Re: apt pining iceweasel και icedove απο experimental
On 19/04/2013 09:04 πμ, Klearchos-Angelos Gkountras wrote: επειδή θέλω να έχω τον τελαυταια εκδοση αυτων των προγραμματών απο το experimental τι πρέπει να βάλω στο /etc/apt/preferences ? ευχαριστώ εκ των προτέρων Με μια επιφύλαξη για το θεωρητικό υπόβαθρο: Δεν χρειάζεται να φτιάξεις /etc/apt/preferences ή να βάλεις κάτι μέσα σε αυτό, γιατί από το experimental δεν τραβιούνται by default πράματα, ούτε κι αν δώσεις aptitude full-upgrade. Οπότε χρειάζεται απλώς στο /etc/apt/sources.list : deb http://ftp.gr.debian.org/debian experimental main Και μετά apt-get update apt-get install -t experimental iceweasel icedove Εγώ έτσι το κάνω επί καιρό και τώρα έχω 17.0.5 icedove και 20.0 iceweasel. cheers! G. Y.Γ. Eπειδή γενικώς προτιμώ το aptitude, ξέρει κανείς την δεύτερη εντολή πως την γράφω σε aptitude; smime.p7s Description: S/MIME Cryptographic Signature
Re: Ανακοινωση ελληνικης κοινοτητας KDE και 3 KDE SC 4.8 Release Parties
On 25/01/2012 08:06 μμ, Thomas Kapoulas wrote: Τι είναι το KDE; Eίναι Windows for Linux. smime.p7s Description: S/MIME Cryptographic Signature
Re: bug στον installer για RAID partitions;
On 17/01/2012 02:01 μμ, Gregory Potamianos wrote: 2012/1/16 Γιώργος Πάλλας gp...@ccf.auth.gr: Καλησπέρα! Έχω το εξής πρόβλημα που μοιάζει λίγο με bug. [...] Το πρόβλημα είναι ότι δεν καταφέρνω να φτιάξω τα partitions που θέλω: Αν πατήσω στο RAID5 device #0 ... δεν γίνεται τίποτα. Αν πατήσω στο #1 που υποτίθεται είναι ένα partition πάνω στο raid device (ποιος του είπε να το φτιάξει; - στην πραγματικότητα από την κονσόλα βλέπω δεν έχει φτιάξει κανένα partition), τότε μου ζητάει να ορίσω χρήση. Άρα μένω κλειδωμένος Μάλλον έχεις μπερδευτεί με τον τρόπο που δουλεύει το mdraid. Το #1 είναι το partition σου, εκεί πάνω μπορείς να βάλεις ext3,4 κλπ. Το mdraid δουλεύει ανά partition. Αν θές να φιτάξεις πολλαπλά partitions πάνω στο raid είτε θα φτιάξεις πολλά md devices, είτε όπως είσαι μέχρι εκείνο το σημείο θα χρησιμοποιήσεις το md0 για lvm physical volume και θα κόψεις τα volumes σου επάνω σε αυτό. Σε ευχαριστώ για την απάντηση. Όμως δεν πείστηκα! Για δυο λόγους: Α. είμαι 91% σίγουρος ότι το mdraid δημιουργεί ένα block device που είναι ακριβώς όπως όλα τα block devices· άρα μπορεί είτε να γίνει φορμάτ με κάποιο fs, είτε να γίνει partitioned και να πάρει σε κάποιο partition fs. B. Το ότι κατάφερα από command line να δημιουργήσω με fdisk τα /dev/md0p1 και /dev/md0p2 και να βάλω στο ένα /boot και στο άλλο το υπόλοιπο σύστημα και να μπουτάρει κανονικά αποδεικνύει ότι έστω κι αν όπως λες δεν είναι ο ενδεδειγμένος τρόπος, εντούτοις λειτουργεί. Γενικώς ξέρω ότι θα μπορούσα να κάνω τη δουλειά ρίχνοντας lvm πάνω στο ένα partition του mdraid, αλλά ήθελα να παίξω με τον installer. G. smime.p7s Description: S/MIME Cryptographic Signature
iSCSI + LUKS over insecure network
A tough one (for me)! I use iSCSI (with CHAP authentication) to get a remote device over an insecure network, then I unlock the LUKS volume and finally I mount the ext4 FS. How (in)secure is that? Data I miss: 1. CHAP encrypts the iSCSI authentication password, but the actual iSCSI data go over the link unencrypted obviously, yes? 2. When I unlock the LUKS volume using a key file, this key file is transmitted over the link, or not? 3. The actual ext4 data go over the link encrypted or not? My pretty educated guesses are: 2. it does not get transmitted, 3. the data data is transmitted encrypted 1. yes, but we don't care because of 2. and 3. any idea how things really are? thanks! G. Pallas smime.p7s Description: S/MIME Cryptographic Signature
iSCSI + LUKS
Hi to all! I searched the web for finding an answer but with no luck. Anyone got an idea about how to first make iSCSI see a remote disk, then LUKS open the LUKS device and finally the filesystem on LUKS gets mounted. The problem now is that first LUKS noearly tries to create the /dev/mapper device but iSCSI has not run yet. Then iSCSI runs and the system sees the remote disk, and last, fstab tries to mount the filesystem, which of course fails because the LUKS /dev/mapper device is not there. How can I make the order: iSCSI, LUKS, fstab ? smime.p7s Description: S/MIME Cryptographic Signature
keyboard (almost) stops working
Hi to all! I'm using squeeze with KDE, fully updated, on a i386, and since about six months a strange thing happens, at least once a day: The keyboard stops functioning. I thought it gets kind of completely disconnected, but then I noticed that if I press continuously a key, the repeat mode starts (i.e. pressing once 'f' gets me nothing, but pressing it continuously, gets me ''). Lame workaround: Logout from KDE, and login again. (as soon as I log out, it types properly again on the login screen) Disconnecting the keyboard and connecting it to another USB port, does not work. Any ideas? smime.p7s Description: S/MIME Cryptographic Signature
sum two disks using RAID or LVM?
Hi all! A short question: I have two 500GB HDDs and I want to use their space as one. Should I aggregate them with software RAID, or using LVM in regard to data safety? If one of them malfunctions, is one of the two approaches better? smime.p7s Description: S/MIME Cryptographic Signature
Re: sum two disks using RAID or LVM?
On 18/03/2011 10:30 πμ, Michael Thompson wrote: 2011/3/18 Γιώργος Πάλλας gp...@ccf.auth.gr: Hi all! A short question: I have two 500GB HDDs and I want to use their space as one. Should I aggregate them with software RAID, or using LVM in regard to data safety? If one of them malfunctions, is one of the two approaches better? Mirrored raid would be the tool for this. LVM will not help should one of the disks fail The 'data safety' I mentioned gave the wrong impression. Let me rephrase: I want to use the two disks aggregated (i.e. 1TB), and the question is if one of the two solutions gives me better chances to get some of my data back, in case of a full or partial failure of one of the two disks... smime.p7s Description: S/MIME Cryptographic Signature
AC adapter state
Hello to all! I have a DELL laptop running debian squeeze where I cannot find whether it is connected on AC power or not. On an EEE laptop, I can check that using the info in the file /proc/acpi/ac_adapter/AC0/state, but such file does not exist on the DELL laptop. But KDE shows me correctly the relevant info on the battery state icon. Where does it take its info from? Thanks! Giorgos smime.p7s Description: S/MIME Cryptographic Signature
AC adapter state
Hello to all! I have a DELL laptop running debian squeeze where I cannot find whether it is connected on AC power or not. On an EEE laptop, I can check that using the info in the file /proc/acpi/ac_adapter/AC0/state, but such file does not exist on the DELL laptop. But KDE shows me correctly the relevant info on the battery state icon. Where does it take its info from? Thanks! Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: AC adapter state
On 21/02/2011 03:32 μμ, Martin Kraus wrote: On Mon, Feb 21, 2011 at 12:40:19PM +0200, Γιώργος Πάλλας wrote: Hello to all! I have a DELL laptop running debian squeeze where I cannot find whether it is connected on AC power or not. On an EEE laptop, I can check that using the info in the file /proc/acpi/ac_adapter/AC0/state, but such file does not exist on the DELL laptop. But KDE shows me correctly the relevant info on the battery state icon. Where does it take its info from? Hi. If you have powermgmt-base installed (and you probably have because pm-utils needs it), you can use on_ac_power script. it's written in shell so you can look at it to see how the ac state is determined if you are interested. mk Thanks! I found it: /sys/class/power_supply/ADP1/online smime.p7s Description: S/MIME Cryptographic Signature
current state of DFS support for cifs ??
Hi! I'm running debian squeeze, fully updated. I can't seem to find how to mount a cifs share which has DFS enabled. I read that after 2.6.26, DFS support for cifs was streamlined into the kernel, yet I can't get it to work: //myserver/dfs/media/shares/dfs cifs iocharset=utf8,uid=1000,gid=1000,credentials=/home/gpall/Documents/samba_credentials.txt 00 It gets mounted, I can see the root of it with its contents, but after that, trying to list any directory further gets me: aris:/home/gpall# ls -la /media/shares/dfs/winHome ls: cannot access /media/shares/dfs/winHome: Resource temporarily unavailable Any ideas? smime.p7s Description: S/MIME Cryptographic Signature
Re: current state of DFS support for cifs ??
On 09/12/2010 01:00 μμ, Stephan Seitz wrote: On Thu, Dec 09, 2010 at 11:39:05AM +0200, Γιώργος Πάλλας wrote: mount a cifs share which has DFS enabled. I read that after 2.6.26, DFS support for cifs was streamlined into the kernel, yet I can't get it to work: Do you have installed the package keyutils? It is needed for DFS. Shade and sweet water! Stephan worked! I only found mentioning of keyutils at a blog post, but nowhere in the documentation... Anyway... thanks for the solution! Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Virtualbox crashes X server after update
Good morning! After today's regular update of my debian squeeze box where kernel was updated (2.6.32-5-amd), if I open up the Virtualbox console and just click on Settings, it crashes the X server and I get the login prompt! Anybody with something similar? smime.p7s Description: S/MIME Cryptographic Signature
Re: Virtualbox crashes X server after update
On 10/20/2010 09:24 AM, Γιώργος Πάλλας wrote: Good morning! After today's regular update of my debian squeeze box where kernel was updated (2.6.32-5-amd), if I open up the Virtualbox console and just click on Settings, it crashes the X server and I get the login prompt! Anybody with something similar? Noticed now that iceweasel also crashes KDE. Right click and 'properties' on a link at the bookmarks toolbar crashes it... I created another user and copied iceweasel's profile, and it doesn't crash. Here are the logs from .xsession-errors. Something almost same happens when it crashes by VirtualBox... This renders the system pretty unusable, so please if anyone has an idea... kdeinit4: preparing to launch /usr/bin/iceweasel kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, initial: 2015729 kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, ASN: 2015729 kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, final: 'ID: 56623165 ;WMCLASS: iceweasel : navigator ;Caption: Iceweasel ' : 2015729 kwin(3557) KWin::Workspace::allowClientActivation: Activation: No client active, allowing kwin(3557) KWin::Workspace::updateClientArea: screens: 1 desktops: 2 kwin(3557) KWin::Workspace::updateClientArea: Done. kwin(3557) KWin::Workspace::allowClientActivation: Activation: Belongs to active application kwin(3557) KWin::Workspace::allowClientActivation: Activation: Belongs to active application kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, initial: 2020777 kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, ASN: 2020777 kwin(3557) KWin::Client::readUserTimeMapTimestamp: User timestamp, final: 'ID: 56624263 ;WMCLASS: iceweasel : dialog ;Caption: ' : 2020777 kwin(3557) KWin::Workspace::allowClientActivation: Activation: Belongs to active application kwin(3557) KWin::Workspace::updateClientArea: screens: 1 desktops: 2 kwin(3557) KWin::Workspace::updateClientArea: Done. kmix: Fatal IO error: client killed kdeinit4: Fatal IO error: client killed kdeinit4: sending SIGHUP to children. kglobalaccel(3559) GlobalShortcutsRegistry::unregisterKey: Unregistering key Volume Up for kmix : increase_volume kglobalaccel(3559) GlobalShortcutsRegistry::unregisterKey: Unregistering key Volume Mute for kmix : mute snipmany of the same style/snip kglobalaccel(3559) GlobalShortcutsRegistry::unregisterKey: Unregistering key Volume Down for kmix : decrease_volume kglobalaccel(3559) GlobalShortcutsRegistry::unregisterKey: Unregistering key Ctrl+Alt+V for klipper : show_klipper_popup klauncher: Exiting on signal 1 kdeinit4: sending SIGTERM to children. kdeinit4: Exit. kdeinit4: Fatal IO error: client killed kdeinit4: sending SIGHUP to children. KCrash: Application 'ksmserver' crashing... kded4: Fatal IO error: client killed Warning: connect() failed: : No such file or directory KCrash cannot reach kdeinit, launching directly. kglobalaccel: Fatal IO error: client killed kwin: Fatal IO error: client killed ProcessControl: Application '/usr/bin/nepomukservicestub' stopped unexpected (Process crashed) Application '/usr/bin/nepomukservicestub' crashed. No restart! ProcessControl: Application '/usr/bin/nepomukservicestub' stopped unexpected (Process crashed) Application '/usr/bin/nepomukservicestub' crashed. No restart! ProcessControl: Application '/usr/bin/nepomukservicestub' stopped unexpected (Process crashed) Application '/usr/bin/nepomukservicestub' crashed. No restart! ProcessControl: Application '/usr/bin/nepomukservicestub' stopped unexpected (Process crashed) Application '/usr/bin/nepomukservicestub' crashed. No restart! ProcessControl: Application '/usr/bin/nepomukservicestub' stopped unexpected (Process crashed) Application '/usr/bin/nepomukservicestub' crashed. No restart! NepomukServer(3586)/kdecore (KConfigSkeleton) KCoreConfigSkeleton::writeConfig: No protocol specified drkonqi: cannot connect to X server :0.0 kdeinit4: sending SIGTERM to children. kdeinit4: Exit. smime.p7s Description: S/MIME Cryptographic Signature
X server crashes after update of squeeze (was Re: Virtualbox crashes X server after update)
On 20/10/2010 10:45 πμ, Γιώργος Πάλλας wrote: On 10/20/2010 09:24 AM, Γιώργος Πάλλας wrote: Good morning! After today's regular update of my debian squeeze box where kernel was updated (2.6.32-5-amd), if I open up the Virtualbox console and just click on Settings, it crashes the X server and I get the login prompt! Anybody with something similar? Noticed now that iceweasel also crashes KDE. Right click and 'properties' on a link at the bookmarks toolbar crashes it... I created another user and copied iceweasel's profile, and it doesn't crash. Here are the logs from .xsession-errors. Something almost same happens when it crashes by VirtualBox... This renders the system pretty unusable, so please if anyone has an idea... Noticed that stopping desktop effects and suspending compositing eliminates the problem. If I find more details I'll keep posted, in case a bug has to be opened... smime.p7s Description: S/MIME Cryptographic Signature
