I removed console-terminus from the stable(sarge) package(250 kByte) and installed console-terminus from the testing(etch) package(565 kByte). Adding SCREEN_FONT=Lat15-Terminus16 in /etc/console-tools/config did not help; rebooting, prompt login as root, the normal font with serifs shows up, if I type unicode_start: answer : already in UTF8 mode after some fiddling around, I remove /etc/console-tools/config.d/fonty and /etc/console-tools/config.d/ and now it boots and changes to Terminus 16 for all consoles. This will probably not survive an update of package fonty! --Meanwhile fonty removed without negative consequences. Small Umlaut "ae,oe,ue" is properly printed on the console Capital Umlaut "Ae,Oe,Ue" and "scharf S"or (sz) like greek beta is not printed on the console, but results in circumflex accent ^^^^ dpkg-reconfigure console-common and dpkg-reconfigure console-data do not help, sometimes after testing these umlauts the console crashes and I get a new boot-prompt. regards I think the events described are severe: pressing Capital Umlauts Ae Oe Ue leads to console possibly exit or crash Is is a bug in the console-terminus package or is it a bug in the keyboard driver or is it a bug in the interfaces associated with these packages and the kernel ? Is it a bug between sarge and etch ? Is it a bug in front of the display ? All that is possible. I believe that I have exhausted all combinations of rebooting, installing fonts and keyboard drivers, .... Do I have to adapt /etc/console-tools/remap ? I add the result of showkey > somefile kb Modus war UNICODE Taste drÃ¼cken (Programm beendet sich 10s nach dem letzten Tastendruck)...) 0x9c There are two funktion keys on the keyboard: Key Enveloppe press 0x00 0x81 0x9b release 0x80 0x81 0x9b key (i) Information symbol 0x00 0x81 0x96 0x80 0x81 0x96 kb Modus war UNICODE Taste drÃ¼cken (Programm beendet sich 10s nach dem letzten Tastendruck)...) key "Ae" 0x28 0xa8 key "Oe" 0x27 0xa7 key "Ue" 0x1a 0x9a key Return 0x1c 0x9c Now from comparing the latter with the first, .. first UTF8 then UNICODE The latter was today immediately after a system boot. Small Ã¤Ã¶Ã¼ (ae oe ue) work fine, capital Ae Oe Ue dont show on the console and result in three circumflex accent ^^^. Console specialists, what are the consequences ? Thanks -- Erich Minderlein <erminderlein(a)locoware.de> MARC Mannheim Rail Commerce GmbH

1 0

T2 Linux build harness
by Mark 29 Jan '07

29 Jan '07

OK, this tip might not be exactly right for grml, but still: since grml is working on new processor architectures like PPC and 64-bit x86 (and don't forget 64-bit PowerPC!). T2 Linux is all about re-targeting. It impressed us by booting obscure machines on which other "ported" distributions croak. It runs many OEM eval boards. There might be ideas here to help porting work. Just fyi. http://www.t2-project.org/about.html "T2 is highly portable. Adding new architectures is easy and can be done within a day!"

1 0

Brittle init.d scripts - tmpfs ramfs
by Mark 28 Jan '07

28 Jan '07

I never thought Linux was so breakable. We're moving mount points into RAM for performance ( /var/log /var/run /var/lock etc. ). Tmpfs or ramfs can stop disk thrashing and speedup a system. (Flash USB needs it even more, to minimize writes.) Example fstab, http://www.debian-administration.org/articles/179 But....brittle /etc/init.d scripts fail without having their private subdirectories persist across boots, in certain places like /var/log and /var/run. If you write or test init.d scripts, please test with tmpfs and ramfs - do not assume pre-existence of application directories. (P.S. Ubuntu Dapper btw already defaults to tmpfs for such directories, though not /var/log. Dapper uses old style init, Edgy/Feisty have something new.) M

