Upgrade etch to lenny on virtual server (Virtuozzo) with 2.6.18 kernel
Hello Debian users, I have a virtual private server (VPS) that is running a minimal install of etch (image used for installation provided by the hosting company). I believe they use Virtuozzo for virtualisation. Since the virtual servers are not run using Xen, this means I am bound by the kernel provided by the hoster since all virtual machines are sharing the kernel. The kernel is: servername:~# uname -a Linux servername.provider.tld 2.6.18-028stab060.2 #1 SMP Tue Jan 13 10:24:09 MSK 2009 i686 GNU/Linux Now my question is: Is it possible (or wise) to try to upgrade to lenny? Is it good practice to combine a lenny system with an etch kernel? Are there security risks? Maybe it's better to stick with etch and keep the system updated with security updates as long as they are available? The main purpose of the machine is being resposible for e-mails (exim, dovecot), hosting a tiny website. But on the other hand, I am playing around with it a lot and experimenting with running other services, so sometimes I'd welcome newer package versions. If you don't see reasons against this update: Would you recommend an installation using debootstrap, or is it fine to do the following? 1) /etc/apt/sources.list : Change occurences of etch to lenny 2) aptitude update 3) apt-get install aptitude 4) aptitude dist-upgrade I have simulated the above, and realised that there are some dependency problems, but they seem to result from the fact that it is a minimal install without X server etc. and thus there are unmet dependencies (depends) even now in the current installation. Thanks for your input! -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: strange gs problem: unicode encoding problem with pdf (?)
Florian Kulzer at 2008-06-03 20:44+02:00: I suspect that something is not quite kosher with the PDFs that your bank generates. The fact that they work with the Adobe reader does not necessarily mean that they conform 100% to the PDF specification. According to the version string, your original file is supposed to be an Acrobat 4 file (i.e. PDF 1.3). I opened your PDF in Adobe Acrobat Professional 8 (on a Macitosh) and used the preflight function to do a little testing/converting: 1) Report PDF syntax issues: This Preflight profile executes a low level check for the current document. Any issues it reports point to possible problems on the PDF syntax level. Not all issues reported necessarily mean that the PDF cannot be used successfuly. In some cases it may be possible to resolve the reported issues by doing 'Save as' in Acrobat or by using PDF Optimizer. The information reported may only be meaningful to developers and PDF experts. This returned No problems found. 2) Compatible with Acrobat 4: Makes the current PDF compatible with Acrobat 4 and saves it as a PDF 1.3 document. Among other things flattens layers and transparency and reduces image bit depth from 16 to 8 bit if necessary. According to the output, this did indeed flatten layers and transparency (and inflated the file from 56 KB to 76 KB). I'm sending you the result of the execution of profile 2) (in a private message) so you can test if that modified version works. (You can put that file online if you wish, to let others have a look, too). I could produce other converted files for you (Acrobat 5 format, ...), but I don't think this would help. -M -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problem with grub
Am 2008-05-15 um 11:14 schrieb Gilles Guiot:
Hello all,
I'm a newbie to Linux. I have a Dell server with two raid arrays
(sda and
sdb), both raid 1. I created and LVM on sdb. After having copied the
filesystem from sda onto sdb, I want to configure grub so that it
boots on
sdb.
Could you clarify what your setup is?
Do you have two disks, sda and sdba, which together form *one* RAID1
array, or
do you have two partitions on each of the two disks and use two RAID1
arrays,
or...? I don't quite understand why you only create LVM on sdb, not
on sda,
and what you mean by copying the filesystem from sda to sdb.
You probably mean you have two partitions on disks sda and sdb, one
small
partition and a larger partition. Then you put the two small partitions
together in RAID1 and use it for /boot. You also put the two larger
partitions
together into another RAID1 array and use this for LVM.
And now you want to make sure you can boot your system either from
sda or sdb,
i.e. you can start the system even if one disk has failed, correct?
when i type grub root (sdb,0) , i get the message : syntax error near
unexpected token '('.
Could someone tell me what's my mistake and how to do it properly ?
Do you type grub root (sdb,0) (without the ) all in one line? If
so, you
have misunderstood the instructions! You first need to type grub,
then press
enter, this will get you to the GRUB command, i.e. you will see a
command
prompt of grub at the left. Only then you have to enter root (sdb,
0) and
all the other commands (each followed by pressing enter and without
the ).