Re: Partitioning a drive with Windows 7 already installed
On 14/10/2010 08:54 πμ, Tom H wrote: 2010/10/14 Γιώργος Πάλλας gp...@ccf.auth.gr: On 14/10/2010 04:43 πμ, Tom H wrote: You can mess with the registry to make another set of recovery disks if the first set is damaged or unduplicated This is really unneeded. What I do on friend's laptops with pre-installed Win7 and no recovery disks (just a recovery partition) who want to install ubuntu or some other linux distro is this: I boot Clonezilla, I open a shell, I delete the recovery partition (using fdisk/cfdisk) and then I take a good compressed disk snapshot on an external usb disk. Then, install debian/ubuntu on the 10Gigs of the recovered space of the ex-recovery partition. And when you sell the box to a Windows-only user, .. ... you also sell him support for when the bad time comes :-P smime.p7s Description: S/MIME Cryptographic Signature
Re: Partitioning a drive with Windows 7 already installed
On 13/10/2010 02:07 μμ, steef wrote: Γιώργος Πάλλας schreef: On 13/10/2010 01:24 πμ, Ogya Chief wrote: If you care about Windows 7 at all (e.g., for playing games) you should back up the Win 7 installation completely before starting. How you go about it depends on your new machine. If it came with recovery options, you should investigate them. With the Ultimate edition, Win 7 comes with a good backup program, you can use that. Otherwise, you can look to the various alternatives for backing up. But I urge you to use something. Most of the programs, free and non-free, work properly to shrink a partition, but it is easy to commit a user error and wipe things out completely. If it were me, I would shrink (resize) the partition first, before running the Debian installer. That is, take it in reversible stages. This is a matter of taste. It isn't strictly necessary. I have Win 7 home premium. I will find out what tools for partitioning/backup came with it and decide on my next course of action. For backing up the whole disk before starting with dangerous operations I suggest Clonezilla (the open-source equivalent of Ghost). It use it regularly and hasn't betrayed me. As for the rest procedures, I also suggest you resize the windows partition with gparted from a ubuntu live cd, and then run the debian installer. could dd_rescue be a suggestion? steef dd_rescue for backup? Isn't it a bit low-level for the simple use of backing up a disk? Clonezilla also does recognize the filesystems and takes up space only for used disk space... smime.p7s Description: S/MIME Cryptographic Signature
Re: Partitioning a drive with Windows 7 already installed
On 14/10/2010 04:43 πμ, Tom H wrote: On Wed, Oct 13, 2010 at 9:11 PM, Mark Allums m...@allums.com wrote: On 10/13/2010 6:43 PM, Tom H wrote: On Wed, Oct 13, 2010 at 9:56 AM, Preston Boyington preston.li...@gmail.com wrote: Ogya Chief wrote: At this stage there is no data to backup. If there is any other thing I can backup, please let me know. Since most newer computers don't come with a Restore disc, I would suggest burning the Restore partition and any associated utility partition to a DVD. Usually it fits on one. That way if your hard drive goes belly up later you can slap a new one in and put the machine back to a factory state if you wish. Don't all boxes ship with a utility to create restore DVDs from the restore partition?! Often, they do, yes. Typically, it allows exactly one copy to be created, so, ensure that good media is used, and follow instructions to the letter. Once that copy is attempted, no more are allowed. You can mess with the registry to make another set of recovery disks if the first set is damaged or unduplicated This is really unneeded. What I do on friend's laptops with pre-installed Win7 and no recovery disks (just a recovery partition) who want to install ubuntu or some other linux distro is this: I boot Clonezilla, I open a shell, I delete the recovery partition (using fdisk/cfdisk) and then I take a good compressed disk snapshot on an external usb disk. Then, install debian/ubuntu on the 10Gigs of the recovered space of the ex-recovery partition. smime.p7s Description: S/MIME Cryptographic Signature
Re: Partitioning a drive with Windows 7 already installed
On 13/10/2010 01:24 πμ, Ogya Chief wrote: If you care about Windows 7 at all (e.g., for playing games) you should back up the Win 7 installation completely before starting. How you go about it depends on your new machine. If it came with recovery options, you should investigate them. With the Ultimate edition, Win 7 comes with a good backup program, you can use that. Otherwise, you can look to the various alternatives for backing up. But I urge you to use something. Most of the programs, free and non-free, work properly to shrink a partition, but it is easy to commit a user error and wipe things out completely. If it were me, I would shrink (resize) the partition first, before running the Debian installer. That is, take it in reversible stages. This is a matter of taste. It isn't strictly necessary. I have Win 7 home premium. I will find out what tools for partitioning/backup came with it and decide on my next course of action. For backing up the whole disk before starting with dangerous operations I suggest Clonezilla (the open-source equivalent of Ghost). It use it regularly and hasn't betrayed me. As for the rest procedures, I also suggest you resize the windows partition with gparted from a ubuntu live cd, and then run the debian installer. smime.p7s Description: S/MIME Cryptographic Signature
how to convert ext3/4 system to btrfs
Ένα mini-howto για όποιον ενδιαφέρεται να μετατρέψει το σύστημά του σε BTRFS που για experimental παίζει αξιοσημείωτα σταθερά! Εμένα πάντως δυο συστήματα που τα τρέχω σε btrfs παίζουν μια χαρά εδώ και κάτι μήνες. -- Υποτίθεται ότι πρέπει το /boot να είναι ξεχωριστό partition 1. Βοοτάρεις το παλιό ext4 σύστημα και εγκαθιστάς btrfs-tools, joe (ή τον αγαπημένο σου editor) 2. Booτάρεις με κάποιο live cd (θα συνιστούσα κάποιο debian live: http://cdimage.debian.org/cdimage/squeeze_live_alpha2/i386/net/) 3. fsck -f /dev/sdX (sdX το υπό μετατροπή root fs) 4. βάζεις joe και btrfs-tools και στο live cd 5. btrfs-convert /dev/sdX 6. mount /dev/sdX /mnt 7. κάνεις populate τα /mnt/proc, /mnt/dev 8. chroot /mnt 9. mount /dev/sdY /boot (sdY το boot partition που θα μείνει ext3) (το /boot είναι τώρα στο chrooted environment 10. joe /etc/fstab και αλλάζεις το ext4 του root partition με btrfs, ενώ τα mount options τα κάνεις defaults,compress (το compress προαιρετικά). Επίσης αντικαθιστάς τα UUIDs (root part + boot part) με κλασσικά /dev/sdXY. 11. mkinitramfs -o /boot/initrd.img-lalala (του τρέχοντος kernel) [ αν όμως το live cd δεν έχει την ίδια έκδοση kernel, τότε πρέπει να γρααφτεί κάτι του στυλ mkinitramfs 2.6.32-5-686 -o /boot/initrd.img-2.6.32-5-686 ] 11.5 πιθανότατα θα χρειαστεί (κάντο, δεν πειράζει και αν δεν χρειαζόταν) ένα grub-install /dev/sdX (προσοχή Το device name εδώ, άνευ partition number! Óxi /dev/sda1 αλλά /dev/sda !!!) 12. update-grub 13. umount /boot 14. exit (βγαίνεις από το chrooted env) 15. umount /mnt/proc, umount /mnt/dev, umount /mnt 16. reboot and pray μετά, αφού σίγουρα η επιχείρηση στέφθηκε με επιτυχία και έχουμε bootάρει στο νέο btrfs σύστημά μας, κάνουμε ένα update-initramfs -t -u -kall για όχι τελείως διευκρινισμένους λόγους (γενικώς για να ξέρει το σύστημα ότι έχει το ίδιο φτιάξει το τρέχον initrd image, και όχι εμείς) Kαι τρέξτε και ένα update-grub για ξορκιστεί το κακό. Μετά, αφού έχετε σιγουρέψει όλη τη δουλειά, μπορείτε να ξανα-αντικαταστήσετε τα ονόματα των partitions στο /etc/fstab με UUIDs. smime.p7s Description: S/MIME Cryptographic Signature
Re: cifs share is not mounted during boot
On 10/06/2010 09:52 PM, Eero Volotinen wrote: 2010/10/6 Γιώργος Πάλλας gp...@ccf.auth.gr: On 10/06/2010 06:50 PM, Pol Hallen wrote: //myserver/myData/data1 /media/cifs-shares/dat1 cifs defaults,user,auto,noserverino,uid=1000,gid=1000,credentials=/home/gpall/Do cuments/samba_credentials.txt 0 mount by shell with these options is success? Po Mounting by shell using these options succeeds, as well as with mount -a. As I wrote in another reply, I'm baffled by the fact that on a freshly installed system, auto mounting during boot works... Nice problem, maybe you can hack some debug info into init script that is mounting cifs on boot? Did that and I think I found the root of the problem. Look: A. freshly installed debian /etc/default/rcS - ASYNCMOUNTNFS=yes (default) init-+-ifup---sh---run-parts---mountnfs---mount---mount.cifs (OK, cifs share mounted) /etc/default/rcS - ASYNCMOUNTNFS=no init-+-init---rc---startpar---startpar---mountnfs.sh---mountnfs---mount---mount.cifs (no network still - ok, logical) B. my machine /etc/default/rcS - ASYNCMOUNTNFS=yes (default) mount.cifs never gets called!!! /etc/default/rcS - ASYNCMOUNTNFS=no init---rc---startpar-+-startpar--mountnfs.sh---mountnfs---mount---mount.cifs (no network still - also logical) I experimented with ASYNCMOUNTNFS but found that it should be left alone to its default - YES. So, at the bottom line, /sbin/ifup which eventually mounts cifs shares never gets called at my machine. And then it occured to me: I'm using wicd which is known (http://goo.gl/8Ysf) to not call ifup scripts. So I'm either going to get rid of wicd, or get wicd scripts options call mount -a... smime.p7s Description: S/MIME Cryptographic Signature
Re: creation of encrypted lvm, suggestions please
On 10/06/2010 08:17 AM, vishnu vardhan wrote: [3] volume group volume group name not found. eventhough i am able to access the partitions, at the boot time it is showing the above message. can i assume the partition scheme is correct ? You mean that you see that as the very first message after grub? That is, you see this message before you are asked for the passphrase? If yes, it is normal. For some reason debian looks for the volume groups even before unlocking the LUKS partition and of course fails. After entering the passphrase, it looks again. I have once entered a bug report for this, as well as for the red letter fail when such a machine shuts down because it cannot close the logical volume which contains the root filesystem. Both were answered that's the way its supposed to work... smime.p7s Description: S/MIME Cryptographic Signature
Re: cifs share is not mounted during boot
On 10/06/2010 08:51 AM, Eero Volotinen wrote: 2010/10/6 Γιώργος Πάλλας gp...@ccf.auth.gr: On 10/05/2010 05:11 PM, Klistvud wrote: Dne, 05. 10. 2010 13:51:39 je Eero Volotinen napisal(a): try mount using ip address ? Are you sure that network is available that time? or add @reboot mount -a on cron crontab ? -- Eero Actually, Eero's answer got me thinking that, at least in your case, perhaps a _netdev mount option in your stab could help? With _netdev option I just get in the dmesg logs: [ 141.008430] CIFS: Unknown mount option _netdev and the share is still not mounted. Trying with just the IP instead of the dns name, also failed. Then, I tried something more radical: I replaced /sbin/mount.cifs with an ad-hoc shell script that I made, which just issues an echo test /tmp/test. So, after boot, I checked for a /tmp/test file but found none. So, it looks like it's not got to do with some bad option, or the network not being available or the credential file not being available or something like that, but actually the binary used to mount cifs shares is not even called during boot. Check from initscripts the way debian mounts shares? /tmp might not be available at that time? I also tried with writing to /root which is on the root filesystem. Still no log... Also, I set up an ad-hoc VM where I tried the same cifs entry in fstab and it works! I'm baffled! smime.p7s Description: S/MIME Cryptographic Signature
Re: cifs share is not mounted during boot
On 10/06/2010 06:50 PM, Pol Hallen wrote: //myserver/myData/data1 /media/cifs-shares/dat1 cifs defaults,user,auto,noserverino,uid=1000,gid=1000,credentials=/home/gpall/Do cuments/samba_credentials.txt 0 3 mount by shell with these options is success? Pol Mounting by shell using these options succeeds, as well as with mount -a. As I wrote in another reply, I'm baffled by the fact that on a freshly installed system, auto mounting during boot works... smime.p7s Description: S/MIME Cryptographic Signature
cifs share is not mounted during boot
Hello to all! I have this cif share in my /etc/fstab which is not mounted during boot and I can't figure out why: //myserver/myData/data1 /media/cifs-shares/dat1 cifs defaults,user,auto,noserverino,uid=1000,gid=1000,credentials=/home/gpall/Documents/samba_credentials.txt 0 3 After boot, giving a mount -a as root mounts it without problem. It is not a matter of the credential file not being available during boot: Even with the credentials supplied in /etc/fstab directly the share is not mounted and nothing ever get logged. I'm running updated debian squeeze and would appreciate some help! smime.p7s Description: S/MIME Cryptographic Signature
Re: cifs share is not mounted during boot
On 10/05/2010 05:11 PM, Klistvud wrote: Dne, 05. 10. 2010 13:51:39 je Eero Volotinen napisal(a): try mount using ip address ? Are you sure that network is available that time? or add @reboot mount -a on cron crontab ? -- Eero Actually, Eero's answer got me thinking that, at least in your case, perhaps a _netdev mount option in your stab could help? With _netdev option I just get in the dmesg logs: [ 141.008430] CIFS: Unknown mount option _netdev and the share is still not mounted. Trying with just the IP instead of the dns name, also failed. Then, I tried something more radical: I replaced /sbin/mount.cifs with an ad-hoc shell script that I made, which just issues an echo test /tmp/test. So, after boot, I checked for a /tmp/test file but found none. So, it looks like it's not got to do with some bad option, or the network not being available or the credential file not being available or something like that, but actually the binary used to mount cifs shares is not even called during boot. Any ideas? smime.p7s Description: S/MIME Cryptographic Signature
Re: [News] Flash player plugin (64 bits) comes back!
