Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Bernard, 5.03.2012: > Tom H wrote: > >On Sun, Mar 4, 2012 at 8:43 AM, Bernard wrote: > >>Selim T. Erdogan wrote: > >>>Andrei POPESCU, 3.03.2012: > On Sb, 03 mar 12, 20:59:30, Bernard wrote: > >Thanks in advance for your help with the 'NetworkManager' or any > >other convenient tool that would operate on Squeeze so as to easily > >turn an Internet connexion ON and OFF > Make sure you have NO definition for eth0 in /etc/network/interfaces and > that you have: > > [ifupdown] > managed=false > > in /etc/NetworkManager/NetworkManager.conf > > If it still doesn't work after a restart[1] please attach both files. > > [1] not sure why, but in some cases new configurations did not work > properly until a full restart > >>>Or you can keep /etc/network/interfaces as you have it now and edit > >>>/etc/NetworkManager/NetworkManager.conf to have > >>> > >>>[main] > >>>plugins=ifupdown,keyfile > >>> > >>>[ifupdown] > >>>managed=true > >>> > >>>(And you have to restart Network Manager.) > >>I just tested both ways, to no avail. > >> > >>At first, I must say that my system did not have any > >> > >>NetworkManager.conf file in /etc/NetworkManager/ > >> > >>The only NetworkManager.conf file that could be found on this system was in > >> > >>/etc/dbus-1/system.d/ > >> > >>I can see the same thing on this other machine still on Lenny, where I am > >>writing this from: no /etc/NetworkManager/NetworkManager.conf file either. > >> > >>the man page for NetworkManager does not mention any such conf file. I did > >>not dare modifying /etc/dbus-1/system.d/NetworkManager.conf > >> > >>But I tested creating an /etc/NetworkManager/NetworkManager.conf, with only > >> > >> > >>[main] > >>plugins=ifupdown,keyfile > >> > >>[ifupdown] > >>managed=false > >>** > >> > >>As a first trial, I had just written: > >> > >>*** > >>[ifupdown] > >>managed=false > >>* > >> > >>in it. But maybe I should have modified permissions and owner ? Owner was > >>root and perms: -rw-r--r-- > >> > >>Besides, my /etc/network/interfaces file did not contain any definition for > >>eth0. Still, I tested removing any mention of eth0, to no avail. > >> > >>Below is my 'etc/network/interfaces' file : > >> > >>* > >> > >># This file describes the network interfaces available on your system > >># and how to activate them. For more information, see interfaces(5). > >> > >># The loopback network interface > >>auto lo > >>iface lo inet loopback > >> > >># The primary network interface > >>allow-hotplug eth0 > >>iface eth0 inet dhcp > >> > >>* > > > >The file in "/etc/dbus-1/system.d/" is useful (to systems) if you have > >systemd installed. > > > >Do you have network access? > > > >With your current setup, "managed=false" and eth0 defined in > >"/etc/network/interfaces" you can control eth0 with ifup and if down. > > > >With "managed=true", you'd have to use NM to bring up or take down eth0. > > > > > I just don't understand what happens. Whatever I try with > 'NetworkManager'... FAILS ! Besides, as previously stated, I have > tried to install 'wicd'... It did install all right, and worked as > far as disconnecting was concerned, but no way to reconnect except > on a re-boot. Now that I wish to get rid of 'wicd', just in case it > would interfer with 'NetworkManager', I can't get rid of it ! > > #apt-get remove wicd > > did remove it, so it said... if I try to remove it again, it says > that it is no longer there... however, I can still see 'wicd network > manager' on my gnome desktop, and, if I launch it, it still > disconnects my Internet link, and is still unable to reconnect it ! There may be other wicd-related packages still installed. Try dpkg --get-selections wicd* > Now, if I try > > #ifup and #ifdown eth0, it seems to work... I repeat: it "SEEMS TO > WORK" but it does not REALLY work !! Here is what I see: > > #ifup eth0 > > .. > ... > receive packet failed on eth0: Network is down > .. > DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 6 > DHCPOFFER from 192.168.1.1 > DHCPREQUEST on eth0... > DHCPACK form 192.168.1.1 > SIOCADDRT: File exists > bound to 192.168.1.15 -- renewal in 40322 seconds. > # > > at this point, if I start again the command: > > #ifup eth0 > > I get this: > > ifup: interface eth0 already configured (which I agree was expected > as normal !) > > ifdown disconfigure eth0 ; ifup reconfigures it... Except that, > whether it is configured or disconfigured, the network (the Internet > network) is up and running, as if it did not care whether ifup had > it configured or not ! (I must state that I have no wireless > connexion on this machine, just one wired ethernet connexion, no > more). > > Ooh... Well... let me see... I
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Bernard wrote: > Now that I wish to get rid of 'wicd', just in case it > would interfer with 'NetworkManager', I can't get rid of it ! > > #apt-get remove wicd > > did remove it, so it said... if I try to remove it again, it says > that it is no longer there... however, I can still see 'wicd network > manager' on my gnome desktop, and, if I launch it, it still > disconnects my Internet link, and is still unable to reconnect it ! It is probably removed but not purged meaning that configuration files that may have been customized by the local admin are not removed. You can tell if you look at it like this: $ dpkg -l wicd and if it lists it as "rc" then there are still configuration files left in the package. To remove the configuration files too it needs to be purged. # dpkg --purge wicd > #ifup and #ifdown eth0, it seems to work... I repeat: it "SEEMS TO > WORK" but it does not REALLY work !! Here is what I see: > > #ifup eth0 > .. > ... > receive packet failed on eth0: Network is down > .. > DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 6 > DHCPOFFER from 192.168.1.1 > DHCPREQUEST on eth0... > DHCPACK form 192.168.1.1 > SIOCADDRT: File exists > bound to 192.168.1.15 -- renewal in 40322 seconds. Looks okay. > at this