-Moritz
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problem with grub
2008-05-15 17:02+0200 Gilles Guiot:
my server has two raid1 arrays, each with two disks, for a total of
four hd. sda1 relates to the biggest partion on the first array
(sda) and sdb1 relates to the only partition on the second raid
array. point is this sdb1 is in lvm so to speak, ie there is an
lv using all of the partition.
2008-05-15 16:43+0200 hh.eu:
I don't quite understand why you only create LVM on sdb, not on sda,
and what you mean by copying the filesystem from sda to sdb.
This is an installed and working server. Because we needed more
space for backuppc, the initial plan was to create an lvm on the
second raid array, boot to it and see if it worked, if so extend it
by incorporating the first raid array.
I am using software RAID, which gives me device names such as md0 and
md1 for
the RAID arrays rather than sda and sdb, so are you using some sort of
hardware RAID (either with a real hardware RAID controller or 'fake
hardware
RAID' using the embedded controller on the motherboard)?
Anyway, if I understand you correctly, you have two partitions on the
old RAID
set and just one partition (with LVM) on the new RAID set. The
problem is
that (legacy) GRUB cannot deal with LVM directly. GRUB can understand
RAID,
but not LVM. That's why people usually have a small RAID1 set on two
disks
that contains just a partition on each disk for /boot, and another
larger
RAID1 set on the same two disks that is entirely filled with LVM.
That's what
I was talking about:
You probably mean you have two partitions on disks sda and sdb,
one small
partition and a larger partition. Then you put the two small
partitions
together in RAID1 and use it for /boot. You also put the two
larger partitions
together into another RAID1 array and use this for LVM.
In other words: You need two partitions on the new RAID set, too. The
size for
the /boot partitions can be quite small, I chose 512 MB which should
be much
more than enough. (I didn't want to make it smaller because hard disk
space is
cheap nowadays and making it bigger isn't so convenient because it's
not on
LVM!).
when i type grub root (sdb,0) , i get the message : syntax error
near
unexpected token '('.
Could someone tell me what's my mistake and how to do it properly ?
Do you type grub root (sdb,0) (without the ) all in one line?
If so, you
have misunderstood the instructions! You first need to type
grub, then press
enter, this will get you to the GRUB command, i.e. you will see a
command
prompt of grub at the left. Only then you have to enter root
(sdb,0) and
all the other commands (each followed by pressing enter and
without the ).
I typed the first grub to enter the grub command, omitted to
specifiy it . Apologies.
I was asking because the error message seems very much like an error
message
from your normal shell, e.g. bash, which you shouldn't get from the GRUB
shell. (I am not in front of a Debian system so I can't test...)
You also need to specify the hard disk name/number in GRUB notation
(which is
something like hd0 or hd1, even for SCSI disks), not in the normal
notation,
i.e. root (sdb,0) doesn't work.
Unfortunately, GRUB uses some sort of guessing to map device names
(e.g. sdb
-- hd0, sda -- hd1) which is not very reliable. See a recent post
from me
for tips and details to find out about the mapping:
http://lists.debian.org/debian-user/2008/05/msg01006.html
Did you read the manual http://www.gnu.org/software/grub/manual/
grub.html?