snip At least this new x64 flashplugin is more stable than the 32bit one under nsplugins, thats all I ask of the stupid thing. Works fine with chrome v7-dev/ff3.6.9 both x64, still pegs a core or so playing videos, sigh! Any idea when it will make it into the multimedia repository? - John Why the multimedia repository? The standard repository doesn't have the flashplugin-nonfree ? smime.p7s Description: S/MIME Cryptographic Signature
Re: iomega 2TB external USB drive
On 09/06/2010 11:01 PM, John Lindsay wrote: Hi Celejar I tried 'mount /dev/sdb1 /mnt' as /sda1 returned drive does not exist. Using 'tail -F /var/log/syslog' when I try the mount command using /sdb1 returns - Sep 6 15:40:38 tux-net kernel: [119018.389495] sdb: sdb1 Sep 6 15:40:38 tux-net kernel: [119018.422500] sd 4:0:0:0: [sdb] Attached SCSI disk Sep 6 15:40:38 tux-net NetworkManager: debug [1283802038.156908] nm_hal_device_added(): New device added (hal udi is '/org/freedesktop/Hal/devices/storage_serial_Iomega_E_xternal_HD_FAF0FDF1FFF1BF600914_0_0'). Sep 6 15:44:34 tux-net kernel: [119289.522770] NTFS volume version 3.1. Sep 6 15:44:34 tux-net kernel: [119289.538772] NTFS-fs error (device sdb1): load_system_files(): $LogFile is not clean. Mounting read-only. Mount in Windows. Sep 6 15:45:00 tux-net ntpd[3215]: synchronized to 132.246.168.148, stratum 2 Sep 6 15:47:15 tux-net ntpd[3215]: synchronized to 199.212.17.34, stratum 2 It shows a NTSF formatted omega drive. It states read only. This much I understand. I am taking a wild guess and thinking that it will show up a an external drive under 'wine' -- I could be totally wrong in that assumption. All the rest is information lost on me!!! The log says Mount in Windows. So, hook it on a windows machine and do a checkdisk on it. smime.p7s Description: S/MIME Cryptographic Signature
Re: best way to backup a full encrypted LUKS system
On 07/27/2010 04:17 AM, David Christensen wrote: Γιώργος Πάλλας wrote: I have a system running off a fully encrypted 160GB HD using LVM over LUKS. Obviously, taking an image of the disk clonezilla-style is very inefficient as I need 160GB per backup. So, regularly, I take a full tar of the system while the system is live. An idea -- boot a live CD, open the LUKS partition(s), configure LVM, run zerofree against the unmounted logical volume, and then dd | tar | gzip | ccrypt the unmounted volume to an external hard drive? Thanks David for the idea! smime.p7s Description: S/MIME Cryptographic Signature
Re: Depreciated warning I can't quite read
On 07/27/2010 10:26 AM, Alan Chandler wrote: Seeing someone else ask a question about where boot messages are logged, reminds me of an issue I have been wondering about for a few weeks During bootup of my system, very early in the process, I am getting some messages appear on the screen very briefly which mention udevd SYSFS depreciated Its a warning of some sort, but it flies past so fast that I can't really read what it is saying. This is my desktop system, which originally ran debian unstable - but a couple of months ago I replaced unstable with squeeze in my sources.list and have been allowing the system to migrate into that mode. I am not sure if this message is a legacy, or everyone is seeing something similar. Anyone any idea what these messages might be and how I can rectify the whatever it is complaining about (just being able to see what the message says might also help). Maybe something like that? http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=582639 smime.p7s Description: S/MIME Cryptographic Signature
best way to backup a full encrypted LUKS system
Hello to everybody! I have a system running off a fully encrypted 160GB HD using LVM over LUKS. Obviously, taking an image of the disk clonezilla-style is very inefficient as I need 160GB per backup. So, regularly, I take a full tar of the system while the system is live. This way, data will not be lost, but in case of a disk failure, it will take some time before I set up manually the LUKS, then the LVM, then restore the data from the tar, restore grub, MBR etc *. Is there a better way? More efficient? How do others deal with that? Thanks! G. * By the way, I haven't tried this procedure, but my educated guess is that it will work... smime.p7s Description: S/MIME Cryptographic Signature
Re: how to make iceweasel icedove look as pretty as in gnome?
On 07/15/2010 05:15 PM, Florian Kulzer wrote: On Thu, Jul 15, 2010 at 09:12:45 +0300, Γιώργος Πάλλας wrote: On 07/14/2010 11:10 PM, Florian Kulzer wrote: On Wed, Jul 14, 2010 at 21:53:36 +0300, Γιώργος Πάλλας wrote: Is there a way? I tried by installing kde-config-gtk-style but both apps look terrible and they also don't exit properly. Which themes did you try? I use Clearlooks from the package gtk2-engines; it works without glitches on my KDE 4.4 desktop for all the GTK applications that I use and I find it quite pleasant to look at. According to the package description, Clearlooks is the default Gnome theme, so I would assume that it makes ice(weasel|dove) under KDE look the same as under Gnome. Hmm Could you please send me the output of aptitude search gtk~i $ aptitude --disable-columns -F '%p %v' search 'gtk~i' alsaplayer-gtk 0.99.80-5 gtk-qt-engine 1:1.1+svn5-4 gtk2-engines 1:2.20.1-1 gtk2-engines-pixbuf 2.20.1-1 ia32-libs-gtk 20090804 kde-config-gtk-style 1:1.1+svn5-4 libgpod4-nogtk 0.7.2-2+b1 libgtk2.0-0 2.20.1-1 libgtk2.0-bin 2.20.1-1 libgtk2.0-common 2.20.1-1 libgtkglext1 1.2.0-1.1 libgtkmm-2.4-1c2a 1:2.20.3-1 libgtkspell0 2.0.16-1 libwxgtk2.6-0 2.6.3.2.2-5 libwxgtk2.8-0 2.8.10.1-3 pinentry-gtk2 0.8.0-1 python-gtk2 2.17.0-2 You set the theme from System Settings - Appearance - Gtk themes etc? Yes, I do (the exact name is GTK Styles and Fonts on my system; I assume that is what you mean). I select the Use another style option in the GTK Styles panel and choose Clearlooks from the drop-down combobox. Clearlooks became available in this combobox after I installed the gtk2-engines package. Thanks Florian. I installed kde-config-gtk-style gtk2-engines gtk-qt-engine and followed your directions, and looks fine now! G. smime.p7s Description: S/MIME Cryptographic Signature
Re: how to make iceweasel icedove look as pretty as in gnome?
On 07/14/2010 11:10 PM, Florian Kulzer wrote: On Wed, Jul 14, 2010 at 21:53:36 +0300, Γιώργος Πάλλας wrote: Is there a way? I tried by installing kde-config-gtk-style but both apps look terrible and they also don't exit properly. Which themes did you try? I use Clearlooks from the package gtk2-engines; it works without glitches on my KDE 4.4 desktop for all the GTK applications that I use and I find it quite pleasant to look at. According to the package description, Clearlooks is the default Gnome theme, so I would assume that it makes ice(weasel|dove) under KDE look the same as under Gnome. Hmm Could you please send me the output of aptitude search gtk~i You set the theme from System Settings - Appearance - Gtk themes etc? smime.p7s Description: S/MIME Cryptographic Signature
how to make iceweasel icedove look as pretty as in gnome?
Is there a way? I tried by installing kde-config-gtk-style but both apps look terrible and they also don't exit properly. All the related messages I googled, are dated from 2008... Thanks, Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: advice on amd64
On 07/06/2010 09:07 AM, Mark Allums wrote: On 7/5/2010 5:40 PM, Jim McCloskey wrote: I recently acquired a new Dell Studio 15 laptop and mean to install Debian (squeeze) on it. The machine has a 1.6GHz Intel Quad Core processor and 6GB of RAM. The GPU is an ATI Mobility Radeon (HD 5470) with 1GB of onboard memory. Do you not mean a 2.6 GHz CPU (rather than 1.6)? I would go with amd64. 32-bit Flash also runs on amd64, and flashplayer-mozilla from Debian-multimedia (rather than flashplugin-nonfree) installs everything you need. And though it seems to imply that it exists for iceweasel/firefox (or possibly epiphany) it works well with Chrome, too. I would also go with amd64. I'm using 1 year now debian squeeze amd64 and never had a problem, except with skype where I had to do something with i32 libraries if I remember correct. But your question comes at the very day where upgrading my squeeze as I regularly do, I saw that flashplugin-nonfree disabled my flash!!! I just tried gnash and it doesn't work, and can't seem to find this flashplayer-mozilla on debian-multimedia*... G. * mordor:/home/gpall# grep multime /etc/apt/sources.list deb http://www.debian-multimedia.org squeeze main mordor:/home/gpall# aptitude search flashplayer-mozilla mordor:/home/gpall# smime.p7s Description: S/MIME Cryptographic Signature
virtuoso consuming too many resources + bug(?)
Hello to everyone! I opened a bug on virtuoso this morning (which doesn't already show up on the bug list) about not being able to stop the virtuoso file indexing. If any of you has the same problem please reply to the bug for extra feedback. Thanks! PS. The bug: I can't seem to be able to suspend virtuoso indexer: Right click at the icon, click on 'Suspend File Indexing' and OK, it is ticked. Hovering over the icon it says 'File indexer is suspended', but using 'top' I see: PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 1986 gpall 20 0 55164 45m 6288 S 85 2.3 40:30.38 virtuoso-t Indexing doesn't seem to stop and it consumes way too many resources on my computer... smime.p7s Description: S/MIME Cryptographic Signature
Re: nepomuk indexer in systray
AG wrote: On 21/05/10 12:37, Γιώργος Πάλλας wrote: Hello to all! I just upgraded my debian testing, installing nepomuk search. My problem is that the indexer every now and then appears in the systray for a second or so, and then disappears, which is very distracting (or am I very sensitive? :-) )... Anyone knows how to stop the icon from getting on my nerves? :-) Thanks! G. Aside from disabling the damn thing from working automatically in the first place - no. KDE4 is becoming increasingly bloated and does seem to be calling in an increasing amount of dependencies and additional database-related libraries which then seemingly take over. Not OK. My $0.02 worth. AG Agree... I also did not understand why some days ago as I upgraded my debian testing + KDE, a hundred packets related to nepomuk were installed, along with mysql server 5.1, and today as I upraded again, mysql server was removed. What was that??? smime.p7s Description: S/MIME Cryptographic Signature
nepomuk indexer in systray
Hello to all! I just upgraded my debian testing, installing nepomuk search. My problem is that the indexer every now and then appears in the systray for a second or so, and then disappears, which is very distracting (or am I very sensitive? :-) )... Anyone knows how to stop the icon from getting on my nerves? :-) Thanks! G. smime.p7s Description: S/MIME Cryptographic Signature
Re: δίσκος PATA WD και stock kern el 2.6.32-3 (SOLVED with 2.6.32-4 from sid)