point, if I start again the command: > > #ifup eth0 > > I get this: > > ifup: interface eth0 already configured (which I agree was expected > as normal !) Again looks okay. > ifdown disconfigure eth0 ; ifup reconfigures it... Except that, > whether it is configured or disconfigured, the network (the Internet > network) is up and running, as if it did not care whether ifup had > it configured or not ! (I must state that I have no wireless > connexion on this machine, just one wired ethernet connexion, no > more). What do you mean? You say the network is up and running but *exactly* are you seeing? Because if you "ifdown eth0" it should disable eth0. > Ooh... Well... let me see... I just finished testing something... As > previously stated, 'wicd' only works in the purpose of disconnecting > ; it never reconnects ; it does that job even though I have > supposedly removed it... I just tested disconnecting with wicd... If ifup and ifdown are working then it means that you have the interface listed in /etc/network/interfaces. If you have that then wicd and network manager will both ignore the interface, as they are programmed to do. If you want wicd to manage the interface then you have to remove it from /etc/network/interfaces. Bob signature.asc Description: Digital signature
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Tom H wrote: On Sun, Mar 4, 2012 at 8:43 AM, Bernard wrote: Selim T. Erdogan wrote: Andrei POPESCU, 3.03.2012: On Sb, 03 mar 12, 20:59:30, Bernard wrote: Thanks in advance for your help with the 'NetworkManager' or any other convenient tool that would operate on Squeeze so as to easily turn an Internet connexion ON and OFF Make sure you have NO definition for eth0 in /etc/network/interfaces and that you have: [ifupdown] managed=false in /etc/NetworkManager/NetworkManager.conf If it still doesn't work after a restart[1] please attach both files. [1] not sure why, but in some cases new configurations did not work properly until a full restart Or you can keep /etc/network/interfaces as you have it now and edit /etc/NetworkManager/NetworkManager.conf to have [main] plugins=ifupdown,keyfile [ifupdown] managed=true (And you have to restart Network Manager.) I just tested both ways, to no avail. At first, I must say that my system did not have any NetworkManager.conf file in /etc/NetworkManager/ The only NetworkManager.conf file that could be found on this system was in /etc/dbus-1/system.d/ I can see the same thing on this other machine still on Lenny, where I am writing this from: no /etc/NetworkManager/NetworkManager.conf file either. the man page for NetworkManager does not mention any such conf file. I did not dare modifying /etc/dbus-1/system.d/NetworkManager.conf But I tested creating an /etc/NetworkManager/NetworkManager.conf, with only [main] plugins=ifupdown,keyfile [ifupdown] managed=false ** As a first trial, I had just written: *** [ifupdown] managed=false * in it. But maybe I should have modified permissions and owner ? Owner was root and perms: -rw-r--r-- Besides, my /etc/network/interfaces file did not contain any definition for eth0. Still, I tested removing any mention of eth0, to no avail. Below is my 'etc/network/interfaces' file : * # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet dhcp * The file in "/etc/dbus-1/system.d/" is useful (to systems) if you have systemd installed. Do you have network access? With your current setup, "managed=false" and eth0 defined in "/etc/network/interfaces" you can control eth0 with ifup and if down. With "managed=true", you'd have to use NM to bring up or take down eth0. I just don't understand what happens. Whatever I try with 'NetworkManager'... FAILS ! Besides, as previously stated, I have tried to install 'wicd'... It did install all right, and worked as far as disconnecting was concerned, but no way to reconnect except on a re-boot. Now that I wish to get rid of 'wicd', just in case it would interfer with 'NetworkManager', I can't get rid of it ! #apt-get remove wicd did remove it, so it said... if I try to remove it again, it says that it is no longer there... however, I can still see 'wicd network manager' on my gnome desktop, and, if I launch it, it still disconnects my Internet link, and is still unable to reconnect it ! Now, if I try #ifup and #ifdown eth0, it seems to work... I repeat: it "SEEMS TO WORK" but it does not REALLY work !! Here is what I see: #ifup eth0 .. ... receive packet failed on eth0: Network is down .. DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 6 DHCPOFFER from 192.168.1.1 DHCPREQUEST on eth0... DHCPACK form 192.168.1.1 SIOCADDRT: File exists bound to 192.168.1.15 -- renewal in 40322 seconds. # at this point, if I start again the command: #ifup eth0 I get this: ifup: interface eth0 already configured (which I agree was expected as normal !) ifdown disconfigure eth0 ; ifup reconfigures it... Except that, whether it is configured or disconfigured, the network (the Internet network) is up and running, as if it did not care whether ifup had it configured or not ! (I must state that I have no wireless connexion on this machine, just one wired ethernet connexion, no more). Ooh... Well... let me see... I just finished testing something... As previously stated, 'wicd' only works in the purpose of disconnecting ; it never reconnects ; it does that job even though I have supposedly removed it... I just tested disconnecting with wicd... and now that it is really disconnected, I can REALLY reconnect with '#ifup eth0' and REALLY disconnect using '#ifdown eth0', back and forth as many times as I wish... provided that I have first disconnected - using 'wicd' - the connexion that occured automatically upon boot ! I hope that these symp
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