-Moritz
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
GRUB device numbers/names, use of device.map (Re: migrating Debian GNU/Linux Etch to second SATA drive)
Am 2008-05-10 um 07:01 schrieb Paul Csanyi: [EMAIL PROTECTED] writes: By the way, if you have a floppy drive, you can install GRUB on a floppy too, then you have a GRUB emergency disk which lets you perform operation such as those you described (in the GRUB shell) (for cases of drive failure etc.). -Moritz That shall I to do, or maybe there is an alternative option, say to install GRUB on the USB stick as emergency disk? I haven't done that, but it should also work. Make sure you test if you can actually boot from your USB stick or not because not all machines can boot from USB devices. E.g., I have a computer that should be able to do it according to the manual and all the settings in the BIOS, but after hours of trying I still couldn't get it to work, it's a buggy BIOS. Also, if it works on one of your machines, it may not work on another or computer of a friend who you are trying to help. If you are frequently using the USB stick, you might occasionally format it, erase it etc., so be careful not to destroy your GRUB install on it. Then, I guess, you have to be careful with the GRUB drive numbers (hd0, ...) when you use the USB stick. E.g., depending on whether or not the USB stick is plugegd in or possibly even which controller it is plugged into, the numbers of all drives may be different. Depending on the BIOS, the USB stick might be considered a floppy drive, so it might be fd1 instead of hd1, for example. So you should be very careful when you try to reinstall GRUB, to ensure you are not (for example) overwriting the MBR on a Windows disk. Generally, GRUB uses some sort of *guessing* to assign the drive numbers, so one always has to be careful.[1] (All of the above of course also applies to the case where you use a floppy. But then you will at least usually know that fd0 is your regular floppy drive.) The GRUB shell offers the find command. You can use it to find a certain file you know exists on a particular drive, which will help you find out which device number GRUB uses for that disk. Another tip is to type something like root ( + tab, it will give you a list of possible devices, then complete the name of one device and use tab again, GRUB should then tell you which file system is on that disk which might also help you. Then there is a file named /boot/grub/device.map where one can define which device should have which number, but the use of this file is quite confusing and poorly documented, and it took me many tries to understand. (The GRUB manual and all sorts of search results couldn't help me clearly.) Summary: (1) If GRUB is started from within Debian: * If started with the command grub, the file device.map is ignored. GRUB assigns device names (hd0 etc.) based on *guesses* it makes.[1] * If started with the command grub --device-map=device.map and the file device.map exists, the file device.map is parsed. * If started with the command grub --device-map=device.map and the file device.map does not exist, GRUB *guesses* the device names [1] and stores the result of the guessing in the file device.map. (2) If GRUB is started directly from the BIOS (GRUB shell): * The file device.map is ignored, the device names are derived from GRUB's guessing.[1] Note that (2) is also the situation you have when you boot your system normally. In other words: Making changes to device.map does not influence device numbers actually used by GRUB when booting. (That's why I don't use that file, I personally find no use for it.) -Moritz [1] This guessing can, of course, never be really consistent and leads to all sorts of confusing situations, e.g. different device numbers depending on whether GRUB is started from the BIOS of from within Debian or different numbers after changing cables in the computer. It is one of the fundamental logical flaws in the design of GRUB and one of the reasons GRUB 0.97 is not developed further. (Development effort goes to GRUB 2, currently at version 1.9something, which has been in the works for years and is still not ready for release and is not documented yet, so for most people is not a viable alternative.) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: migrating Debian GNU/Linux Etch to second SATA drive