On 05/02/2010 10:08 PM, Γιώργος Πάλλας wrote: On 05/02/2010 09:57 PM, George Kiagiadakis wrote: 2010/5/2 Γιώργος Πάλλαςgp...@ccf.auth.gr: On 05/01/2010 09:54 AM, Apollon Oikonomopoulos wrote: On 21:51 Wed 28 Apr , Γιώργος Πάλλας wrote: χελόου! Αναβάθμισα ένα debian squeeze box σε 2.6.32 από 2.6.30 και σταμάτησα να βλέπω έναν WD κλασσικό pata δίσκο. Επ' ευκαιρία, ψάχνοντάς το, σκέφτηκα ότι καλό θα ήταν να μάθω μερικά πράγματα για τα kernel modules. Έχω μαζέψει λοιπόν τα lsmod, lshw, dmesg και lspci τόσο από 2.6.30 όσο και από 2.6.32 για να εντοπίσω τι στραβώνει. Καλημέρα, Δεν κάνεις μια δοκιμή με τον 2.6.32-4, που αφενός είναι ο τρέχων πυρήνας, αφετέρου έχει γυρίσει σε libata drivers για όλους τους PATA controllers; Απόλλων ευχαριστώ Απόλλωνα για την ιδέα - θα προσθέσω στα sources λίγο sid και θα τον εγκαταστήσω. Έχω όμως μια απορία: γιατί ασχολήθηκαν με αλλαγές στους drivers των pata controllers δεδομένου του ότι οι pata δίσκοι σύντομα θα έχουν εξαφανιστεί; Οι παλιοί drivers αφορούν μόνο PATA συσκευές και είναι unmaintained πλέον. Οι καινούριοι αφορούν και PATA αλλά και SATA συσκευές με κάποιο τμήμα του κώδικα κοινό και συντηρούνται κανονικά, οπότε πιστεύω αντιλαμβάνεσαι το λόγο της αλλαγής. Οι drivers αυτοί βεβαίως δεν είναι καινούριοι, απλά το debian τώρα αποφάσισε να κάνει την μετάβαση σε αυτούς. α, οκ, κατανοητόν. Μάλλον αυτή είναι η μετάβαση η οποία ευθύνεται για το ότι το ubuntu έβλεπε από καιρό τους pata δίσκους ως /dev/sdX ενώ το debian κλασσικά ως /dev/hd* Πάντως το ότι ο 2.6.32-3 δεν βλέπει το δισκο μου περισσότερο για bug μου μοιαζει παρά για feature... Τελικά το δοκίμασα με 2.6.32-4 από το sid και τον είδε τον δίσκο (ως /dev/sdb πλέον). Άρα μάλλον bug θα ήταν, απλώς μου κάνει εντύπωση που πέρασε στον kernel του testing... smime.p7s Description: S/MIME Cryptographic Signature
Re: 2.6.32-3 does not recognize PATA WD disk... (SOLVED with 2.6.32-4 from sid)
On 04/28/2010 09:30 PM, Γιώργος Πάλλας wrote: please don't top post! :-) Answer at the end: Anand Sivaram wrote: This is not a kernel bug, but the way debian uses drivers. There are two different drivers in the kernel now. Old ide drivers and newer pata drivers. Old ide drivers are going to get deprecated. For some reason debian stock kernel is taking only ide drivers. I have seen this with both my desktop (ide amd74xx used, pata_amd is not at all compiled) and laptop (both ide piix and pata ata_pii modules are compiled, but ide piix used) Because of that my hard drive is shown as /dev/hda instead of /dev/sda with the debian stock kernel. What I did was to compile a custom kernel by including only the pata drivers. Now with this custom kernel everything is fine and consistent. 2010/4/28 Γιώργος Πάλλαςgp...@ccf.auth.gr: Hi to everybody! I just got a WD caviar blue, with the classic IDE interface and tried to connect it to my debian squeeze pc which already has a WD SATA device, where the OS lives. The problem is that 2.6.32-3 does not recognize the disk and although I'm not an expert, I tend to think that it is a kernel bug. I booted with 2.6.30-2-686 and it sees the disk: [1.560374] hda: WDC WD3200AAJB-56R1A0, ATA DISK drive About the IDE controller, here's what lspci -v says: on both 2.6.30 and 2.6.32 : 00:14.1 IDE interface: ATI Technologies Inc SB600 IDE (prog-if 8a [Master SecP PriP]) Subsystem: Giga-byte Technology Gigabyte GA-MA69G-S3H Motherboard Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 16 I/O ports at 01f0 [size=8] I/O ports at 03f4 [size=1] I/O ports at 0170 [size=8] I/O ports at 0374 [size=1] I/O ports at f900 [size=16] Kernel driver in use: ATIIXP_IDE The only differences between lspci -v for 2.6.30 and 2.6.32 are the following: # diff 2.6.30-2-lspci 2.6.32-3-lspci 23c23 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 38c38 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 95d94 Kernel driver in use: piix4_smbus where the pcieport differences refer to 00:02.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (external gfx0 port A) (prog-if 00 [Normal decode]) and 00:0a.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (PCI express gpp port F) (prog-if 00 [Normal decode]) The lshw utility shows almost perfect match between 2.6.30 and 2.6.32 with the difference that under the ide section which is identical in both kernels: *-ide description: IDE interface product: SB600 IDE vendor: ATI Technologies Inc physical id: 14.1 bus info: p...@:00:14.1 version: 00 width: 32 bits clock: 66MHz capabilities: ide bus_master configuration: driver=ATIIXP_IDE latency=32 resources: irq:16 ioport:1f0(size=8) ioport:3f6 ioport:170(size=8) ioport:376 ioport:f900(size=16) *-ide description: IDE Channel 0 physical id: 0 bus info: i...@0 logical name: ide0 clock: 66MHz but in 2.6.32 nothing disk relevant follows, where with 2.6.30, the disk section follows: *-disk description: ATA Disk product: WDC WD3200AAJB-56R1A0 vendor: Western Digital physical id: 0 bus info: i...@0.0 logical name: /dev/hda version: 01.03E01 serial: WD-WCAV2H175667 size: 298GiB (320GB) capabilities: ata dma lba iordy smart security pm partitioned partitioned:dos configuration: mode=udma5 signature=0804 smart=on bit strange, isn't it? Any ideas? As well as that, I can't report it via reportbug, because with: reportbug linux-image-2.6.32-3-686 I always get Querying Debian BTS for reports on linux-2.6 (source)... Unable to connect to Debian BTS; continue [y|N|?]? after a couple of minutes... Giorgos That means that there is no way to get the disk working by using the stock kernel? It is also strange that the drivers are there in the 2.6.32, but they do not seem to work: [1.256871] atiixp :00:14.1: IDE controller (0x1002:0x438c rev 0x00) [1.256894] ATIIXP_IDE :00:14.1: PCI INT A - GSI 16 (level, low) - IRQ 16 [1.256909] atiixp :00:14.1: not 100% native mode: will probe irqs later [1.256915] ide0: BM-DMA at 0xf900-0xf907 [1.256923] Probing IDE interface ide0... [1.265866] ide0: no devices on the port Any kernel expert to explain
Re: δίσκος PATA WD και stock kern el 2.6.32-3
On 05/01/2010 09:54 AM, Apollon Oikonomopoulos wrote: On 21:51 Wed 28 Apr , Γιώργος Πάλλας wrote: χελόου! Αναβάθμισα ένα debian squeeze box σε 2.6.32 από 2.6.30 και σταμάτησα να βλέπω έναν WD κλασσικό pata δίσκο. Επ' ευκαιρία, ψάχνοντάς το, σκέφτηκα ότι καλό θα ήταν να μάθω μερικά πράγματα για τα kernel modules. Έχω μαζέψει λοιπόν τα lsmod, lshw, dmesg και lspci τόσο από 2.6.30 όσο και από 2.6.32 για να εντοπίσω τι στραβώνει. Καλημέρα, Δεν κάνεις μια δοκιμή με τον 2.6.32-4, που αφενός είναι ο τρέχων πυρήνας, αφετέρου έχει γυρίσει σε libata drivers για όλους τους PATA controllers; Απόλλων ευχαριστώ Απόλλωνα για την ιδέα - θα προσθέσω στα sources λίγο sid και θα τον εγκαταστήσω. Έχω όμως μια απορία: γιατί ασχολήθηκαν με αλλαγές στους drivers των pata controllers δεδομένου του ότι οι pata δίσκοι σύντομα θα έχουν εξαφανιστεί; smime.p7s Description: S/MIME Cryptographic Signature
Re: δίσκος PATA WD και stock kern el 2.6.32-3
On 05/02/2010 09:57 PM, George Kiagiadakis wrote: 2010/5/2 Γιώργος Πάλλαςgp...@ccf.auth.gr: On 05/01/2010 09:54 AM, Apollon Oikonomopoulos wrote: On 21:51 Wed 28 Apr , Γιώργος Πάλλας wrote: χελόου! Αναβάθμισα ένα debian squeeze box σε 2.6.32 από 2.6.30 και σταμάτησα να βλέπω έναν WD κλασσικό pata δίσκο. Επ' ευκαιρία, ψάχνοντάς το, σκέφτηκα ότι καλό θα ήταν να μάθω μερικά πράγματα για τα kernel modules. Έχω μαζέψει λοιπόν τα lsmod, lshw, dmesg και lspci τόσο από 2.6.30 όσο και από 2.6.32 για να εντοπίσω τι στραβώνει. Καλημέρα, Δεν κάνεις μια δοκιμή με τον 2.6.32-4, που αφενός είναι ο τρέχων πυρήνας, αφετέρου έχει γυρίσει σε libata drivers για όλους τους PATA controllers; Απόλλων ευχαριστώ Απόλλωνα για την ιδέα - θα προσθέσω στα sources λίγο sid και θα τον εγκαταστήσω. Έχω όμως μια απορία: γιατί ασχολήθηκαν με αλλαγές στους drivers των pata controllers δεδομένου του ότι οι pata δίσκοι σύντομα θα έχουν εξαφανιστεί; Οι παλιοί drivers αφορούν μόνο PATA συσκευές και είναι unmaintained πλέον. Οι καινούριοι αφορούν και PATA αλλά και SATA συσκευές με κάποιο τμήμα του κώδικα κοινό και συντηρούνται κανονικά, οπότε πιστεύω αντιλαμβάνεσαι το λόγο της αλλαγής. Οι drivers αυτοί βεβαίως δεν είναι καινούριοι, απλά το debian τώρα αποφάσισε να κάνει την μετάβαση σε αυτούς. α, οκ, κατανοητόν. Μάλλον αυτή είναι η μετάβαση η οποία ευθύνεται για το ότι το ubuntu έβλεπε από καιρό τους pata δίσκους ως /dev/sdX ενώ το debian κλασσικά ως /dev/hd* Πάντως το ότι ο 2.6.32-3 δεν βλέπει το δισκο μου περισσότερο για bug μου μοιαζει παρά για feature... smime.p7s Description: S/MIME Cryptographic Signature
Re: Unable to connect to my home wireless
On 05/02/2010 09:24 AM, Marc Shapiro wrote: I am running an up to date Squeeze system using LXDE. I use wicd as my network manager. Since doing a full upgrade on Monday I have been unable to connect to my home wireless network. I can connect to various unsecured networks at work, at local restaurants and coffee shops. I just can't connect to my home network. Here, I have to connect a cat5 cable. Since the only place to do this is sitting directly in front of my desktop system it kind of defeats the purpose. I have verified that the WEP key is correct and I really don't know what else to look for. Hi! Do an attempt to connect, and then as root, go to /var/log and give an ' ls -lat' - up on the list are the log files which were changed recently and contain the logs you want. Check them and send the relevant parts. By the way, my brother who was running ubuntu 9.10 just upgraded to 10.04 and has exactly the same issue. Can you tell us what is the version you are currently running? (aptitude show wicd) G. smime.p7s Description: S/MIME Cryptographic Signature
2.6.32-3 does not recognize PATA WD disk...
Hi to everybody! I just got a WD caviar blue, with the classic IDE interface and tried to connect it to my debian squeeze pc which already has a WD SATA device, where the OS lives. The problem is that 2.6.32-3 does not recognize the disk and although I'm not an expert, I tend to think that it is a kernel bug. I booted with 2.6.30-2-686 and it sees the disk: [1.560374] hda: WDC WD3200AAJB-56R1A0, ATA DISK drive About the IDE controller, here's what lspci -v says: on both 2.6.30 and 2.6.32 : 00:14.1 IDE interface: ATI Technologies Inc SB600 IDE (prog-if 8a [Master SecP PriP]) Subsystem: Giga-byte Technology Gigabyte GA-MA69G-S3H Motherboard Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 16 I/O ports at 01f0 [size=8] I/O ports at 03f4 [size=1] I/O ports at 0170 [size=8] I/O ports at 0374 [size=1] I/O ports at f900 [size=16] Kernel driver in use: ATIIXP_IDE The only differences between lspci -v for 2.6.30 and 2.6.32 are the following: # diff 2.6.30-2-lspci 2.6.32-3-lspci 23c23 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 38c38 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 95d94 Kernel driver in use: piix4_smbus where the pcieport differences refer to 00:02.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (external gfx0 port A) (prog-if 00 [Normal decode]) and 00:0a.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (PCI express gpp port F) (prog-if 00 [Normal decode]) The lshw utility shows almost perfect match between 2.6.30 and 2.6.32 with the difference that under the ide section which is identical in both kernels: *-ide description: IDE interface product: SB600 IDE vendor: ATI Technologies Inc physical id: 14.1 bus info: p...@:00:14.1 version: 00 width: 32 bits clock: 66MHz capabilities: ide bus_master configuration: driver=ATIIXP_IDE latency=32 resources: irq:16 ioport:1f0(size=8) ioport:3f6 ioport:170(size=8) ioport:376 ioport:f900(size=16) *-ide description: IDE Channel 0 physical id: 0 bus info: i...@0 logical name: ide0 clock: 66MHz but in 2.6.32 nothing disk relevant follows, where with 2.6.30, the disk section follows: *-disk description: ATA Disk product: WDC WD3200AAJB-56R1A0 vendor: Western Digital physical id: 0 bus info: i...@0.0 logical name: /dev/hda version: 01.03E01 serial: WD-WCAV2H175667 size: 298GiB (320GB) capabilities: ata dma lba iordy smart security pm partitioned partitioned:dos configuration: mode=udma5 signature=0804 smart=on bit strange, isn't it? Any ideas? As well as that, I can't report it via reportbug, because with: reportbug linux-image-2.6.32-3-686 I always get Querying Debian BTS for reports on linux-2.6 (source)... Unable to connect to Debian BTS; continue [y|N|?]? after a couple of minutes... Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: 2.6.32-3 does not recognize PATA WD disk...