On Sun, Mar 4, 2012 at 8:43 AM, Bernard wrote: > Selim T. Erdogan wrote: >> Andrei POPESCU, 3.03.2012: >>> On Sb, 03 mar 12, 20:59:30, Bernard wrote: Thanks in advance for your help with the 'NetworkManager' or any other convenient tool that would operate on Squeeze so as to easily turn an Internet connexion ON and OFF >>> >>> Make sure you have NO definition for eth0 in /etc/network/interfaces and >>> that you have: >>> >>> [ifupdown] >>> managed=false >>> >>> in /etc/NetworkManager/NetworkManager.conf >>> >>> If it still doesn't work after a restart[1] please attach both files. >>> >>> [1] not sure why, but in some cases new configurations did not work >>> properly until a full restart >> >> Or you can keep /etc/network/interfaces as you have it now and edit >> /etc/NetworkManager/NetworkManager.conf to have >> >> [main] >> plugins=ifupdown,keyfile >> >> [ifupdown] >> managed=true >> >> (And you have to restart Network Manager.) > > I just tested both ways, to no avail. > > At first, I must say that my system did not have any > > NetworkManager.conf file in /etc/NetworkManager/ > > The only NetworkManager.conf file that could be found on this system was in > > /etc/dbus-1/system.d/ > > I can see the same thing on this other machine still on Lenny, where I am > writing this from: no /etc/NetworkManager/NetworkManager.conf file either. > > the man page for NetworkManager does not mention any such conf file. I did > not dare modifying /etc/dbus-1/system.d/NetworkManager.conf > > But I tested creating an /etc/NetworkManager/NetworkManager.conf, with only > > > [main] > plugins=ifupdown,keyfile > > [ifupdown] > managed=false > ** > > As a first trial, I had just written: > > *** > [ifupdown] > managed=false > * > > in it. But maybe I should have modified permissions and owner ? Owner was > root and perms: -rw-r--r-- > > Besides, my /etc/network/interfaces file did not contain any definition for > eth0. Still, I tested removing any mention of eth0, to no avail. > > Below is my 'etc/network/interfaces' file : > > * > > # This file describes the network interfaces available on your system > # and how to activate them. For more information, see interfaces(5). > > # The loopback network interface > auto lo > iface lo inet loopback > > # The primary network interface > allow-hotplug eth0 > iface eth0 inet dhcp > > * The file in "/etc/dbus-1/system.d/" is useful (to systems) if you have systemd installed. Do you have network access? With your current setup, "managed=false" and eth0 defined in "/etc/network/interfaces" you can control eth0 with ifup and if down. With "managed=true", you'd have to use NM to bring up or take down eth0. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAOdo=sxg5pw_moy3c-kvjz6pzu0z+xug+8pnug6_fd+y4uy...@mail.gmail.com
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Bernard, 4.03.2012: > Selim T. Erdogan wrote: > >Andrei POPESCU, 3.03.2012: > >>On Sb, 03 mar 12, 20:59:30, Bernard wrote: > >>>Thanks in advance for your help with the 'NetworkManager' or any > >>>other convenient tool that would operate on Squeeze so as to easily > >>>turn an Internet connexion ON and OFF > >>Make sure you have NO definition for eth0 in > >>/etc/network/interfaces and that you have: > >> > >>[ifupdown] > >>managed=false > >> > >>in /etc/NetworkManager/NetworkManager.conf > >> > >>If it still doesn't work after a restart[1] please attach both files. > >> > >>[1] not sure why, but in some cases new configurations did not > >>work properly until a full restart > > > >Or you can keep /etc/network/interfaces as you have it now and edit > >/etc/NetworkManager/NetworkManager.conf to have > > > >[main] > >plugins=ifupdown,keyfile > > > >[ifupdown] > >managed=true > > > >(And you have to restart Network Manager.) > > > > > I just tested both ways, to no avail. > > At first, I must say that my system did not have any > > NetworkManager.conf file in /etc/NetworkManager/ > > The only NetworkManager.conf file that could be found on this system was in > > /etc/dbus-1/system.d/ I have that file too but it's not related to what we're trying to do. > I can see the same thing on this other machine still on Lenny, where > I am writing this from: no /etc/NetworkManager/NetworkManager.conf > file either. > > the man page for NetworkManager does not mention any such conf file. > I did not dare modifying /etc/dbus-1/system.d/NetworkManager.conf > > But I tested creating an /etc/NetworkManager/NetworkManager.conf, > with only > > > > [main] > plugins=ifupdown,keyfile > > [ifupdown] > managed=false > ** I wrote managed=true in my message above. (Andrei's message had false.) > As a first trial, I had just written: > > *** > [ifupdown] > managed=false > * > > in it. But maybe I should have modified permissions and owner ? Owner was > root and perms: -rw-r--r-- My permissions and the owner are the same as yours. > Besides, my /etc/network/interfaces file did not contain any definition for > eth0. Still, I tested removing any mention of eth0, to no avail. My /e/n/i file has eth0. That's why we are changing NetworkManager.conf, to get it to cooperate with /e/n/i. If you don't want to define eth0 in /e/n/i, then you can try Andrei's suggestion. (If you do, you might have to set specifics of your wired interface using the NetworkManager applet.) However, your /e/n/i file below defines eth0! > Below is my 'etc/network/interfaces' file : > > * > > # This file describes the network interfaces available on your system > # and how to activate them. For more information, see interfaces(5). > > # The loopback network interface > auto lo > iface lo inet loopback > > # The primary network interface > allow-hotplug eth0 > iface eth0 inet dhcp > > * That last line defines eth0. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120304165317.ga...@cs.utexas.edu