I am not sure if I understood you correctly, so please correct me if I am wrong, but I assume: * you have basically moved all needed partitions for a full Debian system from one disk to another, * and now you want to install GRUB to this second disk so that GRUB can boot from that disk (and you can eventually remove the old disk from the system). Can you still (temporarily) boot from your first disk where GRUB is still installed? You don't need to be able to boot Debian from the old disk, you just need a working GRUB install there. Does GRUB display a menu when you boot from your old disk? Then you can use the GRUB shell, and within the GRUB shell you can run several commands to install GRUB to the second, new hard disk. You get to the GRUB shell by pressing the c key when GRUB displays the menu. (Maybe you have to press Esc first, I am not sure and can't try it now.) Let me know if this would work for you, then I could tell you more about which commands to run in the GRUB shell. Am 2008-05-08 um 23:12 schrieb Paul Csanyi: Hello! I tried to migrating from the first to second SATA drive My Debian Etch OS with Gparted. I copied the /, /usr, /var, swap, /tmp, /home partitions of the system to the second SATA: sdb with Gparted. Then I tried to set up grub to boot this Debian system from the first SATA drive sda, and did the following: - boot the install CD of the Etch Netinstall - at prompt use expert - ... - detecting disks - did not disk partition - run a shell Here I follow the mini howto: http://wiki.archlinux.org/index.php/Reinstalling_GRUB cd / mount -t ext3 /dev/discs/disc1/part1 /mnt # this is the / partition of the sdb mount -t ext3 /dev/discs/disc1/part2 /mnt/usr # this is the /usr partition of the sdb mount -t proc proc /mnt/proc mount -t sysfs sys /mnt/sys mount -o bind /dev /mnt/dev chroot /mnt /bin/bash grub-install /dev/sda Error: /dev/scsi/host1/bus0/target0/lun0/disc does not have any corresponding BIOS drive. I can't here as root to use nano because of the bterm error. I did the following to get some editor, namely mcedit: TERM=xterm export TERM mcedit /etc/fstab I can to use here mcedit, but with difficulties. Still can to edit fstab, however. But, I can't to install grub to boot sdb root from the sda MBR. Any advices will be appreciated! -- Regards, Paul Csanyi http://www.freewebs.com/csanyi-pal/index.htm -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: migrating Debian GNU/Linux Etch to second SATA drive
Am 2008-05-09 um 22:13 schrieb Paul Csanyi: At last I have success! :D I don't use grub-install command but run grub on the command line, and: grub root (hd1,0) grub setup (hd0) grub quit OK, I hadn't seen this last message from you when I wrote mine, but I see that you have had success with exactly what I was going to suggest to you. I am just wondering if you are aware that you still need your first disk installed to boot the second. So far, GRUB is not installed on the second disk, so if your first disk breaks, you will not (easily) be able to boot from the second disk. I am not sure about what you are doing with those two disks, but maybe it is wise to install GRUB also to the second disk, in case you remove the first one at some point. (Or you could directly boot from the second disk.) By the way, if you have a floppy drive, you can install GRUB on a floppy, too, then you have a GRUB emergency disk which lets you perform operation such as those you described (in the GRUB shell) (for cases of drive failure etc.). -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Kontron 986LCD-M/mITX
I was looking at the mini-ITX board Kontron 986LCD-M/mITX http://emea.kontron.com/products/boards+and+mezzanines/embedded +motherboards/ miniitx+motherboards/986lcdmmitx.html and am wondering if anyone has used this for a Debian system. Are the chipsets supported, or do you recognise any that are not working well/at all? * Intel 945GM Intel ICH7R Embedded Chipset consisting of: + Intel 82945G Graphics and Memory Controller Hub (GMCH) + Intel ICH7R I/O Controller Hub (ICH7R) + 8 Mbit Firmware Hub (FWH) * graphics: Intel GMA950 Controller with 2D/3D Graphics Engine * ethernet: Realtek RTL8111B (with 3 ethernet connectors!) * Winbond W83627THF LPC Bus I/O Controller * Intel 82801G ICH7 USB Universal Host Controller * Intel 82801GR/GH SATA RAID Controller * Intel 82801G ICH7 Ultra ATA Storage Controllers The manual also says: OS support: Linux: Feodora Core 5, Suse 10.01 (limitations may apply) and The SATA controller supports AHCI mode and has integrated RAID functionality with support for RAID modes 0, 1, 5 and 10 (Linux O/S only support for RAID 0 and 1). Any ideas? Thanks, -Moritz P.S.: A Celeron M ULV does NOT have Intel VT-x technology (for virtualisation), right? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Kontron 986LCD-M/mITX / kernel upgrade with etch stable?