please don't top post! :-) Answer at the end: Anand Sivaram wrote: This is not a kernel bug, but the way debian uses drivers. There are two different drivers in the kernel now. Old ide drivers and newer pata drivers. Old ide drivers are going to get deprecated. For some reason debian stock kernel is taking only ide drivers. I have seen this with both my desktop (ide amd74xx used, pata_amd is not at all compiled) and laptop (both ide piix and pata ata_pii modules are compiled, but ide piix used) Because of that my hard drive is shown as /dev/hda instead of /dev/sda with the debian stock kernel. What I did was to compile a custom kernel by including only the pata drivers. Now with this custom kernel everything is fine and consistent. 2010/4/28 Γιώργος Πάλλας gp...@ccf.auth.gr: Hi to everybody! I just got a WD caviar blue, with the classic IDE interface and tried to connect it to my debian squeeze pc which already has a WD SATA device, where the OS lives. The problem is that 2.6.32-3 does not recognize the disk and although I'm not an expert, I tend to think that it is a kernel bug. I booted with 2.6.30-2-686 and it sees the disk: [1.560374] hda: WDC WD3200AAJB-56R1A0, ATA DISK drive About the IDE controller, here's what lspci -v says: on both 2.6.30 and 2.6.32 : 00:14.1 IDE interface: ATI Technologies Inc SB600 IDE (prog-if 8a [Master SecP PriP]) Subsystem: Giga-byte Technology Gigabyte GA-MA69G-S3H Motherboard Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 16 I/O ports at 01f0 [size=8] I/O ports at 03f4 [size=1] I/O ports at 0170 [size=8] I/O ports at 0374 [size=1] I/O ports at f900 [size=16] Kernel driver in use: ATIIXP_IDE The only differences between lspci -v for 2.6.30 and 2.6.32 are the following: # diff 2.6.30-2-lspci 2.6.32-3-lspci 23c23 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 38c38 Kernel driver in use: pcieport-driver --- Kernel driver in use: pcieport 95d94 Kernel driver in use: piix4_smbus where the pcieport differences refer to 00:02.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (external gfx0 port A) (prog-if 00 [Normal decode]) and 00:0a.0 PCI bridge: ATI Technologies Inc RD790 PCI to PCI bridge (PCI express gpp port F) (prog-if 00 [Normal decode]) The lshw utility shows almost perfect match between 2.6.30 and 2.6.32 with the difference that under the ide section which is identical in both kernels: *-ide description: IDE interface product: SB600 IDE vendor: ATI Technologies Inc physical id: 14.1 bus info: p...@:00:14.1 version: 00 width: 32 bits clock: 66MHz capabilities: ide bus_master configuration: driver=ATIIXP_IDE latency=32 resources: irq:16 ioport:1f0(size=8) ioport:3f6 ioport:170(size=8) ioport:376 ioport:f900(size=16) *-ide description: IDE Channel 0 physical id: 0 bus info: i...@0 logical name: ide0 clock: 66MHz but in 2.6.32 nothing disk relevant follows, where with 2.6.30, the disk section follows: *-disk description: ATA Disk product: WDC WD3200AAJB-56R1A0 vendor: Western Digital physical id: 0 bus info: i...@0.0 logical name: /dev/hda version: 01.03E01 serial: WD-WCAV2H175667 size: 298GiB (320GB) capabilities: ata dma lba iordy smart security pm partitioned partitioned:dos configuration: mode=udma5 signature=0804 smart=on bit strange, isn't it? Any ideas? As well as that, I can't report it via reportbug, because with: reportbug linux-image-2.6.32-3-686 I always get Querying Debian BTS for reports on linux-2.6 (source)... Unable to connect to Debian BTS; continue [y|N|?]? after a couple of minutes... Giorgos That means that there is no way to get the disk working by using the stock kernel? It is also strange that the drivers are there in the 2.6.32, but they do not seem to work: [1.256871] atiixp :00:14.1: IDE controller (0x1002:0x438c rev 0x00) [1.256894] ATIIXP_IDE :00:14.1: PCI INT A - GSI 16 (level, low) - IRQ 16 [1.256909] atiixp :00:14.1: not 100% native mode: will probe irqs later [1.256915] ide0: BM-DMA at 0xf900-0xf907 [1.256923] Probing IDE interface ide0... [1.265866] ide0: no devices on the port Any kernel expert to explain what exactly is happening
Re: Different ways of creating a USB-install stick
James Stuckey wrote: Hello, I'm curious as to the different ways one can create a USB install stick, for the purpose of installing Debian. It is possible to do it by a) acquiring (where?) a boot.img.gz file, and then doing zcat boot.img.gz /dev/sdc. Then load a netboot iso to the disc. b) making a FAT/FAT32 partition on the stick, and then doing syslinux /dev/sdXX, copying vmlinuz and initrd.gz to the stick c) using Unetbootin with an .iso file I like the way Unetbootin allows one to use a larger .iso (dvd, or full cd iso). The downside of a) and b) is that you can only use the smaller netboot .isos, and if I wanted to load a few more files onto the disc for use I wouldn't have the room to do so. Does anyone know another method? I'm particularly interested in knowing the way to do what Unetbootin does manually, so that I could load a cd/dvd iso onto the stick after making it boot-able. Regards, James About c), I know that ubuntu can work that way, but has failed in the past with debian, because the debian installer believes it runs from a cd and tries to find the packages to be installed from the host's cdrom. I dont know if something has changed recently. Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: Can't read DVD
James Stuckey wrote: I'm not able to read DVDs in squeeze. I burned this disc on the same drive in squeeze. Thereafter it worked fine until a week or so ago (the disc is less than a month old). I tested the disc last week on a windoze and mac osx and it worked without a problem. When I try to mount the cd/dvd rom drive: stuc...@debian:~$ mount /dev/scd0 mount: block device /dev/sr0 is write-protected, mounting read-only mount: wrong fs type, bad option, bad superblock on /dev/sr0, missing codepage or helper program, or other error (could this be the IDE device where you in fact use ide-scsi so that sr0 or sda or so is needed?) In some cases useful info is found in syslog - try dmesg | tail or so Could be many things... First of all try the disc on another machine... smime.p7s Description: S/MIME Cryptographic Signature
Re: crypt question/server hotel
Osamu Aoki wrote: Hi, On Sat, Apr 17, 2010 at 10:49:20AM +0200, Jozsi Vadkan wrote: I want to put my server in a server hotel. But: I don't trust my server hotel owner. What can I do? I am no expert on this issue but this is my common sense. Do not use such untrusted servers for the sensitive data. You can put measures to remote break-in etc. But whoever have local hysical access can get tou your data on the system. (I do not quite understand what kind of server arrangement ... virtualized or rack moiunted dedicated server... either way, it is the same thing.) I can crypt my partition/hdd's that contains the data. Ok. But: then my operating system will not be encrypted. Not Ok. Well, once booted, and if they have some kind of hardware access before you boot into your system, you are doomed. Because they can have backdoor access. If I crypt my operating system too, then when a reboot comes, I have to type a password to decrypt. But my server will be at a server hotel I can't directly use a keyboard [no service cpu]. All these methods protect against casual break-in but if system is run under some super-server like xen etc., your security measure stopps there. What can I do [on technical side] to ensure a little more security to my server [e.g: crypt my partition/slice/whatever, that has the operating system, but without the type password problem] If they have monitoring system pre-installed, ... even with this protection is no good. Thank you for any tips/help. Keep sensitive data where you have full trust. The remote untrusted servers are good for web gateway only. But even for that, you should have some trust to them. Osamu you may be interested in this: Unlocking a LUKS encrypted root partition via ssh http://www.debian-administration.org/articles/579 smime.p7s Description: S/MIME Cryptographic Signature
Re: Error: Read from socket failed: Connection reset by peer
vikram wrote: Hi, This is the output of scp -vv $ scp -vv files.tar.gz r...@192.168.1.11:/root/Desktop Executing: program /usr/bin/ssh host 192.168.1.11, user root, command scp -v -t /root/Desktop OpenSSH_5.1p1 Debian-5, OpenSSL 0.9.8g 19 Oct 2007 debug1: Reading configuration data /etc/ssh/ssh_config debug1: Applying options for * debug2: ssh_connect: needpriv 0 debug1: Connecting to 192.168.1.11 [192.168.1.11] port 22. debug1: Connection established. debug1: identity file /home/user/.ssh/identity type -1 debug1: identity file /home/user/.ssh/id_rsa type -1 debug1: identity file /home/user/.ssh/id_dsa type -1 debug1: Remote protocol version 2.0, remote software version OpenSSH_5.1p1 Debian-5 debug1: match: OpenSSH_5.1p1 Debian-5 pat OpenSSH* debug1: Enabling compatibility mode for protocol 2.0 debug1: Local version string SSH-2.0-OpenSSH_5.1p1 Debian-5 debug2: fd 3 setting O_NONBLOCK debug1: SSH2_MSG_KEXINIT sent Read from socket failed: Connection reset by peer lost connection I get this not matter which file i try to copy. Hmmm, this could be an active RST denial of service. Could you possibly get a tcpdump of the session and check if you receive an RST packet just before the connection reset by peer? smime.p7s Description: S/MIME Cryptographic Signature
Re: Error: Read from socket failed: Connection reset by peer
vikram wrote: Hi, I get this while trying to scp onto a local machine(on the same subnet, both machines running Lenny). Read from socket failed: Connection reset by peer lost connection I am able to ping the machine fine. How can this be fixed? Thanks Vikram It starts copying and then this, or you get this from the beginning? Run scp with option -vv and send the output. smime.p7s Description: S/MIME Cryptographic Signature
Re: unable to connect to Debian BTS
Ron Johnson wrote: On 2010-04-08 06:40, Γιώργος Πάλλας wrote: The last few days, when using reportbug, I get (from two machines, on two different networks): Querying Debian BTS for reports on linux-2.6 (source)... Unable to connect to Debian BTS; continue [y|N|?]? except some very rare cases. Anybody else having the same problem? I just tried it from the southern US, and it worked like a charm. I just noticed that it only fails on a certain package: linux-image-2.6.32-3-686 which has over 800 bugs. Must be some timeout... ... Found a reportbug filed bug: #572316: reportbug: timeout for querying the BTS should be configurable by user for which a --timeout option is to be added. smime.p7s Description: S/MIME Cryptographic Signature
unable to connect to Debian BTS
The last few days, when using reportbug, I get (from two machines, on two different networks): Querying Debian BTS for reports on linux-2.6 (source)... Unable to connect to Debian BTS; continue [y|N|?]? except some very rare cases. Anybody else having the same problem? smime.p7s Description: S/MIME Cryptographic Signature
Re: host trying to connect using IPv6 without having IPv6 global address
Mark Kamichoff wrote: On Tue, Mar 02, 2010 at 11:50:09PM +0200, Γιώργος Πάλλας wrote: It sits behind an ordinary adsl router, with no IPv6 capabilities, it gets its private 192.168.x.x address etc. Now the problem is that it tries to connect to hosts returning quad-A records using IPv6, but it has no IPv6 global address. Ifconfig shows only the normal 'link' IPv6 address. So, eventually it does not connect. For example, using the browser I can't go to http://www.ipv6.org Aptitude update stalls, writing: 84% [Connecting to security.debian.org (2001:610:1908:a000::149:225)] Or even: $ telnet security.debian.org Trying 2001:610:1908:a000::149:225... I would appreciate any help, because this problem starts to get me! Below, in the bz2 file, I give a transcript of various useful command outputs, if someone bothers to dive into it! I include a telnet strace, netstat IPv6 output, ifconfig, my IPv6 routing table, and other things. Something I suspect is this entry in my routing table: ::/0 fe80::4432:f215:1283:2320 UGDAe 1024 090 eth2 should that be there? This is the problem. Issue an ip -6 route del default and you should be all set. I suspect at some point your host accepted an RA packet (possibly with the 'M' flag set, which may explain the lack of an autoconfigured global address) and set the default route to the LL address of the advertising router. - Mark You were right! I deleted the route, and all was OK - until 10 minutes later that the route reappeared! I deleted it again and it reappeared once more. I then set up tcpdump running and checked the packets I received when the route reappeared. So I discovered something interesting. This is the packet that was setting up the route: 20:30:25.828992 IP6 fe80::4432:f215:1283:2320 ff02::1: ICMP6, router advertisement, length 32 My adsl router did not have such a mac address, so I checked my wife's laptop which is running the evil Windows 7. I found out that the wireless interface had the address fe80::4432:f215:1283:2320 as link local, and was sending in a 7-10 minutes interval this router advertisement... This advertisement comes among other messages related to IPv6 (in the following, fe80::222:5fff:fe6e:cc63 is my debian's laptop link local address on the wireless interface): 19:49:12.259649 IP6 fe80::222:5fff:fe6e:cc63 fe80::4432:f215:1283:2320: ICMP6, neighbor solicitation, who has fe80::4432:f215:1283:2320, length 32 19:49:12.263671 IP6 fe80::4432:f215:1283:2320 fe80::222:5fff:fe6e:cc63: ICMP6, neighbor advertisement, tgt is fe80::4432:f215:1283:2320, length 32 19:49:17.294542 IP6 fe80::4432:f215:1283:2320 fe80::222:5fff:fe6e:cc63: ICMP6, neighbor solicitation, who has fe80::222:5fff:fe6e:cc63, length 32 19:49:17.294571 IP6 fe80::222:5fff:fe6e:cc63 fe80::4432:f215:1283:2320: ICMP6, neighbor advertisement, tgt is fe80::222:5fff:fe6e:cc63, length 24 19:49:55.387376 IP6 fe80::4432:f215:1283:2320 ff02::1: ICMP6, router advertisement, length 32 Anybody else has heard of this win 7 behavior? G. smime.p7s Description: S/MIME Cryptographic Signature
host trying to connect using IPv6 without having IPv6 global address
Hello to all! I am having the following strange problem on a fully updated squeeze. It sits behind an ordinary adsl router, with no IPv6 capabilities, it gets its private 192.168.x.x address etc. Now the problem is that it tries to connect to hosts returning quad-A records using IPv6, but it has no IPv6 global address. Ifconfig shows only the normal 'link' IPv6 address. So, eventually it does not connect. For example, using the browser I can't go to http://www.ipv6.org Aptitude update stalls, writing: 84% [Connecting to security.debian.org (2001:610:1908:a000::149:225)] Or even: $ telnet security.debian.org Trying 2001:610:1908:a000::149:225... I would appreciate any help, because this problem starts to get me! Below, in the bz2 file, I give a transcript of various useful command outputs, if someone bothers to dive into it! I include a telnet strace, netstat IPv6 output, ifconfig, my IPv6 routing table, and other things. Something I suspect is this entry in my routing table: ::/0 fe80::4432:f215:1283:2320 UGDAe 1024 090 eth2 should that be there? Thanks for any help, Giorgos log.txt.bz2 Description: Binary data smime.p7s Description: S/MIME Cryptographic Signature
προσπάθεια σύνδεσης με IPv6
Καλησπέρα σε όλους! Τις τελευταίες μέρες, σε ένα fully updated debian squeeze, πάνω σε siemens adsl router με forthnet ISP, έχω το πρόβλημα ότι προσπαθεί να μιλήσει με IPv6 σε IPv6 enabled sites. Ουσιαστικά δηλαδή, σε sites των οποίων το DNS επιστρέφει record. Φυσικά, ο ISP δεν δίνει IPv6, ούτε και ο routerάκος, και ενώ το interface μου είναι ως εξής: eth2 Link encap:Ethernet HWaddr 00:22:5f:6e:cc:63 inet addr:192.168.1.200 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::222:5fff:fe6e:cc63/64 Scope:Link παρόλα αυτά ο υπολογιστής προσπαθεί να μιλήσει με IPv6. Αρχικά το παρατήρησα με το security.debian.org κατά τη διάρκεια ενός aptitude update: 84% [Connecting to security.debian.org (2001:8d8:2:1:6564:a62:0:2)] Αλλά παρομοίως συμπεριφέρεται και αλλού: mordor:/home/gpall# telnet www.ipv6.org 80 Trying 2001:6b0:1:ea:202:a5ff:fecd:13a6... Τί να συμβαίνει; Πως παίρνει την απόφαση το λειτουργικό για το αν θα προσπαθήσει να συνδεθεί με IPv6; To αν έχει IPv6 global address? smime.p7s Description: S/MIME Cryptographic Signature
Re: how to convince that debian is one the three major choices for a stable server environment?
Michal wrote: Well, we 're not Google or Facebook, and we would like to stick with linux... I'm not 100% sure what that has to do with anything...I'm taking a a guess at maybe your thinking of high performance web sites in which case it has a lot of uses over just that, if that's what you think. quote What applications are you planning on running? What kind of performance do you need? As others have suggested, *BSD have a great security and stability track record. However, Linux (all distros) have much more capability than *BSD in many areas. File system choice is limited in the *BSD variants to UFS/FFS, the BSD Fast filesystem, and SFS, which reserves odd numbered inodes for ACL information, but otherwise is identical to FFS. *BSD do not often alternative high performance file systems such as XFS, the fastest filesystem available for most applications. We are a large university's NOC, with thousands of users and the requirement that our servers are interoperable with other universities also running linux. In short, we must stick to linux. Anyway, I found tons of references pro Debian on the web so I think I'm done with my argument! G. smime.p7s Description: S/MIME Cryptographic Signature
Re: how to convince that debian is one the three major choices for a stable server environment?
Mihamina Rakotomandimby wrote: Γιώργος Πάλλας gp...@ccf.auth.gr : (it is, isn't it? :-) ) So, yes, we are moving on from our 10year experience with gentoo What reasons have you collected to decide to move from Gentoo? I want 'x' program installed, and I want it *now* to solve a burning situation. But I can't have it! Because I'm a happy gentoo user! I have to get it, compile it, oh I forgot the snmp flag, recompile it, and there I am! The drop that filled the bucket (or whatever the expression is), was that we can't have the latest PHP with the latest SNMP module because of a certain PHP bug that is refused to be dealt with in Gentoo - in debian it works fine. Anyway, don't ask me any further Gentoo specific questions, I don't want to know more about Gentoo, I just want to forget it! smime.p7s Description: S/MIME Cryptographic Signature
how to convince that debian is one the three major choices for a stable server environment?
(it is, isn't it? :-) ) So, yes, we are moving on from our 10year experience with gentoo, and are searching for our new environment. From my personal experience I would say debian stable - any hard evidence to support the claim? Server OS statistics? Statistics for stableness? Bugs? Any white papers showing debian's superiority? I am also doing my google research, but I'm asking if someone can point me to something like real hard evidence... Thanks, G. smime.p7s Description: S/MIME Cryptographic Signature
Re: how to convince that debian is one the three major choices for a stable server environment?