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Selim T. Erdogan wrote: Andrei POPESCU, 3.03.2012: On Sb, 03 mar 12, 20:59:30, Bernard wrote: Thanks in advance for your help with the 'NetworkManager' or any other convenient tool that would operate on Squeeze so as to easily turn an Internet connexion ON and OFF Make sure you have NO definition for eth0 in /etc/network/interfaces and that you have: [ifupdown] managed=false in /etc/NetworkManager/NetworkManager.conf If it still doesn't work after a restart[1] please attach both files. [1] not sure why, but in some cases new configurations did not work properly until a full restart Or you can keep /etc/network/interfaces as you have it now and edit /etc/NetworkManager/NetworkManager.conf to have [main] plugins=ifupdown,keyfile [ifupdown] managed=true (And you have to restart Network Manager.) I just tested both ways, to no avail. At first, I must say that my system did not have any NetworkManager.conf file in /etc/NetworkManager/ The only NetworkManager.conf file that could be found on this system was in /etc/dbus-1/system.d/ I can see the same thing on this other machine still on Lenny, where I am writing this from: no /etc/NetworkManager/NetworkManager.conf file either. the man page for NetworkManager does not mention any such conf file. I did not dare modifying /etc/dbus-1/system.d/NetworkManager.conf But I tested creating an /etc/NetworkManager/NetworkManager.conf, with only [main] plugins=ifupdown,keyfile [ifupdown] managed=false ** As a first trial, I had just written: *** [ifupdown] managed=false * in it. But maybe I should have modified permissions and owner ? Owner was root and perms: -rw-r--r-- Besides, my /etc/network/interfaces file did not contain any definition for eth0. Still, I tested removing any mention of eth0, to no avail. Below is my 'etc/network/interfaces' file : * # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet dhcp * -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f53718f.2020...@teaser.fr
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
On Sat, Mar 3, 2012 at 4:25 PM, Andrei POPESCU wrote: > On Sb, 03 mar 12, 20:59:30, Bernard wrote: >> >> Thanks in advance for your help with the 'NetworkManager' or any >> other convenient tool that would operate on Squeeze so as to easily >> turn an Internet connexion ON and OFF > > Make sure you have NO definition for eth0 in /etc/network/interfaces and > that you have: > > [ifupdown] > managed=false > > in /etc/NetworkManager/NetworkManager.conf For the record, you can have "managed=true" and only define "lo" in "/etc/network/interfaces" and NM'll manage the interfaces. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAOdo=sy2mjqc8-2mexqnedllfa9ooryr55bcjuegbys7gj9...@mail.gmail.com
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Andrei POPESCU, 3.03.2012: > On Sb, 03 mar 12, 20:59:30, Bernard wrote: > > > > Thanks in advance for your help with the 'NetworkManager' or any > > other convenient tool that would operate on Squeeze so as to easily > > turn an Internet connexion ON and OFF > > Make sure you have NO definition for eth0 in /etc/network/interfaces and > that you have: > > [ifupdown] > managed=false > > in /etc/NetworkManager/NetworkManager.conf > > If it still doesn't work after a restart[1] please attach both files. > > [1] not sure why, but in some cases new configurations did not work > properly until a full restart Or you can keep /etc/network/interfaces as you have it now and edit /etc/NetworkManager/NetworkManager.conf to have [main] plugins=ifupdown,keyfile [ifupdown] managed=true (And you have to restart Network Manager.) -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120303225129.ga13...@cs.utexas.edu
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
On Sb, 03 mar 12, 20:59:30, Bernard wrote: > > Thanks in advance for your help with the 'NetworkManager' or any > other convenient tool that would operate on Squeeze so as to easily > turn an Internet connexion ON and OFF Make sure you have NO definition for eth0 in /etc/network/interfaces and that you have: [ifupdown] managed=false in /etc/NetworkManager/NetworkManager.conf If it still doesn't work after a restart[1] please attach both files. [1] not sure why, but in some cases new configurations did not work properly until a full restart Kind regards, Andrei -- Offtopic discussions among Debian users and developers: http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic signature.asc Description: Digital signature
Re: Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
On Sat 03 Mar 2012 at 20:59:30 +0100, Bernard wrote: > So, everything seems to work fine... except for the 'NetworkManager' ! > On that machine, I only have a wired ethernet connexion ; there is no > wlan chipset for now ; I will install one soon. So, the Internet > connexion happens automatically upon booting. But, as a difference with > my other machines (still running Lenny), the NetworkManager tool on the > Gnome desktop does not operate as an ON and OFF switch as I like to use > it. Probably because the interface is not being managed by NM but by /etc/network/interfaces. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120303203338.GK13677@desktop
Problem with NetworkManager - WAS: Re: Help needed to repair a damaged dual boot Debian System
Andrei POPESCU wrote: On Ma, 28 feb 12, 19:35:00, Bernard wrote: stuff deleted.. /dev/hdc3, but now my only CDROM drive does work on /dev/hda (hda as is, not on hda1 or whatever, which I did find weird though). It does work, This is normal, since CD-ROMs don't have partitions. and also I am now able to boot on a system CD !! So, I could possibly decide to re-install... I could also decide to use my machine "as is" since it now operates almost as it did before, with only one CDROM and no CD writer (but I don't need that on this particular machine) ; I just changed the /boot/grub/menu.lst and replaced "/dev/sda3 by "/dev/hdc3". Security support for lenny has ended. The Release Team is preparing a final point release before archiving it. It's high time you upgrade to squeeze. Part of the squeeze upgrade grub1 will be upgraded to grub2 and configurations referencing hard disks will be migrated to UUID (instead of device names, which are not reliable). Since you are considering a clean re-install anyway may I suggest you first upgrade to squeeze, and then, if you are still not satisfied with the