Am 02.05.2008 um 11:01 schrieb Sylvain Le Gall: On 02-05-2008, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I was looking at the mini-ITX board Kontron 986LCD-M/mITX http://emea.kontron.com/products/boards+and+mezzanines/embedded +motherboards/ miniitx+motherboards/986lcdmmitx.html and am wondering if anyone has used this for a Debian system. Are the chipsets supported, or do you recognise any that are not working well/at all? * Intel 945GM Intel ICH7R Embedded Chipset consisting of: + Intel 82945G Graphics and Memory Controller Hub (GMCH) + Intel ICH7R I/O Controller Hub (ICH7R) + 8 Mbit Firmware Hub (FWH) * graphics: Intel GMA950 Controller with 2D/3D Graphics Engine * ethernet: Realtek RTL8111B (with 3 ethernet connectors!) * Winbond W83627THF LPC Bus I/O Controller * Intel 82801G ICH7 USB Universal Host Controller * Intel 82801GR/GH SATA RAID Controller * Intel 82801G ICH7 Ultra ATA Storage Controllers The manual also says: OS support: Linux: Feodora Core 5, Suse 10.01 (limitations may apply) and The SATA controller supports AHCI mode and has integrated RAID functionality with support for RAID modes 0, 1, 5 and 10 (Linux O/S only support for RAID 0 and 1). I have this kind of motherboard inside a Hush B3. It takes time at the beginning to work. Most of the part are working except the network (RTL8111B). The solution is to upgrade your kernel to at least 2.6.22.6, to have the working r8169 kernel module. Since network issue is solved, this motherboard is working great. I don't have tested graphics and RAID. Regards, Sylvain Le Gall Thanks a lot for this valuable info, I wouldn't have expected the network to be the biggest problem! I've been seeing different comments on this list recently regarding mixing stable with testing or unstable or backports etc. Generally this should be avoided. Would this mean, since I need a newer kernel than the one in etch stable, that I should not run Etch stable but go for lenny or sid instead? If anyone else has any tips or ideas for problems that might occur with the chipsets, please let me know! As for the RAID: I plan to use software RAID rather than use 'hardware' RAID of the controller. That should be possible, shouldn't it? Thanks again! -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: installing scanner canon
Are you aware of the SANE website and its list of supported devices http://www.sane-project.org/sane-mfgs.html#Z-CANON and the sane-devel mailing list http://lists.alioth.debian.org/mailman/listinfo/sane- devel which might be more appropriate places to consult with your problem? You are referring to sane-find-scanner etc., so you're trying to get the scanner to work using SANE, however your scanner is listed as not supported on the above mentioned list of compatible hardware. The reference on the Ubuntu Portugal page you gave claiming that your scanner works with Ubuntu doesn't list any proof/links/background etc., so seems a bit unreliable, if you ask me. A quick Google search also only reveals results that show that the scanner doesn't work in any other OS than Windows. E.g., a German Ubuntu page specifically lists the device as not supported on its blacklist: http://wiki.ubuntuusers.de/Hardware_blacklist I believe the problem is that your scanner uses a chipset for which no one has written a SANE backend yet. So I am afraid your scanner is (still) unsupported, and your conclusion that after all these years of not being supported it should work now was a bit premature. Most (all?) scanner producers usually only provide binary drivers for Windows. This is what you have on the CD. Of course, unfortunately, these are useless for all other OS like Debian. -Moritz P.S.: I now see you did look at the SANE documentation before. Maybe try the mailing list, but I think they won't be able to help you. Am 27.04.2008 um 14:48 schrieb Shams Fantar: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I'm trying to install my scanner canon canoscan 3000F. Few years ago, I wasn't able because there was no driver for this scanner, now, it exists. And I've found this[1], and the scanner is in this page, so, that confirms what I said. - - Xsane doesn't find my scanner : no devices available. - - # lsusb : Bus 003 Device 006: ID 04a9:2215 Canon, Inc. CanoScan 3000/3000F/3000ex - - # sane-find-scanner found USB scanner (vendor=0x04a9, product=0x2215, chip=GL660 +GL646?) at libusb:003:006 - - My /etc/sane.d/canon.conf[2] I think I have to install the driver of the scanner, but, how to do ? I don't find it in the CD-ROM for installing this scanner (the cd only works under windows, of course but I read that if xsane doesn't find the scanner, we need to look for the driver on the CD-ROM...right ?). And I didn't find information in the sane documentation. [1] : http://www.ubuntupt.org/wiki/index.php? title=Lista_de_scanners_reconhecidos_pelo_Ubuntu_7.04#Canon [2] : http://snurf.info/sfantar/scanner/canon.conf Any ideas? Regards, - -- Shams Fantar (Website : http://snurf.info) My public GPG Key : http://snurf.info/sfantar.gpg « A book is like a garden carried in the pocket. » -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFIFHY55ChwvXmalbURAjWYAJ9h7ZoG/09nEoTljqGSaCwL8Ln3IQCePZ6b aOHNV0+nufglc7+chLUmJps= =l0Gb -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Kontron 986LCD-M/mITX / kernel upgrade with etch stable?