Tim Clewlow wrote: On 22/02/2010 13:01, ΓιώÏγος Πάλλας wrote: (it is, isn't it? :-) ) So, yes, we are moving on from our 10year experience with gentoo, and are searching for our new environment. From my personal experience I would say debian stable - any hard evidence to support the claim? Server OS statistics? Statistics for stableness? Bugs? Any white papers showing debian's superiority? I am also doing my google research, but I'm asking if someone can point me to something like real hard evidence... Thanks, G. Depending on what you want to use the servers for, OpenBSD I second this - *bsd is very good for servers. Gnu/Linux is adequate for small scale enterprises but will not give you the hard core grunt that *bsd will. Also OpenBSD has a seriously secure core Well, we 're not Google or Facebook, and we would like to stick with linux... smime.p7s Description: S/MIME Cryptographic Signature
Re: Bug in file command magic
Giovanni Funchal wrote:
Hi,
I've just noticed that the file command returns wrong mime for a file.
In fact, if a LaTeX file uses the command \bibligraphystyle somewhere
in the beginning, file will return mime text/vnd.graphviz instead of
text/x-tex. I took a look at /usr/share/file/magic (on Lenny), and
it has something like :
0regex/100[\r\n\t\ ]*graph[\r\n\t\ ]*.*\\{graphviz graph text
!:mimetext/vnd.graphviz
This regular expression matches \biblioGRAPHystyle in the first
thousand characters or so. A good fix would be :
- 0regex/100[\r\n\t\ ]*graph[\r\n\t\ ]*.*\\{graphviz graph text
+ 0regex/100(^|[\r\n\t\ ]+)graph[\r\n\t\ ]*.*\\{graphviz graph text
Since in the dot file format[1], the keyword graph must be preceded
by a white space or be in the beginning of the file.
Anyone knows where I can report this?
-- Giovanni
[1] http://www.graphviz.org/doc/info/lang.html
I think this belongs to package 'file'...
smime.p7s
Description: S/MIME Cryptographic Signature
knetworkmanager on debian testing broken
After yesterday'ss upgrade to version 0.7.999-2, I cannot connect to any network, wireless or not. Related bug: 567354 - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567354 smime.p7s Description: S/MIME Cryptographic Signature
Re: knetworkmanager on debian testing broken
Γιώργος Πάλλας wrote: After yesterday'ss upgrade to version 0.7.999-2, I cannot connect to any network, wireless or not. Related bug: 567354 - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567354 And by the way, I have a question: I have seen on a kubuntu machine, a nice network manager GUI. Can I install this on my debian machine? Does it support openvpn, vpnc, mobile, etc? And most importantly, what is the package's name? Thanks! G. smime.p7s Description: S/MIME Cryptographic Signature
Re: knetworkmanager on debian testing broken
tv.deb...@googlemail.com wrote: Γιώργος Πάλλας wrote: Γιώργος Πάλλας wrote: After yesterday'ss upgrade to version 0.7.999-2, I cannot connect to any network, wireless or not. Related bug: 567354 - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567354 And by the way, I have a question: I have seen on a kubuntu machine, a nice network manager GUI. Can I install this on my debian machine? Does it support openvpn, vpnc, mobile, etc? And most importantly, what is the package's name? Thanks! G. Hi, not sure about Kubuntu but it may be NetworkManager's plasmoïd, it's in Debian experimental currently under the name plasma-widget-networkmanagement. So it's nm under the hood with vpn support (but nm brokenness too) , and the package is currently in experimental for a reason: buggy (tried it a few weeks ago). Another gui is wicd, it's nice and works well, but I am not aware of a plugging for vpn that goes with it. You can use another gui application maybe, like kvpnc (never tried that one, name's suggest it should be up to the job...). Can't comment on connection through mobile. You're right, its called plasma-widget-networkmanagement but it is buggy... I tried wicd which tortured me a little because it seemed to have a problem either with dhclient3 or network-manager which remained installed after I removed network-manager-kde. After I removed both dhclient3 and network-manager, I reinstalled wicd and it worked fine. Thanks! G. smime.p7s Description: S/MIME Cryptographic Signature
Re: KMS, vga=791 and intel xorg driver
Matthew Moore wrote: On Saturday February 6 2010 2:36:02 pm Γιώργος Πάλλας wrote: So I removed vga=791 and problems with getting a console were resolved. In order to have a good resolution for my console while the pc is booting, I used the GRUB_GFXMODE=1440x900 directive in the grub2 config, but while it starts at this resolution, after the message 'Loading the initial image' (or something like that), it reverts again to a very low analysis. How can I fix that? I think that you need to append video=i915 to the linux command line, via the variable GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub. This specifies the KMS driver for the kernel to use. When KMS gets loaded, it resets the console's scrollback buffer. Since the KMS driver gets loaded late in the boot process in Debian (later than in most other distro's), you won't be able to see any of your boot messages once your machine reaches login. Apparently this is going to be fixed in the future. MM Thanks for your answers everybody! So, I got it working by adding video=i915:modeset=1 to the GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub. I also want to add that I don't see happening what Matthew wrote. I can see all the boot messages while the machine boots, and after the machine reaches login prompt. Even now, from the X session, if I press CTRL-F1 I get to the console where there are the boot messages, and at the bottom, the login prompt. Cheers, Giorgos smime.p7s Description: S/MIME Cryptographic Signature
KMS, vga=791 and intel xorg driver
Hi to all! After I upgraded last week to latest xorg drivers my squeeze box, I had problem getting a console: the X session just froze. I searched a bit and found bug #567563. So I discovered that the problem was the 'vga=791' setting that I used to boot with, so that I have a decent resolution while booting, for the kernel messages. So I removed vga=791 and problems with getting a console were resolved. In order to have a good resolution for my console while the pc is booting, I used the GRUB_GFXMODE=1440x900 directive in the grub2 config, but while it starts at this resolution, after the message 'Loading the initial image' (or something like that), it reverts again to a very low analysis. How can I fix that? Thanks! G. smime.p7s Description: S/MIME Cryptographic Signature
το επίπεδο αφαίρεσης υ λισμικού βαίνει προς το τέ λος του;
...ή στα ελληνικά: ο HAL πάει να γίνει obsolete; Έχει εδώ και μήνες κάποιο περίεργο bug που έχει αναφερθεί από πολύ κόσμο, το οποίο προκαλεί ένα segmentation fault όταν κάνεις τον hal restart, καθώς και όταν ξεκινάει η υπηρεσία, αλλά δεν λέει να διορθωθεί - κανείς δεν ασχολείται; Άκουσα δε, αλλά δεν θυμάμαι που, ότι ο hal θα γίνει obsolete γιατί θα αντικατασταθεί από το udev (αν θυμάμαι). Iσχύει; Στέλνω το mail και στην debian-users γιατί θέλω να ρωτήσω επίσης, που μπορεί κανείς να βρεί ειδικά για το debian κάποιο επίσημο forum ας πούμε όπου συζητούνται τέτοιου είδους αποφάσεις πριν ληφθούν. Μήπως οι ηγέτες του debian συναντιούνται κάπου μυστικά ντυμένοι με περίεργα ρούχα;; :-P smime.p7s Description: S/MIME Cryptographic Signature
Re: problem installing nvidia-glx on a debian testing with 2.6.32 kernel
Γιώργος Πάλλας wrote: Hello to everybody! I'm trying to install nvidia-glx in order to make my GeForce FX 5200 play, but I bump on this curious situation: What is this irrelevant 2.6.26-2-686 kernel that is mentioned? My system runs 2.6.32-trunk-686... any ideas? thanks to everybody who answered but since I did not want to enable the unstable repository, I installed the nvidia driver from its site which compiled and played like a charm... G. smime.p7s Description: S/MIME Cryptographic Signature
Re: shutdown after kernel upgrade
Márcio H. Parreiras wrote: Hi, I have an Acer Aspire 5315 laptop, Intel GMA965 chipset, with Debian Testing installed. Today morning I've made an system upgrade. Since then, when kernel 2.6.32-trunk-686 is loaded, the laptop is powering off suddenly, few minutes after boot. If I choose the old kernel the symptom do not happens. Is any bug or there are some config to do? In time: another machine, a desktop, with the same distro, was upgraded too and didn't show any problem with the new kernel. same symptom here on a Dell studio laptop (amd64) and on a classical pentium pc, but shutdown does not happen in 2 minutes but needs a couple of hours... No trace left in syslog and kernel log though... smime.p7s Description: S/MIME Cryptographic Signature
Re: shutdown after kernel upgrade
John wrote: On 28/01/10, ?? (gp...@ccf.auth.gr) wrote: | Márcio H. Parreiras wrote: | ... Acer Aspire 5315 laptop ... kernel 2.6.32-trunk-686 ... | powering off suddenly, few minutes after boot. If I | choose the old kernel the symptom do not happen ... | same symptom here on a Dell studio laptop (amd64) and on a classical | pentium pc... Same here with both Debian and vanilla versions of all 2.6.32 kernels, home-compiled, running sid on an old IBM Thinkpad A31, with cpufreqd and cpufrequtils. On my machine, the shutdown is caused by runaway overheating after only a few minutes, so there's not much time for diagnosing. 2.6.31.6 with same basic .config runs fine. how did you understand it was due to heating?? smime.p7s Description: S/MIME Cryptographic Signature
hangs with 2.6.32 on debian squeeze?
I'm experiencing strange system hangs after squeeze was upgraded from
2.6.30-2 to 2.6.32.
That happens on two systems, where one is 686 and the other is amd64.
Anybody else with overall strange behaviour with the new kernel?
So, I reinstalled the 2.6.30-2 kernel which prematurely I admit, I purged.
Now for a couple of modules which I need, I need also to install the
headers but it doesn't seem to work.
It says: linux-headers-2.6.30-2-686: Depends: linux-kbuild-2.6.30 which
is a virtual package.
How can I get over that?
Thanks!
G.
aris:/home/encmp/gpall# aptitude install linux-headers-2.6.30-2-686
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initializing package states... Done
The following packages are BROKEN:
linux-headers-2.6.30-2-686
The following NEW packages will be installed:
linux-headers-2.6.30-2-common{a}
0 packages upgraded, 2 newly installed, 0 to remove and 8 not upgraded.
Need to get 4,029kB of archives. After unpacking 25.6MB will be used.
The following packages have unmet dependencies:
linux-headers-2.6.30-2-686: Depends: linux-kbuild-2.6.30 which is a
virtual package.
The following actions will resolve these dependencies:
Keep the following packages at their current version:
linux-headers-2.6.30-2-686 [Not Installed]
Score is -9881
Accept this solution? [Y/n/q/?]
No packages will be installed, upgraded, or removed.
0 packages upgraded, 0 newly installed, 0 to remove and 8 not upgraded.
Need to get 0B of archives. After unpacking 0B will be used.
smime.p7s
Description: S/MIME Cryptographic Signature
Re: Problem with tty when debian testing(2.6.32-trunk kernel) starts up
Jeffrey Cao wrote: When I upgraded to 2.6.32-trunk kernel in debian testing, I got a problem with tty on startup. After power on, it jumpts to tty4 or tty5(one time to tty4, and another time to tty5) before printing the following message: INIT: Entering runlevel 2 Why does it jumpt to tty4/tty5 rather than just stay on tty1? Jeffrey I noticed that too... smime.p7s Description: S/MIME Cryptographic Signature
Re: hangs with 2.6.32 on debian squeeze?
godo wrote:
Γιώργος Πάλλας wrote:
...
So, I reinstalled the 2.6.30-2 kernel which prematurely I admit, I
purged.
Now for a couple of modules which I need, I need also to install the
headers but it doesn't seem to work.
It says: linux-headers-2.6.30-2-686: Depends: linux-kbuild-2.6.30
which is a virtual package.
How can I get over that?
Thanks!
G.
aris:/home/encmp/gpall# aptitude install linux-headers-2.6.30-2-686
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initializing package states... Done
The following packages are BROKEN:
linux-headers-2.6.30-2-686
The following NEW packages will be installed:
linux-headers-2.6.30-2-common{a}
0 packages upgraded, 2 newly installed, 0 to remove and 8 not upgraded.
Need to get 4,029kB of archives. After unpacking 25.6MB will be used.
The following packages have unmet dependencies:
linux-headers-2.6.30-2-686: Depends: linux-kbuild-2.6.30 which is a
virtual package.
The following actions will resolve these dependencies:
Keep the following packages at their current version:
linux-headers-2.6.30-2-686 [Not Installed]
Score is -9881
Accept this solution? [Y/n/q/?]
No packages will be installed, upgraded, or removed.
0 packages upgraded, 0 newly installed, 0 to remove and 8 not upgraded.
Need to get 0B of archives. After unpacking 0B will be used.
Hi,
I have installed linux-headers-2.6.30-2-686 and linux-kbuild-2.6.30 in
Squeeze.
I can't recall whether I have installed through Synaptic or manually
download linux-kbuild-2.6.30.
http://packages.debian.org/lenny-backports/linux-kbuild-2.6.30
Maybe it is not the best way but it work's in my case.
Thank you! That worked fine (dpkg -i ...)!
G.
smime.p7s
Description: S/MIME Cryptographic Signature
Re: boot hangs on pass phrase to unlock encrypted root
k b wrote: hello, i can no longer boot my computer due to the problem in the subject line. i´ve looked in the initramfs-tools bug reports, but can not find anything exactly like this. but im not completly sure what package i should look for the problem in. the problem is the boot hangs on the luks passphrase. i key the passphrase and press enter, and nothing happens. before, when the system booted, upon entering the passphrase and pressing enter, i would get a newline, and soon after the message that the disk was unlocked or that the passphrase is wrong. but now i dont even get the newline. im running debian/squeeze, kernel linux-image-2.6.26-2-amd64, Intel(R) Celeron(R) CPU 2.80GHz by the way, i can mount the encrypted disk from a live cd with out any problems using cryptsetup, vgscan and mount. the only thing related i found on the internet, is related to a usb keyboard, but im using the integrated keyb of my lap. http://ubuntuforums.org/showthread.php?t=1228063 thanks Did you upgrade lately? Does the laptop keyboard work at grub? Try connecting a usb keyboard to your laptop and see what happens. Reset to default your BIOS. Take out the battery from your laptop for a couple of hours. Pray to this, void of meaning universe... smime.p7s Description: S/MIME Cryptographic Signature
problem installing nvidia-glx on a debian testing with 2.6.32 kernel
Hello to everybody!
I'm trying to install nvidia-glx in order to make my GeForce FX 5200
play, but I bump on this curious situation: What is this irrelevant
2.6.26-2-686 kernel that is mentioned? My system runs 2.6.32-trunk-686...
any ideas?
aris:/home/encmp/gpall# aptitude install nvidia-glx
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initializing package states... Done
The following packages are BROKEN:
nvidia-kernel-2.6.26-2-686 xserver-xorg-core
The following NEW packages will be installed:
nvidia-glx nvidia-kernel-common{a}
0 packages upgraded, 3 newly installed, 0 to remove and 0 not upgraded.
Need to get 10.3MB of archives. After unpacking 30.8MB will be used.