outcome, do a clean *squeeze* install. Kind regards, Andrei Thanks for this advice. I just finished upgrading to Squeeze on this system. What a surprise ! Now on this new system, my HD is again reckognized as sda3 ! Still, one of my CD drives is not working, but I don't really care since I don't need two CD drives on that one PC. So, everything seems to work fine... except for the 'NetworkManager' ! On that machine, I only have a wired ethernet connexion ; there is no wlan chipset for now ; I will install one soon. So, the Internet connexion happens automatically upon booting. But, as a difference with my other machines (still running Lenny), the NetworkManager tool on the Gnome desktop does not operate as an ON and OFF switch as I like to use it. (Indeed, if I leave the computer running and unattended for hours, I hate if the machine remains connected all the time. Under MSWIN, you can't easily switch connexions ON and OFF (or you have to disconnect cables... or use complicate processes...), but I found it very convenient on Linux). But now, this no longer seems to operate ! The NetworkManager icon is still there, but it always shows 'not connected', even if a connexion is up and running. Whether I 'activate' of 'deactivate' the network with this tool doesn't change anything ! So, I installed 'wicd'... Well, 'wicd' IS able to disconnect my network, but then, once disconnected, there is no way I can reactivate the connexion ! Reactivating does not operate ! To reconnect, I did not find any convenient way other than rebooting ! Thanks in advance for your help with the 'NetworkManager' or any other convenient tool that would operate on Squeeze so as to easily turn an Internet connexion ON and OFF -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f527822.6060...@teaser.fr
Re: Help needed to repair a damaged dual boot Debian System
On Ma, 28 feb 12, 19:35:00, Bernard wrote: > > What did I do since my last post ? As Richard suggested, I opened up > the box and checked cables around the cdrom hardware. I noticed > something that I thought weird at first: the CDROM blocks were connected > to a cable where one could read: "HD cable", while the HD was linked to > a cable that read "CDROM cable". I then swaped cables, and tested. The Markings don't matter, but type of cable (IDE 40 wires / 80 wires or SATA) and position on IDE cable (in case of Cable Select jumper settings). > only difference I got was that my Linux system now had to be booted on > /dev/hdc3, while it did boot on /dev/hda3 before. Still, none of my CDs > did operate on Debian, one did operate on MSWIN, none would boot an > autoboot system CD. Checking jumper configuration(s) makes sense, as Keith suggested. > This being done, I decided to unplug one of my two CDROM drives, the one > that did not work on either system. I removed all cables from that > drive. This helped a lot, so it seems. My Debian system still boots on A defective drive or just mis-configured (jumpers) can create havoc. > /dev/hdc3, but now my only CDROM drive does work on /dev/hda (hda as is, > not on hda1 or whatever, which I did find weird though). It does work, This is normal, since CD-ROMs don't have partitions. > and also I am now able to boot on a system CD !! So, I could possibly > decide to re-install... I could also decide to use my machine "as is" > since it now operates almost as it did before, with only one CDROM and > no CD writer (but I don't need that on this particular machine) ; I just > changed the /boot/grub/menu.lst and replaced "/dev/sda3 by "/dev/hdc3". Security support for lenny has ended. The Release Team is preparing a final point release before archiving it. It's high time you upgrade to squeeze. Part of the squeeze upgrade grub1 will be upgraded to grub2 and configurations referencing hard disks will be migrated to UUID (instead of device names, which are not reliable). Since you are considering a clean re-install anyway may I suggest you first upgrade to squeeze, and then, if you are still not satisfied with the outcome, do a clean *squeeze* install. Kind regards, Andrei -- Offtopic discussions among Debian users and developers: http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic signature.asc Description: Digital signature
Re: Help needed to repair a damaged dual boot Debian System
On 28/02/12 16:31, Bernard wrote: What did I do since my last post ? As Richard suggested, I opened up the box and checked cables around the cdrom hardware. I noticed something that I thought weird at first: the CDROM blocks were connected to a cable where one could read: "HD cable", while the HD was linked to a cable that read "CDROM cable". I then swaped cables, and tested. The only difference I got was that my Linux system now had to be booted on /dev/hdc3, while it did boot on /dev/hda3 before. Still, none of my CDs did operate on Debian, one did operate on MSWIN, none would boot an autoboot system CD. This being done, I decided to unplug one of my two CDROM drives, the one that did not work on either system. I removed all cables from that drive. This helped a lot, so it seems. My Debian system still boots on /dev/hdc3, but now my only CDROM drive does work on /dev/hda (hda as is, not on hda1 or whatever, which I did find weird though). It does work, and also I am now able to boot on a system CD !! So, I could possibly decide to re-install... I could also decide to use my machine "as is" since it now operates almost as it did before, with only one CDROM and no CD writer (but I don't need that on this particular machine) ; I just changed the /boot/grub/menu.lst and replaced "/dev/sda3 by "/dev/hdc3". However... Something still has to be wrong with this machine. Why does it now reckognises my HD and CD drives as /dev/hdc (hdc1 for MSWIN, hdc2 for swap space and hdc3 for Debian system) and /dev/hda for the now unique CDROM drive while it previously did reckognise these pieces of hardware as SCSI drives (/dev/sda1, sda2, sda3 for the HD... Nothing shows if I type: Thanks in advance for you explanations and advices. Sounds like your cdrom drives may have their jumpers set to primary rather than secondary. Possible explanation for sd becoming hd; you altered the cables between your drives. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f4d21d2.5090...@gmail.com