Am 02.05.2008 um 14:09 schrieb Sylvain Le Gall: On 02-05-2008, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Am 02.05.2008 um 11:01 schrieb Sylvain Le Gall: The solution is to upgrade your kernel to at least 2.6.22.6, to have the working r8169 kernel module. I've been seeing different comments on this list recently regarding mixing stable with testing or unstable or backports etc. Generally this should be avoided. Would this mean, since I need a newer kernel than the one in etch stable, that I should not run Etch stable but go for lenny or sid instead? You have 2 solutions: - use r1000 (a driver from realtek for this card) with kernel 2.6.18 in etch For anyone else looking for this (Realtek's website is a bit disorganised and searching for the driver on it took me a while), here's the link: http://www.realtek.com.tw/downloads/downloadsView.aspx? Langid=1PNid=13PFid=5Level=5Conn=4DownTypeID=3GetDown=false - just recompile a newer kernel or use one from backports.org (http://packages.debian.org/etch-backports/linux-image-2.6-686), which is version 2.6.22 OK, good, thanks for clarifying. In the two cases, you doesn't need to mix unstable/stable. I think going to lenny just for the kernel is not a good idea. Yes, that's what I thought, too! As for the RAID: I plan to use software RAID rather than use 'hardware' RAID of the controller. That should be possible, shouldn't it? This is possible and far more better than using fake intel RAID... Performance will be equal and you will have something working better. Great, thanks again for backing up my thoughts and choice against fake hardware RAID! Now I'm researching which case to get. Ideally I want a tiny one but which allows two or more 2,5 HDDs in it. (Perhaps I later also want to install a TV (DVB-T) card in it, too.) So far I haven't found a good one. -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
XFS / file system for backup server (Re: JFS / Unsupported file systems)
Ron Johnson wrote at 2008-04-29 11:34 +0200: On 04/29/08 01:24, Alex Samad wrote: On Mon, Apr 28, 2008 at 09:52:16PM -0500, Ron Johnson wrote: On 04/28/08 19:22, Alex Samad wrote: some reason that xfs is not being talked about? We hate it because Dr. Evil uses it. has this got something to do with elderberries ? A little, but not as much as you think... Hm, some of us, I am sure, have to look up what elderberries are in their language, conclude that they might have something to do with a shrubbery, but still do not quite get what you're talking about... :-) Anyway, are you just joking or are there serious reasons against XFS? I am not very knowledgeable with respect to file systems on GNU/Linux, coming from a different background, and it's a few weeks ago that I made my decision so I don't remember all the details of my reasoning, but I have decided to try XFS for an rsync backup disk, mostly because it seemed to be the best choice for larger files and to hold metadata from different operating/file systems (extended attributes, ...). I haven't implemented this in all detail and am currently testing it in various situation, so could change if you elaborated a bit or gave me pointers to resources on how to choose a good file system for a backup server holding data from several file systems. From this thread, XFS can cause trouble when it is not cleanly unmounted? Thanks! -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Put server to sleep (hibernate/suspend) automatically after inactivity? / problems with suspend via GUI
(Sorry for the double-post, but my previous post to debian-user didn't get any replies, and since this is an energy settings question, I'm hoping the laptop users might be able to help me. Here's my (shortened) previous post: My question boils down to this: Can I tell the server, using a mechanism available when there is no user logged in, to automatically go to sleep (suspend to disk/hibernate) after a certain amount of inactivity? I.e., I am looking for settings such as those available in the GNOME power manager, but to be used for the system in general. I guess one of the problems is the definition of inactivity on a server as there will probably always be things going on due to some services running. Read on for the details and background, which might help in narrowing down criteria for inactivity. These are my requirements: 1) Main use of machine (for now): file server, mostly for rsync backups or storing of archive (files used not regularly). 2) Machine to be run 'headless' (without monitor and physical access, i.e. stored away in the basement). 