The following packages have unmet dependencies:
nvidia-kernel-2.6.26-2-686: Depends: linux-image-2.6.26-2-686 which is
a virtual package.
xserver-xorg-core: Conflicts: xserver-xorg-video-2 which is a virtual
package.
The following actions will resolve these dependencies:
Keep the following packages at their current version:
nvidia-glx [Not Installed]
nvidia-kernel-2.6.26-2-686 [Not Installed]
Score is -9871
Accept this solution? [Y/n/q/?]
No packages will be installed, upgraded, or removed.
0 packages upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Need to get 0B of archives. After unpacking 0B will be used.
Do you want to continue? [Y/n/?]
Writing extended state information... Done
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initializing package states... Done
aris:/home/encmp/gpall#
smime.p7s
Description: S/MIME Cryptographic Signature
Re: problem installing nvidia-glx on a debian testing with 2.6.32 kernel
Jerome BENOIT wrote: In Sid, there is a package ready to be built to do so. thanks for the answers! I guess you mean the nvidia-kernel-source package. Well, I installed it, and then: # module-assistant auto-install nvidia That fails, and I cant understand why... Again any ideas would be very appreciated! aris:/var/cache/modass# cat nvidia-kernel-source.buildlog.2.6.32-trunk-686.1264422219 /usr/bin/make -f debian/rules clean make[1]: Entering directory `/usr/src/modules/nvidia-kernel' # select which makefile to use. rm -f /usr/src/modules/nvidia-kernel/Makefile || true if [ 6 = 6 ]; then \ ln -s Makefile.kbuild Makefile ; \ fi if [ 6 = 4 ]; then \ ln -s Makefile.nvidia Makefile ; \ fi if [ -e patch-stamp ]; then \ dpatch deapply-all ; \ rm -rf patch-stamp debian/patched ; \ fi if [ -f /usr/src/modules/nvidia-kernel/debian/control.template ]; then \ cp /usr/src/modules/nvidia-kernel/debian/control.template /usr/src/modules/nvidia-kernel/debian/control; \ fi dh_testroot rm -f build-stamp configure-stamp /usr/bin/make clean SYSSRC=/lib/modules/2.6.32-trunk-686/build -C /usr/src/modules/nvidia-kernel/ -f Makefile make[2]: Entering directory `/usr/src/modules/nvidia-kernel' make[2]: Leaving directory `/usr/src/modules/nvidia-kernel' rm -f /usr/src/modules/nvidia-kernel//Makefile || true; rm /usr/src/modules/nvidia-kernel//gcc-check rm /usr/src/modules/nvidia-kernel//cc-sanity-check dh_clean rm /usr/src/modules/nvidia-kernel/debian/control rm /usr/src/modules/nvidia-kernel/debian/dirs rm /usr/src/modules/nvidia-kernel/debian/override make[1]: Leaving directory `/usr/src/modules/nvidia-kernel' echo ROOT_CMD = ROOT_CMD = /usr/bin/make -f debian/rules binary_modules make[1]: Entering directory `/usr/src/modules/nvidia-kernel' # select which makefile to use. rm -f /usr/src/modules/nvidia-kernel/Makefile || true if [ 6 = 6 ]; then \ ln -s Makefile.kbuild Makefile ; \ fi if [ 6 = 4 ]; then \ ln -s Makefile.nvidia Makefile ; \ fi if ! gcc-4.3 -v 2 /dev/null ; then \ echo Compiler gcc-4.3 does not exist on the system ; \ exit 1; \ fi if [ -f /usr/src/modules/nvidia-kernel/debian/control.template ]; then \ cp /usr/src/modules/nvidia-kernel/debian/control.template /usr/src/modules/nvidia-kernel/debian/control; \ fi if [ i686 = x86_64 ]; then \ cp /usr/src/modules/nvidia-kernel/nv-kernel.o.x86_64 /usr/src/modules/nvidia-kernel/nv-kernel.o ; \ fi touch
Re: problem installing nvidia-glx on a debian testing with 2.6.32 kernel
Vincent Lefevre wrote: Hi, On 2010-01-25 14:26:56 +0200, Γιώργος Πάλλας wrote: I guess you mean the nvidia-kernel-source package. Well, I installed it, and then: # module-assistant auto-install nvidia That fails, and I cant understand why... Again any ideas would be very appreciated! I don't have that in the installation steps: # apt-get install nvidia-kernel-common module-assistant # m-a -i -t prepare # m-a clean,a-i -i -t -f nvidia-kernel-source # depmod -a # apt-get install nvidia-glx nvidia-glx-dev nvidia-xconfig # dpkg-reconfigure xserver-xorg # nvidia-xconfig When I need to reinstall the drivers (e.g. after a kernel upgrade), I rerun steps 2 to 4. This has always worked for me. Same thing happens. This time, just to be sure, I purged all nvidia related files first, before going through the above steps. Again, the building process fails without any specific reason (maybe some incompatibility with the gcc version?): aris:/usr/src# aptitude install nvidia-kernel-common module-assistant Reading package lists... Done Building dependency tree Reading state information... Done Reading extended state information Initializing package states... Done The following NEW packages will be installed: module-assistant nvidia-kernel-common 0 packages upgraded, 2 newly installed, 0 to remove and 0 not upgraded. Need to get 113kB/117kB of archives. After unpacking 618kB will be used. Writing extended state information... Done Get:1 http://postmortem.csd.auth.gr testing/main module-assistant 0.11.2 [113kB] Fetched 113kB in 0s (3,865kB/s) Selecting previously deselected package module-assistant. (Reading database ... 107512 files and directories currently installed.) Unpacking module-assistant (from .../module-assistant_0.11.2_all.deb) ... Selecting previously deselected package nvidia-kernel-common. Unpacking nvidia-kernel-common (from .../nvidia-kernel-common_20080825+1_all.deb) ... Processing triggers for man-db ... Setting up module-assistant (0.11.2) ... Setting up nvidia-kernel-common (20080825+1) ... update-rc.d: warning: nvidia-kernel stop runlevel arguments (1) do not match LSB Default-Stop values (none) Reading package lists... Done Building dependency tree Reading state information... Done Reading extended state information Initializing package states... Done Writing extended state information... Done aris:/usr/src# m-a -i -t prepare Getting source for kernel version: 2.6.32-trunk-686 Kernel headers available in /lib/modules/2.6.32-trunk-686/build apt-get -y install build-essential Reading package lists... Done Building dependency tree Reading state information... Done build-essential is already the newest version. 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. Done! aris:/usr/src# m-a clean,a-i -i -t -f nvidia-kernel-source . Updated infos about 1 packages Getting source for kernel version: 2.6.32-trunk-686 Kernel headers available in /lib/modules/2.6.32-trunk-686/build apt-get -y install build-essential Reading package lists... Done Building dependency tree Reading state information... Done build-essential is already the newest version. 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. Done! download Reading package lists... Done Building dependency tree Reading state information... Done The following extra packages will be installed: debhelper dpatch html2text intltool-debian po-debconf Suggested packages: dh-make libmail-box-perl Recommended packages: fakeroot patchutils nvidia-glx kernel-package devscripts libmail-sendmail
Re: LVM+RAID+CRYPT
Stan Hoeppner wrote: Sjors van der Pluijm put forth on 1/8/2010 5:13 AM: 3. Is it ok to have swap and /boot on an encrypted LVM? Never run encryption on swap. Doing so merely burdens performance. I doubt even NSA, CIA, MI6 encrypt swap partitions on workstations. I've never tried to boot from an encrypted /boot, so I really can't say if it would work or not. Why can't/won't you create 3 partitions? [boot] 100MB mounted as /boot normal ext2 [swap] 1-8GB mounted as normal swap partition [root] [remaining space] mounted as /root and encrypted however you like I run a couple of identical machines, some with full disk encryption (i.e. everything including swap except /boot which you cannot encrypt) and some where only home is encrypted with LUKS. Never noticed any performance impact. I think that swap encryption is *mandatory* for the reason of there being written many things that shouldn't in case they are sensitive. And I guess this why the approach of the debian installer should you choose to encrypt includes swap encryption. G. smime.p7s Description: S/MIME Cryptographic Signature
KDE auto-login not working after kernel upgrade (2.6.30-2 - 2.6.32)
A strange problem is occuring: After a kernel upgrade from 2.6.30-2 to 2.6.32 on an updated debian testing system, the user auto-login feature of KDE stopped working - instead I am presented with the kdm login screen. If I reboot with the old kernel, kde auto-logins the user fine. The problem is reproduced: Booting with the new kernel always gives me the kdm login screen. Booting with the old one, it works. Any ideas how could I investigate something like this? thanks and a happy new year! G. smime.p7s Description: S/MIME Cryptographic Signature
chattr to a file after converting to ext4 gives scary trace in syslog
Hello all! I just converted a data partition from ext3 to ext4, and after that I tried to convert a file to use extents: mordor:/home/gpall# lsattr file --- file mordor:/home/gpall# chattr +e file mordor:/home/gpall# lsattr file -e- file I calculated the md5 of the file before and after, and it was OK, but syslog had written this after the chattr command: Dec 27 10:24:48 mordor kernel: [ 2456.957666] [ cut here ] Dec 27 10:24:48 mordor kernel: [ 2456.957683] WARNING: at /build/buildd/linux-2.6-2.6.30/debian/build/source_amd64_none/fs/inode.c:1185 generic_delete_inode+0x64/0x168() Dec 27 10:24:48 mordor kernel: [ 2456.957689] Hardware name: Studio 1737 Dec 27 10:24:48 mordor kernel: [ 2456.957692] Modules linked in: ext4 jbd2 crc16 michael_mic arc4 ecb i915 drm i2c_algo_bit vboxnetadp vboxnetflt vboxdrv acpi_cpufreq cpufreq_c Dec 27 10:24:48 mordor kernel: al fan thermal_sys [last unloaded: pcmcia_core] Dec 27 10:24:48 mordor kernel: [ 2456.957861] Pid: 4704, comm: chattr Tainted: P 2.6.30-2-amd64 #1 Dec 27 10:24:48 mordor kernel: [ 2456.957866] Call Trace: Dec 27 10:24:48 mordor kernel: [ 2456.957875] [802d1aba] ? generic_delete_inode+0x64/0x168 Dec 27 10:24:48 mordor kernel: [ 2456.957882] [802d1aba] ? generic_delete_inode+0x64/0x168 Dec 27 10:24:48 mordor kernel: [ 2456.957892] [80242393] ? warn_slowpath_common+0x77/0xa3 Dec 27 10:24:48 mordor kernel: [ 2456.957899] [802d1aba] ? generic_delete_inode+0x64/0x168 Dec 27 10:24:48 mordor kernel: [ 2456.957930] [a0704f17] ? ext4_ext_migrate+0x637/0x6b4 [ext4] Dec 27 10:24:48 mordor kernel: [ 2456.957955] [a06f5935] ? ext4_ioctl+0x2cf/0x621 [ext4] Dec 27 10:24:48 mordor kernel: [ 2456.957963] [802cc136] ? vfs_ioctl+0x21/0x6c Dec 27 10:24:48 mordor kernel: [ 2456.957970] [802cc5ac] ? do_vfs_ioctl+0x42b/0x464 Dec 27 10:24:48 mordor kernel: [ 2456.957976] [802cc636] ? sys_ioctl+0x51/0x70 Dec 27 10:24:48 mordor kernel: [ 2456.957985] [8020fa42] ? system_call_fastpath+0x16/0x1b Dec 27 10:24:48 mordor kernel: [ 2456.957990] ---[ end trace 817d37482a92ff42 ]--- Has anybody else bumped on this? Is this safe? Is it to be expected? I am running updated debian squeeze with 2.6.30-2-amd64 stock kernel. Ciao! G. smime.p7s Description: S/MIME Cryptographic Signature
debian mirroring πρόβλημα
Καλησπέρα σε όλους και χρόνια πολλά! Τις τελευταίες μέρες (3-4) παίρνω το εξής στα logs από το script που ενημερώνει το mirror μου από κάποιο άλλο. rsync: connection unexpectedly closed (250 bytes received so far) [generator] rsync error: error in rsync protocol data stream (code 12) at io.c(600) [generator=3.0.6] To mirroring διακόπτεται βέβαια και παρόλο που άλλαξα τo mirror από το οποίο τραβάω, εξακολουθεί πάντα το ίδιο error. Μήπως τυχαίνει κανείς σας να έχει πετύχει το ίδιο πρόβλημα; Σημειώνω ότι για την ανανέωση του mirror χρησιμοποιώ το ftpsync που δίνει το debian. G. smime.p7s Description: S/MIME Cryptographic Signature
Re: debian mirroring πρόβλημα
Γιώργος Πάλλας wrote: Καλησπέρα σε όλους και χρόνια πολλά! Τις τελευταίες μέρες (3-4) παίρνω το εξής στα logs από το script που ενημερώνει το mirror μου από κάποιο άλλο. rsync: connection unexpectedly closed (250 bytes received so far) [generator] rsync error: error in rsync protocol data stream (code 12) at io.c(600) [generator=3.0.6] To mirroring διακόπτεται βέβαια και παρόλο που άλλαξα τo mirror από το οποίο τραβάω, εξακολουθεί πάντα το ίδιο error. Μήπως τυχαίνει κανείς σας να έχει πετύχει το ίδιο πρόβλημα; Σημειώνω ότι για την ανανέωση του mirror χρησιμοποιώ το ftpsync που δίνει το debian. G. χίλια συγγνώμη: δικιά μου ανοησία - ήμουν σίγουρος ότι είχα χώρο στο δίσκο και το απέκλεισα χωρίς να το κοιτάξω... smime.p7s Description: S/MIME Cryptographic Signature
Re: debian mirroring πρόβλημα
Christos Ricudis wrote: Giwrgo, elpizw na peirakses to ftpsync script tou debian wste na pairneis mirror *mono* apo to branch pou se endiaferei kai *mono* gia tis arxitektonikes pou se endiaferei, giati alliws katebazei to sympan... προφανώς... τα 250G του i386 και του amd64 είναι αρκετά ;-) smime.p7s Description: S/MIME Cryptographic Signature
Re: Monitor file system free space
m...@neidorff.com wrote: I had a problem with my (small) disk filling up and crashing my mail server. So, I wrote a simple perl script to take the output of 'df -h' and look for a % full that is greater than 97% and send me an e-mail if the % is greater than 97. I run the script daily from crontab. It works fine. Yesterday I downloaded an iso image (which brought the file system fill from 93% to 100% so that the mail server could not send me the e-mail warning. Is there another way to approach this problem (a different tool?)so that I can be notified BEFORE the file system is too full? Thanks for any thoughts, Mark Hello Mark! A general solution is to have your user's home directories (/home/user1, /home/user2) on a separate partition. This way, when you fill it, no system process will crash... smime.p7s Description: S/MIME Cryptographic Signature
merging LVM snapshots back to origin
Does anybody know the status of this wish item? For the time being, on a squeeze server, I see that lvconvert can only convert a linear volume to a snapshot and not vice versa. G. smime.p7s Description: S/MIME Cryptographic Signature
Re: eth1 disappears after kernel upgrade