2 4

grml-quickconfig?
by Michael Schierl 28 Jan '07

28 Jan '07

Hi, when I use grml (the live cd), there are a few tasks that I do quite often directly after boot. For example, loading keyboard layout (always), configuring non-dhcp network (sometimes) or starting X or running grml-info for looking something up (rarely). Using cheatcodes for configuring it does not "help", since I have to type the kernel name first, and sometimes I am not even there when it boots (I give the CD to the friend/relative that has problems and tell him that he should boot from it until I am back, and just press Return when "an image of a huge keyboard" appears). So, I'd like to write a script (grml-quickconfig) started at bootup on console 1 that shows a nice screen (with highlighted letters) that allows to do things like that by pressing a single key; the Return key will exit it. Like: Welcome to GRML. Press one of the highlighted keys. Press Return to exit this menu. Configure keyboard language: [d]e [a]t [c]h Configure your [n]etwork card. Show [i]nformation Start [x] d: grml-lang de a: grml-lang at c: grml-lang ch n: netcardconfig i: grml-info x: grml-x (asking for the windowmanager) Windowmanager would be (taken from the completion menu): [d]wm [e]vilwm [f]luxbox f[v]wm fvwm[2] fvwm-[c]rystal ion[3] [j]wm pe[k]wm [p]wm3 [r]atpoison [t]wm w[9]wm [w]indowlab w[m]i wm[i]i wm-[n]g I guess even for people that type really fast this is an improvement (if not, it requires a single keystroke to get out) - and even better for those that cannot type that fast (typing Â»su grml -c 'grml-x fluxbox'Â« is a bit more than typing Â»xfÂ«). The main question now is if something like that would be included into grml and if yes, what other things to add to it (there are still some keys "free"). Michael PS: Recently I tried the "soundtest" script (I wanted to test if a sound card is supported by Linux) and unfortunately it worked... Unfortunately because it got me into a discussion about bad pronunciation of English. I don't want to say that my English pronunciation is much better, but if any native English speaker could re-record the sample file, it would be great. In the meantime, I'm going to use the emacspeak intro sound for such tests...

4 9

grml user survey - please participate
by Michael Prokop 28 Jan '07

28 Jan '07

Hi, I am pleased to announce that I am running a grml user survey. The purpose of this is to help grml developers gain a better idea of what the grml user community is really doing and using, and what aspects people would most like to see improved. This user survey is heavily based on the survey done by Bryan O'Sullivan for the Mercurial Project: http://www.selenic.com/pipermail/mercurial/2006-April/007513.html Mercurial is a great distributed version control system used at grml, see http://grml.org/mercurial/ Please participate! If you know grml users who do not read the grml mailing list, please forward the survey to them and ask them to take part. The more people that respond, the more useful the survey will be! The user survey itself is available online at: http://grml.org/user-survey/grml.txt as well. Please send your responses directly to me (mika(a)grml.org), and do *not* CC the mailing list. Regarding privacy: I will not share your name or contact information with anyone else unless you explicitly allow me to, but I may quote you anonymously (unless you would prefer me not to). At the end of February 2007 I will post a detailed report to the grml mailing list and the grml wiki. Thanks again for taking part! regards, -mika- ---->8 cut here 8<---- About you 1. What country are you in? What is your preferred language? Getting started with grml 1. How did you hear about grml? 2. Did you encounter any problems with grml when starting to use it and if so which ones? 3. Since which release do you use grml? 4. Which version(s) of grml do you currently use? Your habits 1. Do you use grml for private use, work, unpaid projects, or even all of them? 2. How do you obtain grml? Internet (http, ftp, bittorent or rsync?), magazines (Linux-Magazin, Linux-User,...), CD from friends? 3. Do you use any other Live-CDs? * If yes: which ones and why? * If yes: how does grml compare to other Live-CDs you have used? 4. Which operating systems/distribution do you usually work with? 5. Do you use any platforms not x86 based? * If yes, which ones and how often? (daily, weekly,...) 6. Which languages do you speak and which ones do you use via lang=... bootoption with grml? What you think of grml 1. Overall, how happy are you with grml? 2. What do you like about using grml? 3. What would you most like to see improved about grml? (features, bugs, plugins, documentation, ... - whatever you think of) 4. If you want to see grml more widely used, what do you think we could do to make this happen? Documentation 1. Do you use the grml wiki? If yes, do you find it useful? 2. Do you find available documentation useful? 3. What is your favourite user documentation for any software projects or products you have used? Which tools do you use 1. Which tools do you use most frequently on grml? 2. Which scripts of the following ones do you use? [add yes/no] * grml-ap: * grml-bridge: * grml-crypt: * grml-debootstrap: * grml-network: * grml-router: * grml-vpn: * grml-x: 3. Do you use grml as harddisk installation (grml2hd)? * If yes: why? 4. Do you use grml on your usb device through grml2usb? 5. Do you use any special bootoption on regular base? Getting help, staying in touch 1. Have you tried to get grml help from other people? * If yes, did you get these problems resolved quickly and to your liking? 2. Have you used the bug database? * If yes, did you find it useful, easy to use? 3. Do you subscribe to the mailing list? * If yes, do you find it useful, and traffic levels OK? 4. Do you use the IRC channel (#grml on irc.freenode.net)? * If no, did you know that grml developers use IRC, and that there's almost 24-hour help available? Open forum 1. What other comments or suggestions do you have that are not covered by the questions above?