3) Machine to be put in 'suspend to disk' mode when inactive, to be woken up via magic packet over ethernet (wake-on LAN) when one of the clients needs access (i.e. to perform an rsync backup). My machine is using an AMD Athlon processor (1 GHz) with 768 MB of RAM on an ASUS A7V133 board. I am using two SATA disks on a PCI controller (SiI 3512A) to run a software RAID with LVM for Debian etch on it. I did install a desktop environment (GNOME), too, for various other reasons, and could not get the suspend features of gnome-power-manager to work. I.e. the settings for put computer to sleep when it is inactive for ... do nothing, and I can't manually suspend the system via the GUI, either (the little power plug symbol in the menu in the upper right). I can, however, suspend the machine with sudo s2ram -f or sudo s2disk. I've been searching the internet for solutions to why I can't use the nice interface in the GUI, and am puzzled by the plethora of different power management packages etc. that are available and all the problems associated with them. I have tried several things suggested to make the GUI method work (made sure my user account is in the admin group in /etc/sudoers , added my account to the powerdev group etc.), but nothing helped. When I power up the machine, I don't want an automatic login into a desktop session, instead I usually want to contact the machine over ssh (and occasionally over VNC) or on other ports (rsync daemon). So ideally I don't need to the GUI solution to work, but I thought that maybe an alternative would be to automatically let the system boot into a desktop session under a (dedicated) user account and then use the features of the GUI. The least ideal solution, I guess, would be to let the clients explicitly tell the server I don't need you anymore, go to sleep (over ssh). It's a home server, I don't expect more than one or two clients to be connected simultaneously, for a limited number of purposes such as rsync, and the clients could execute commands via sudo. When given the command to sleep, the machine could check if another user is using the server or if another person is communicating with the the rsync daemon to see it the machine can be put to sleep. Or is all this too unreliable and complicated? Do you have ideas? The purpose, obviously, is to avoid running a 100 W device all the time when usually it is not needed. Thanks for your thoughts! -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mutt + mailings list ( + vim)
Am 21.04.2008 um 23:45 schrieb Alex Samad: so whilst viewing an email, I press shift-l, this starts vim with the emails, I then use up and down arrows and v to highlight some text, I would then like to press some key combo and have the text replaced by cr[snip]cr I haven't used this before, but try putting something like this into your ~/.vimrc file: map C-S-F1 dEsciCR[snip]CREsc This will create a shortcut for Ctrl+Shift+F1 (think of a better one, I used this for testing to make sure I didn't interfere with the defaults). You make a selection in vim, i.e. press Shift+v, then go down a few lines, and then press the shortcut, which will delete the selection (d), go into insert mode (i) and put in your text and then leave insert mode again (Esc). So actually it's pretty easy. Type :help key-mapping in vim for more info. You'll have to play around a bit to see how you want things when you don't select full lines. -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Put server to sleep (hibernate/suspend) automatically after inactivity?
Hello experts, I've been setting up my first real (albeit small) home server on Debian (etch) in the past days, and I have some problems regarding the implementation of my specific goals. My question boils down to this: Can I tell the server, using a mechanism available when there is no user logged in, to automatically go to sleep (suspend to disk/hibernate) after a certain amount of inactivity? I.e., I am looking for settings such as those available in the GNOME power manager, but to be used for the system in general. (Let me know if the people on debian-laptop might the ones to ask here, I din't want to cross-post to two lists.) I guess one of the problems is the definition of inactivity on a server as there will probably always be things going on due to some services running. Read on for the details and background, which might help in narrowing down criteria for inactivity. These are my requirements: 1) Main use of machine (for now): file server, mostly for rsync backups or storing of archive (files used not regularly). 2) Machine to be run 'headless' (without monitor and physical access, i.e. stored away in the basement). 3) Machine to be put in 'suspend to disk' mode when inactive, to be woken up via magic packet over ethernet (wake-on LAN) when one of the clients needs access (i.e. to perform an rsync backup). My machine is using an AMD Athlon processor (1 GHz) with 768 MB of RAM on an ASUS A7V133 board. I am using two SATA disks on a PCI controller (SiI 3512A) to run a software RAID with LVM for Debian etch on it. I did install a desktop environment (GNOME), too, for various other reasons, and could not get the suspend features of gnome-power-manager to work. I.e. the settings for put computer to sleep when it is inactive for ... do nothing, and I can't manually suspend the system via the GUI, either (the little power plug symbol in the menu in the upper right). I can, however, suspend the machine with sudo s2ram -f or sudo s2disk. I've been searching the internet for solutions to why I can't use the nice interface in the GUI, and am puzzled by the plethora of different power management packages etc. that are available and all the problems associated with them. I have tried several things suggested to make the GUI method work (made sure my user account is in the admin group in /etc/sudoers , added my account to the powerdev