Re: Help needed to repair a damaged dual boot Debian System
Christofer C. Bell wrote: On Fri, Feb 24, 2012 at 6:08 PM, Bernard wrote: Bob Proulx wrote: Hopefully you would be able to boot a rescue cdrom. The debian installer disk has a rescue mode. That can be very useful to repair a system such as yours. If the above grub selection of a newer kernel does not work then I recommend using the debian-install disk as a rescue disk. Well, but I can't get any of my cdrom drives to work ! Your CDROM drive doesn't work from *within the installed operating system* (and this point is important). This doesn't mean that the BIOS can't see it and that the kernel on the rescue CD won't be able to access it. There is still value in burning an installation CD of Lenny, trying to boot it, and seeing what works. Thanks to everyone for your help. So far, I have reached a point where I wonder if I should quit or test furthermore. Why quit ? Because my system now * seems * to be working almost satisfactorily, even though it does so in a different scheme as it did before. Why thinking about testing furthermore ? Because one of my CD/DVD driver is out of order, also because I still see some bizarre messages while booting, which makes me wonder whether I might soon discover some more problems. What did I do since my last post ? As Richard suggested, I opened up the box and checked cables around the cdrom hardware. I noticed something that I thought weird at first: the CDROM blocks were connected to a cable where one could read: "HD cable", while the HD was linked to a cable that read "CDROM cable". I then swaped cables, and tested. The only difference I got was that my Linux system now had to be booted on /dev/hdc3, while it did boot on /dev/hda3 before. Still, none of my CDs did operate on Debian, one did operate on MSWIN, none would boot an autoboot system CD. This being done, I decided to unplug one of my two CDROM drives, the one that did not work on either system. I removed all cables from that drive. This helped a lot, so it seems. My Debian system still boots on /dev/hdc3, but now my only CDROM drive does work on /dev/hda (hda as is, not on hda1 or whatever, which I did find weird though). It does work, and also I am now able to boot on a system CD !! So, I could possibly decide to re-install... I could also decide to use my machine "as is" since it now operates almost as it did before, with only one CDROM and no CD writer (but I don't need that on this particular machine) ; I just changed the /boot/grub/menu.lst and replaced "/dev/sda3 by "/dev/hdc3". However... Something still has to be wrong with this machine. Why does it now reckognises my HD and CD drives as /dev/hdc (hdc1 for MSWIN, hdc2 for swap space and hdc3 for Debian system) and /dev/hda for the now unique CDROM drive while it previously did reckognise these pieces of hardware as SCSI drives (/dev/sda1, sda2, sda3 for the HD... Nothing shows if I type: cat /proc/scsi/scsi while it previously showed my HD. However, If I plug an USB mass storage device (key or HD), then it does automount on /dev/sda1 and shows in /proc/scsi/scsi ! Maybe you could have an eye on my /var/log/syslog or /var/log/messages files ? http://bdebreil.free.fr/logFeb28/syslog http://bdebreil.free.fr/logFeb28/messages On the first file, you could check first on keywords "collisions" (device not available because of BAR 0 collisions...), and "scsi0 sata promise. sata link down..." Thanks in advance for you explanations and advices. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f4d1e54.5060...@teaser.fr
Re: Help needed to repair a damaged dual boot Debian System
Christofer C. Bell wrote: On Fri, Feb 24, 2012 at 6:08 PM, Bernard wrote: Bob Proulx wrote: Hopefully you would be able to boot a rescue cdrom. The debian installer disk has a rescue mode. That can be very useful to repair a system such as yours. If the above grub selection of a newer kernel does not work then I recommend using the debian-install disk as a rescue disk. Well, but I can't get any of my cdrom drives to work ! Your CDROM drive doesn't work from *within the installed operating system* (and this point is important). This doesn't mean that the BIOS can't see it and that the kernel on the rescue CD won't be able to access it. There is still value in burning an installation CD of Lenny, trying to boot it, and seeing what works. Thanks to everyone for your help. So far, I have reached a point where I wonder if I should quit or test furthermore. Why quit ? Because my system now * seems * to be working almost satisfactorily, even though it does so in a different scheme as it did before. Why thinking about testing furthermore ? Because one of my CD/DVD driver is out of order, also because I still see some bizarre messages while booting, which makes me wonder whether I might soon discover some more problems. What did I do since my last post ? As Richard suggested, I opened up the box and checked cables around the cdrom hardware. I noticed something that I thought weird at first: the CDROM blocks were connected to a cable where one could read: "HD cable", while the HD was linked to a cable that read "CDROM cable". I then swaped cables, and tested. The only difference I got was that my Linux system now had to be booted on /dev/hdc3, while it did boot on /dev/hda3 before. Still, none of my CDs did operate on Debian, one did operate on MSWIN, none would boot an autoboot system CD. This being done, I decided to unplug one of my two CDROM drives, the one that did not work on either system. I removed all cables from that drive. This helped a lot, so it seems. My Debian system still boots on /dev/hdc3, but now my only CDROM drive does work on /dev/hda (hda as is, not on hda1 or whatever, which I did find weird though). It does work, and also I am now able to boot on a system CD !! So, I could possibly decide to re-install... I could also decide to use my machine "as is" since it now operates almost as it did before, with only one CDROM and no CD writer (but I don't need that on this particular machine) ; I just changed the /boot/grub/menu.lst and replaced "/dev/sda3 by "/dev/hdc3". However... Something still has to be wrong with this machine. Why does it now reckognises my HD and CD drives as /dev/hdc (hdc1 for MSWIN, hdc2 for swap space and hdc3 for Debian system) and /dev/hda for the now unique CDROM drive while it previously did reckognise these pieces of hardware as SCSI drives (/dev/sda1, sda2, sda3 for the HD... Nothing shows if I type: cat /proc/scsi/scsi while it previously showed my HD. However, If I plug an USB mass storage device (key or HD), then it does automount on /dev/sda1 and shows in /proc/scsi/scsi ! Maybe you could have an eye on my /var/log/syslog or /var/log/messages files ? http://bdebreil.free.fr/logFeb28/syslog http://bdebreil.free.fr/logFeb28/messages On the first file, you could check first on keywords "collisions" (device not available because of BAR 0 collisions...), and "scsi0 sata promise. sata link down..." Thanks in advance for you explanations and advices. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f4d0174.6040...@teaser.fr
Re: Help needed to repair a damaged dual boot Debian System
On Fri, Feb 24, 2012 at 6:08 PM, Bernard wrote: > Bob Proulx wrote: >> >> Hopefully you would be able to boot a rescue cdrom. The debian >> installer disk has a rescue mode. That can be very useful to repair a >> system such as yours. If the above grub selection of a newer kernel >> does not work then I recommend using the debian-install disk as a >> rescue disk. >> > > Well, but I can't get any of my cdrom drives to work ! > Your CDROM drive doesn't work from *within the installed operating system* (and this point is important). This doesn't mean that the BIOS can't see it and that the kernel on the rescue CD won't be able to access it. There is still value in burning an installation CD of Lenny, trying to boot it, and seeing what works. -- Chris -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAOEVnYu1Sfd1snmZQr_3joiaF9i_0W288Zfqtyv=f=woem4...@mail.gmail.com
Re: Help needed to repair a damaged dual boot Debian System
Bob Proulx wrote: Bernard wrote: was fitted with MSWIN XP ; I repartitioned the HD to have a dual boot (GRUB) with Debian Lenny and the original OS ; it worked very well in such config for more than 2 years or so ; I had a mysql server on it, with mysql databases. Here you say Lenny. But then... just waited indefinitely without doing nothing. Error messages appeared after 4 or 5 long minutes of idling, they mostly said that /dev/sda3 did not (no longer) exist, and that, perhaps, I should try /dev/hda3. I did And /dev/sda would be right for Lenny. In Lenny's 2.6.26 kernel all device names work under the scsi naming. But then you say /dev/hda and the hda names are the old kernel names. This leads me to believe that you are booting an older kernel. The previous Etch kernel used hd names. I think that is part of your problem. I must state that this Desktop was never fitted with any Linux system else than Lenny. When I mentioned the 'original OS', I meant MSWIN XP. The Debian partition was on /dev/sda3 from start. This is not until I had a graphic card collapse and a repair in a shop, that I found that my Linux system now had to be booted on /dev/hda3 The first thing I would check is to see which kernels are available and which one you are booting. You can do this from the grub command line. I only have one kernel to boot, which is 2.6.26 Other choices in my GRUB menu are to boot the same kernel on a "Single User Mode", or to boot MSWIN XP... no more. I usually use TAB for filename completion expansion and use it like an ls command. (The new version of grub includes a real ls command but older ones did not.) Find out if you are booting the right kernel. If you find that you are not booting the right kernel then modify the grub boot on the fly as you have already been doing successfully to specify both the kernel and the associated initrd for it. If that works it should get you to a good booting system. But then you will need to report that problem and fix up any other modifications setting things back to hd* that you may have made such as /etc/fstab. Everything should use /dev/sd* names now. By the way, below is a copy/paste of the /eetc/fstab on that computer: # /etc/fstab: static file system information. # # proc/proc procdefaults0 0 /dev/sda3 / ext3errors=remount-ro 0 1 /dev/sda2 noneswapsw 0 0 /dev/hdc/media/cdrom0 udf,iso9660 user,noauto 0 0 /dev/hda/media/cdrom1 udf,iso9660 user,noauto 0 0 discovered so far. For once, the CDROM drivers did not operate at all on Debian environment ; under MSWIN one of the two drivers did operate ; If my guess is right about the kernel then that would explain this part not working too. none of the two drives would boot a bootable CD (whatever settings in the BIOS SETUP), so there is no way that I could possibly launch a SystemRescueCD bootdisk. Hopefully you would be able to boot a rescue cdrom. The debian installer disk has a rescue mode. That can be very useful to repair a system such as yours. If the above grub selection of a newer kernel does not work then I recommend using the debian-install disk as a rescue disk. Well, but I can't get any of my cdrom drives to work ! Thanks for your message Bernard -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f482694.6010...@teaser.fr
Re: Help needed to repair a damaged dual boot Debian System