Anthony Campbell wrote: On 10 Nov 2009, Γιώργος Πάλλας wrote: Anthony Campbell wrote: On 10 Nov 2009, Γιώργος Πάλλας wrote: Anthony Campbell wrote: When I installed kernel 2.6.30-2-686-bigmem my eth1 disappeared, so no wireless. Not a problem in practice because I simply revert to 2.6.30-1-686-bigmem and it's back, but I'd like to understand what is happening. Does it need a bug report or is it a feature? Anthony It seems to me strange to lose your wireless just from 2.6.30-1 to 2.6.30-2. Do you remember if your wless was playing out of the box with 2.6.30-1 ? Do you happen to use any of those broadcom sta modules? Boot into 2.6.30-1 where your wireless is recognized and send us an lspci -v (just the wireless part) Giorgos Yes, you are right. I'd forgotten this was a Broadcom wireless. I'm not sure if the modules are available for the latest kernel. Anthony I also have such a card. I have noted the following instructions for doing after a kernel upgrade: aptitude purge everything-with-broadcom-in-the-package-name aptitude install broadcom-sta-common broadcom-sta-source # module-assistant prepare broadcom-sta # module-assistant auto-install broadcom-sta and reboot! (you need internet to get the packages, so first boot the kernel where wifi functions, do the first two commands, then reboot to the new and issue the next to commands) I seem to remember doing that in the past but the broadcom packages are apparently not available at the moment. Anthony Oh, this must be related to the thread 2.6.31 kernels about the broadcom packages disappearing from the repositories. Why the heck was that?? smime.p7s Description: S/MIME Cryptographic Signature
Re: packages cannot be authenticated
Sthu Deus wrote: Good day. I think after $ sudo apt-get update i have got a message on $ sudo -t stable apt-get upgrade stating that some packages can not be authenticated. Before - all was all right: i was updating and upgrading - no problems, but now I have the problem. Sometimes this is caused by something transferred wrongly to your pc, or even worse to the mirror you're using. Try again tomorrow. smime.p7s Description: S/MIME Cryptographic Signature
Re: eth1 disappears after kernel upgrade
Anthony Campbell wrote: When I installed kernel 2.6.30-2-686-bigmem my eth1 disappeared, so no wireless. Not a problem in practice because I simply revert to 2.6.30-1-686-bigmem and it's back, but I'd like to understand what is happening. Does it need a bug report or is it a feature? Anthony It seems to me strange to lose your wireless just from 2.6.30-1 to 2.6.30-2. Do you remember if your wless was playing out of the box with 2.6.30-1 ? Do you happen to use any of those broadcom sta modules? Boot into 2.6.30-1 where your wireless is recognized and send us an lspci -v (just the wireless part) Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: eth1 disappears after kernel upgrade
Anthony Campbell wrote: On 10 Nov 2009, Γιώργος Πάλλας wrote: Anthony Campbell wrote: When I installed kernel 2.6.30-2-686-bigmem my eth1 disappeared, so no wireless. Not a problem in practice because I simply revert to 2.6.30-1-686-bigmem and it's back, but I'd like to understand what is happening. Does it need a bug report or is it a feature? Anthony It seems to me strange to lose your wireless just from 2.6.30-1 to 2.6.30-2. Do you remember if your wless was playing out of the box with 2.6.30-1 ? Do you happen to use any of those broadcom sta modules? Boot into 2.6.30-1 where your wireless is recognized and send us an lspci -v (just the wireless part) Giorgos Yes, you are right. I'd forgotten this was a Broadcom wireless. I'm not sure if the modules are available for the latest kernel. Anthony I also have such a card. I have noted the following instructions for doing after a kernel upgrade: aptitude purge everything-with-broadcom-in-the-package-name aptitude install broadcom-sta-common broadcom-sta-source # module-assistant prepare broadcom-sta # module-assistant auto-install broadcom-sta and reboot! (you need internet to get the packages, so first boot the kernel where wifi functions, do the first two commands, then reboot to the new and issue the next to commands) smime.p7s Description: S/MIME Cryptographic Signature
Re: Slow booting
Kevin Buhr wrote: Kamil Kułaga teodoz...@gmail.com writes: After init of soundcard and before swap activation i have 90 seconds hole. Is it normal? Can I configure kernel to somewhat avoid this? Unfortunately, I think adding a large swap partition (yours is 6GB) can be a slow operation under Linux. Google for slow swapon and you'll see several complaints. Does this happen under specific conditions? I'm asking because my swap is 5GB on a debian squeeze 64bit box, and it inits in an instant. Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: 2.6.31 kernels
Colin wrote: Alex Samad wrote: isn't there some issue with the broadcom-sta drivers and .31 ? The broadcom-sta source code that was included in non-free Debian would no longer compile with kernel 2.6.31. That's why they've been working on the b43 driver extensively in the 2.6.32 kernel so the broadcom-sta driver will no longer be necessary. So, if I understand correctly, I should not upgrade to 2.6.31 stock kernel, when that is released for testing, right? Giorgos smime.p7s Description: S/MIME Cryptographic Signature
Re: Squeeze: Sub - Process rred received segmentation fault
Jason Filippou wrote: Hello, I was executing aptitude update and noticed the followin output after a sequence of 'hit's, 'ign's and 'get's: Hit http://http.us.debian.org stable/non-free Packages Get:2 http://http.us.debian.org unstable/main 2009-11-02-1520.49.pdiff [18.3kB] Get:3 http://http.us.debian.org unstable/contrib 2009-11-04-0016.39.pdiff [74.3kB] Get:4 http://http.us.debian.org testing/main 2009-11-04-0016.39.pdiff [10.2MB] Get:5 http://http.us.debian.org unstable/non-free 2009-11-04-0016.39.pdiff [151kB] Get:6 http://http.us.debian.org experimental/main 2009-11-02-2358.21.pdiff [3,694B] Get:7 http://http.us.debian.org unstable/main 2009-11-02-1520.49.pdiff [18.3kB] Get:8 http://http.us.debian.org testing/main 2009-11-04-0016.39.pdiff [10.2MB] Get:9 http://http.us.debian.org unstable/contrib 2009-11-04-0016.39.pdiff [74.3kB] Get:10 http://http.us.debian.org unstable/non-free 2009-11-04-0016.39.pdiff [151kB] Get:11 http://http.us.debian.org experimental/main 2009-11-02-2358.21.pdiff [3,694B] E: Method rred has died unexpectedly! E: Sub-process rred received a segmentation fault. Am I to assume that there's some kind of issue with aptitude on squeeze? Should I wait until the issue is resolved before performing a safe-upgrade? Thank you. There is already a relevant bug submitted for package 'apt'. smime.p7s Description: S/MIME Cryptographic Signature
Re: Tgz Backups
Merciadri Luca wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I often backup everything of my Debian systems with the tgz utility. Is it possible to boot a LiveCD, to put the whole content of the tgz (assuming it is complete) on the partition (assuming there is enough free space), and to reboot (assuming GRUB is okay)? I would like to hear others' experience about this. yes, it is possible. I also keep backup of my system using tar* and I have at least once successfully restored a working, booting system after a failure. * tar use for backup: tar cvSj --exclude=/tmp --exclude=/proc --exclude=/lost+found --exclude=/dev --exclude=/sys /tmp/backup.tar.bz2 / and this is what I actually use in order to store the system backup in 4GB encrypted chunks, along with an encrypted index of all the files and an md5sum checksum of the archive (I love one-liners :-) ): time tar cvSj --exclude=/home/gpall/.local/share/Trash --exclude=/tmp --exclude=/proc --exclude=/lost+found --exclude=/dev --exclude=/mnt --exclude=/media --exclude=/sys / 2 (gpg -z 9 -e -r gpall -o sysbackup-filelist.gpg) | gpg -z 0 -e -r gpall | tee (split -b 4000m - sysbackup.tbg.split_) | md5sum sysbackup.tbg.md5 smime.p7s Description: S/MIME Cryptographic Signature
Re: Tgz Backups
Merciadri Luca wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Γιώργος Πάλλας gp...@ccf.auth.gr writes: Merciadri Luca wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I often backup everything of my Debian systems with the tgz utility. Is it possible to boot a LiveCD, to put the whole content of the tgz (assuming it is complete) on the partition (assuming there is enough free space), and to reboot (assuming GRUB is okay)? I would like to hear others' experience about this. yes, it is possible. I also keep backup of my system using tar* and I have at least once successfully restored a working, booting system after a failure. * tar use for backup: tar cvSj --exclude=/tmp --exclude=/proc --exclude=/lost+found --exclude=/dev --exclude=/sys /tmp/backup.tar.bz2 / and this is what I actually use in order to store the system backup in 4GB encrypted chunks, along with an encrypted index of all the files and an md5sum checksum of the archive (I love one-liners :-) ): time tar cvSj --exclude=/home/gpall/.local/share/Trash --exclude=/tmp --exclude=/proc --exclude=/lost+found --exclude=/dev --exclude=/mnt --exclude=/media --exclude=/sys / 2 (gpg -z 9 -e -r gpall -o sysbackup-filelist.gpg) | gpg -z 0 -e -r gpall | tee (split -b 4000m - sysbackup.tbg.split_) | md5sum sysbackup.tbg.md5 Nice. How do you restore your system with these archives? restore?? never thought about that... :-P restore is easy: 1. live cd boot 2. create partitions and filesystems 3. restore system: cat /media/MYUSBDISK/sysbackup.tbg_split* | gpg -d | tar xvjS 4. fix GRUB PS. I would recommend backup and restoring with something like Clonezilla. I use my own way because I run a fully encrypted LUKS system and bit-to-bit copy has a little overhead ;-) smime.p7s Description: S/MIME Cryptographic Signature
Re: reportbug reportbug?
jagginess wrote: Is there a way to determine if reportbug is misbehaving? I dont like having to copy and paste the template made from console everytime to my icedove thunderbird.. are there any alternatives? I'm at a lost.. I had reportbug working and now i only get smtp-auth.. I have no idea if its my isp's fault or if its reportbug :/.. it's funny its been working dozens of times.. Most probably something changed in the way reportbug was sending the email report. Send us the error message please. BTW, here is my /etc/reportbug.cfg that works (comment lines omitted): submit query-bts cc config-files compress email gp...@ccf.auth.gr realname Giorgos Pallas smtphost localhost verify smime.p7s Description: S/MIME Cryptographic Signature
Re: Debian 5
Tanco wrote: Hi, I am writing to report that the ia64 dvd image is broken, somehow after about 3-4 separate downloads none of the images was bootable, you put the DVD in / mount it to vbox , the power on and nothing happends , then the system hangs.. after that i downloaded the i386 img and installed the 32 bit version, this is actually an old issue, It happend to me somewhere before the official release but I didn't give it much thought since it was still in rc. but now I think it is a big issue... best regards Tanco I guess you already know that ia64 arch is for some rare special Intel CPUs. Are you 100% positive that you don't want amd64 instead? smime.p7s Description: S/MIME Cryptographic Signature
Re: Regular high load peaks on servers
Julien wrote: Hi, Since quite a long time now, we observe the same phenomenon on three web servers we have on two different places. They regularly have high load peaks, until 20 to 50. These peaks append very regularly (from once a day to several an hour), and we can't explain why. It doesn't seem to be linked to a special increase in traffic or web requests. Two of the three servers are load-balanced web frontends running apache with nfs mounts. The third is an autonomous server with web, mail, mysql… services, without nfs. The three run under Debian Lenny. I know nobody could really solve this problem without access to the machines and logs, but I wonder if someone already experienced this sort of regular load peaks. Thanks in advance for any help, Try installing sysstat, and use the iostat utility to check your disk's usage during those peaks. High load is caused from high cpu utilization, and from I/O util. Try also to stop cron for several hours, in order to be sure that no cron job causes the load. G. smime.p7s Description: S/MIME Cryptographic Signature
Re: Could you recommend CD/DVD writer program?
Matthew Moore wrote: On Friday October 16 2009 10:56:29 am Boyd Stephen Smith Jr. wrote: On Friday 16 October 2009 11:03:13 Matthew Moore wrote: On Friday October 16 2009 9:11:05 am Lee Winter wrote: And you are adding to it in order to shorten it? This thread is much like any infinite set. Adding a finite number of elements to it does not change it's size. By size do you mean cardinality or ordinality? The size of a set is a cardinal number. Generally, when you talk about the size of a set, you are referring to it's cardinality. You can take a set of (infinite) ordinality aleph-naught and add single new element that is greater than all the others (generally denoted 1*) and get a new set of the same cardinality but an ordinality of aleph-naught + 1. I suppose that you are thinking of the length of the thread as the length of the maximal ascending (or descending) ordered chain (here ab if and only if b is a reply to a). I admit that I interpreted this thread has to stop! as this thread must stop having additional members added to it! and not this thread must have a finite maximal chain!. MM stop the damn thread! --a kitty smime.p7s Description: S/MIME Cryptographic Signature
Re: How to REALLY restart networking?
Klistvud wrote: Howdie, fellow Debianites! I'm running Lenny with Gnome on two boxes, both connected wirelessly via network-manager to my home router. Problem: about once to thrice a day, one or both of the boxes randomly disconnect (they are using the proprietary Broadcom wl.ko and ndiswrapper, respectively) and I can't seem to make them reconnect. The nm-applet doesn't show the available networks anymore. Which are the services I should restart to make networking restart from scratch? What I've tried so far and *doesn't work*: -- running /etc/init.d/networking restart -- running /etc/init.d/networking-dispatcher restart -- killing the Gnome nm-applet and starting another instance thereof from a Gnome-Terminal -- running all of the above in every picturesque combination I could think of -- running ifup, ifdown -- logging out and logging in again -- turning wireless off and on with the hardware button on the box which has one -- all to no avail -- rebooting - oddly enough, this helps. After a reboot, network- manager reconnects flawlessly, and everything runs fine for another day or so. How does one REALLY restart wireless in Debian without rebooting? TIA Maybe the kernel module blacks out - try to find it, unload it and reload it using modprobe... [I could also very well just talk BS!] G. smime.p7s Description: S/MIME Cryptographic Signature
Re: How to REALLY restart networking?
Klistvud wrote: No wonder I'm afraid to touch anything in my Debian config files ... At least until I get to know my OS a little better ... Don't be afraid! # tar cf /root/etc_bckp.tar /etc and then change /etc/ files as you please! (it also helps to remember which was the change that broke your system ;-) ) smime.p7s Description: S/MIME Cryptographic Signature