group etc.), but nothing helped. When I power up the machine, I don't want an automatic login into a desktop session, instead I usually want to contact the machine over ssh (and occasionally over VNC) or on other ports (rsync daemon). So ideally I don't need to the GUI solution to work, but I thought that maybe an alternative would be to automatically let the system boot into a desktop session under a (dedicated) user account and then use the features of the GUI. The least ideal solution, I guess, would be to let the clients explicitly tell the server I don't need you anymore, go to sleep (over ssh). It's a home server, I don't expect more than one or two clients to be connected simultaneously, for a limited number of purposes such as rsync, and the clients could execute commands via sudo. When given the command to sleep, the machine could check if another user is using the server or if another person is communicating with the the rsync daemon to see it the machine can be put to sleep. Or is all this too unreliable and complicated? Do you have ideas? The purpose, obviously, is to avoid running a 100 W device all the time when usually it is not needed. Thanks for your thoughts! -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Configure grub for Windows XP on second HDD
Am 19.04.2008 um 17:54 schrieb Michael Wilson: I have Linux on hd0 and Windows XP Home on hd1 (I installed XP with the Linux disk disconnected so Windows couldn't mess with it). If I configure the BIOS to make the Windows drive the first, Windows boots fine, but I can't figure out how to configure GRUB to boot windows when the Linux disk is first. At the moment, at the end of my menu.lst, I have: title Windows XP map (hd0) (hd1) map (hd1) (hd0) rootnoverify(hd0,0) chainloader +1 makeactive Which is close as I can get, but still no cigar. Can anyone help? Try rootnoverify (hd1,0) instead of rootnoverify (hd0,0), that should work. As others have noted, searching the web gives you fuzzy results as a lot of people have typos or don't quite understand how GRUB works (it's pretty hard and at times unpredictable!), and the example configuration given in GRUB's official documentation regarding this toping at http://www.gnu.org/software/grub/manual/grub.html#DOS_002fWindows is more confusing than helpful, in my mind. Anyway, in the rootnoverify command, you always use the hard disk number (i.e. hd0 or hd1 or hd2 etc.) according to the order given by the BIOS, no matter if you use the map command or not. GRUB always uses the order of disks it gets from the BIOS. If you have two ATA drives in your computer labelled A and B and tell the BIOS to try to boot from A before B, the A disk will correspond to hd0 and the B one to hd1. If you tell the BIOS to boot from B disk before A disk, it will be the other way round (hd0 for B, hd1 for A). You then tell GRUB from which device to boot. If it is a system GRUB natively supports (such as Debian), you use the root command. If it is a 'foreign' system (such as Windows, but also others), you use rootnoverify. Windows is, again, special in that it wants to be on the first drive only, so you have to use the map commands as in your example. However, this doesn't affect the syntax in rootnoverify, so you still need to use rootnoverify (hd1,0) if Windows is on the second drive (according to BIOS order). As far as I know, you could even reverse the order of the rootnoverify and the map commands, i.e.: title Windows XP rootnoverify(hd1,0) map (hd0) (hd1) map (hd1) (hd0) makeactive chainloader +1 I think that is what I did (don't have access to the file at the moment) because this appears more logical when reading the file later. -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: wakeonlan
Am 19.04.2008 um 22:51 schrieb Paul Csanyi: Andrei Popescu [EMAIL PROTECTED] writes: On Sat, Apr 19, 2008 at 09:41:14PM +0200, Paul Csanyi wrote: When I run the command wakeonlan the server box wan't not to wake up. I never used wake-on-lan, but I see from the description of the wakeonlan package that it used UDP packets in order to avoid the need for root. Maybe you should also try etherwake? Maybe I don't understand the description of wakeonlan: You can wake up WOL compliant Computers which have been powered down to sleep mode or start WOL compliant Computers with a BIOS feature. Well, if I power down the server with shutdown -t 1 -h now then I can't to wake it up with wakeonlan? Have you tried suspending (hibernating) the system and then waking it up? If I understand you correctly, you are trying to power up a system that is not in suspend-to-disk or suspend-to-ram mode, but off. I have never used a system with a NIC that actually has a special cable for Wake-on LAN (my cheap NICs were always regular PCI ones without that connection) and I haven't read the docs for wakeonlan, so I am not sure if that makes a difference, but I always thought (and verified by experiment for my hardware) that I needed to suspend the system for WOL to work; starting a machine which had simply been powered off with the shutdown command never worked. Again, maybe this is possible with certain hardware, but it would certainly be special. -Moritz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