1 0

udev rule for usb boot device
by Mark 26 Jan '07

26 Jan '07

What is the best/approved udev rule for a usb hard drive to escape random drive letter variations /dev/sd[a|b|c|d]. The drives have ATTRS{serial} but I want to walk carefully amidst grml's built-in rules. For example, if sandwiched between other grml rules, what lexical filename to use. Second question: will the udev name work from /etc/fstab during boot, or should we stick to mount-by-label. (Idea for grml2hd - use hostname to auto-generate a related udev naming rule for the drive.)

1 1

sarge and kernel 2.6.18-grml
by Erich Minderlein 26 Jan '07

26 Jan '07

_console and keyboard_ Hallo , I removed console-terminus from the stable(sarge) package(250 kByte) and installed console-terminus from the testing(etch) package(565 kByte). Adding SCREEN_FONT=Lat15-Terminus16 in /etc/console-tools/config did not help; rebooting, prompt login as root, the normal font with serifs shows up, if I type unicode_start: answer : already in UTF8 mode after some fiddling around, I remove /etc/console-tools/config.d/fonty and /etc/console-tools/config.d/ and now it boots and changes to Terminus 16 for all consoles. This will probably not survive an update of package fonty! Small Umlaut Ã¤Ã¶Ã¼ is properly printed on the console Capital Umlaut Ã„Ã–Ãœ and "scharf S"~ÃŸ is not printed on the console, but results in ^^^^ dpkg-reconfigure console-common and dpkg-reconfigure console-data do not help, sometimes after testing these umlauts the console crashes and I get a new boot-prompt. regards -- Erich Minderlein <erminderlein(a)locoware.de> MARC Mannheim Rail Commerce GmbH

1 0

Write barriers for journaling filesystems
by Mark 22 Jan '07

22 Jan '07

Testing exercises with XFS, ext3, and power failure. I found advice to use write barriers. It appears that any journaling file system (XFS/Reiser/ext3) will benefit w.r.t. power failures. Almost all hard drives, IDE, SATA, USB, SCSI have a ~multimegabyte write cache nowadays. "OK, enough theory. Do this: mount all of your filesystems with I/O barrier turned on. Something like this (example line from my /etc/fstab): /dev/sda5 / ext3 defaults,barrier=1 1 1" http://kerneltrap.org/node/6173#comment-193357 "Fortunately, you can have your cake and eat it too. The trick is to implement IO barriers using the CACHE FLUSH and/or FUA commands. That way you can have the performance and MTBF benefits of write-back caching while still having a safe fsync() (safe as in doesn't return before data is on the platters)." http://lwn.net/Articles/181393/ It seems like maybe a good default setting for journaled filesystems, especially on USB (which users often yank out...). M P.S. Update on ZFS. Apple will include it in their next OS release (mika: not as default/boot, still HFS+ for that). Also, Sun may release ZFS under GPLv3 "eventually." http://apple.slashdot.org/apple/06/12/18/1155226.shtml http://www.itweek.co.uk/vnunet/news/2172823/sun-kills-gpl3-rumours

2 2