John L. Cunningham wrote: > Bob Proulx wrote: > > And /dev/sda would be right for Lenny. In Lenny's 2.6.26 kernel all > > device names work under the scsi naming. > > > > But then you say /dev/hda and the hda names are the old kernel names. > > This leads me to believe that you are booting an older kernel. The > > previous Etch kernel used hd names. I think that is part of your > > problem. > > That's not how it works for me. I'm running Lenny with the 2.6.26 kernel > and it uses the old names. In fact, the reason why I haven't upgraded > yet is because the new names will foobar my backup scripts. Did I get that wrong? I did. Sorry. Thanks for the correction. I even looked on a Lenny machine before posting that and it was using 2.6.26 with /dev/sd* naming. But looking on a different Lenny machine I do see /dev/hd* naming. (Why is one of my Lenny machines different from the rest? Because it is a T43 and has a IDE-SATA bridge chip and presents the IDE hard drive as a SATA drive. Sorry for the confusion.) In any case I think using a debian-installer boot cdrom disk as a rescue image would be best. Then re-install grub again. If the system can be booted then I would use grub to repair the mbr. Bob signature.asc Description: Digital signature
Re: Help needed to repair a damaged dual boot Debian System
On Fri, Feb 24, 2012 at 12:02:14PM -0700, Bob Proulx wrote: > Bernard wrote: > > Here you say Lenny. But then... > > > just waited indefinitely without doing nothing. Error messages appeared > > after 4 or 5 long minutes of idling, they mostly said that /dev/sda3 did > > not (no longer) exist, and that, perhaps, I should try /dev/hda3. I did > > And /dev/sda would be right for Lenny. In Lenny's 2.6.26 kernel all > device names work under the scsi naming. > > But then you say /dev/hda and the hda names are the old kernel names. > This leads me to believe that you are booting an older kernel. The > previous Etch kernel used hd names. I think that is part of your > problem. That's not how it works for me. I'm running Lenny with the 2.6.26 kernel and it uses the old names. In fact, the reason why I haven't upgraded yet is because the new names will foobar my backup scripts. John -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120224193427.GA6478@infotech
Re: Help needed to repair a damaged dual boot Debian System
On Fri, 24 Feb 2012 18:35:54 +0100 Bernard wrote: > It is a 3-4 yrs old machine that belongs to my Genealogical > Association. (Intel Pentium 4 - 3.2 GHz - RAM 2048 MB). At the time > of purchase, it was fitted with MSWIN XP ; I repartitioned the HD to > have a dual boot (GRUB) with Debian Lenny and the original OS ; it > worked very well in such config for more than 2 years or so ; I had a > mysql server on it, with mysql databases. > > Two weeks ago, while trying to boot the system, the monitor screen had > remained black. Same thing when trying another monitor. So, we took > the machine to a repair shop, where our Graphic card was being > replaced ; it costed close to 200 bucks. Once tested, the machine > booted OK on the MSWIN partition, but it no longer booted on Debian > Linux: the system just waited indefinitely without doing nothing. > Error messages appeared after 4 or 5 long minutes of idling, they > mostly said that /dev/sda3 did not (no longer) exist, and that, > perhaps, I should try /dev/hda3. I did so, editing the GRUB boot > command. Debian Linux then booted all-right, after about 3 minutes of > idling. Once booted, the Linux system seemed to behave normally, as > far as I have been able to investigate: Mysqld, OpenOffice, Firefox > etc... seemed to behave as they should. A few things were abnormal > though, may be more many things than those I have discovered so far. > For once, the CDROM drivers did not operate at all on Debian > environment ; under MSWIN one of the two drivers did operate ; none > of the two drives would boot a bootable CD (whatever settings in the > BIOS SETUP), so there is no way that I could possibly launch a > SystemRescueCD bootdisk. > > I could try re-partitioning and re-installing, but I am not sure that > it is the best thing to do, since it could end up that some piece of > hardware is disfunctioning or simply disconnected. I would hate having > to return to the repair shop, since I doubt if the tech man is > trustworthy. > I bet your MS repair guy let windows repair itself: So your boot loader has been overwritten. As for the CDROMs, open it up and check all the cables are properly seated on the CDROMs, as your description sounds mechanical. There's nothing in there than can bite, turn the power off first . Your first priority is to get the CDROMs functional then you can sort the system out as suggested by others. -- Best wishes / 73 Richard Bown e-mail: rich...@g8jvm.com or richard.b...@blueyonder.co.uk nil carborundum a illegitemis ## Ham Call G8JVM . OS Fedora FC16 x86_64 on a Dell Insiron N5030 laptop Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W ( degs mins ) QRV HF + VHF Microwave 23 cms:140W,13 cms:100W,6 cms:10W & 3 cms:5W ## -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120224191707.05c2c...@lap.g8jvm.com
Help needed to repair a damaged dual boot Debian System
It is a 3-4 yrs old machine that belongs to my Genealogical Association. (Intel Pentium 4 - 3.2 GHz - RAM 2048 MB). At the time of purchase, it was fitted with MSWIN XP ; I repartitioned the HD to have a dual boot (GRUB) with Debian Lenny and the original OS ; it worked very well in such config for more than 2 years or so ; I had a mysql server on it, with mysql databases. Two weeks ago, while trying to boot the system, the monitor screen had remained black. Same thing when trying another monitor. So, we took the machine to a repair shop, where our Graphic card was being replaced ; it costed close to 200 bucks. Once tested, the machine booted OK on the MSWIN partition, but it no longer booted on Debian Linux: the system just waited indefinitely without doing nothing. Error messages appeared after 4 or 5 long minutes of idling, they mostly said that /dev/sda3 did not (no longer) exist, and that, perhaps, I should try /dev/hda3. I did so, editing the GRUB boot command. Debian Linux then booted all-right, after about 3 minutes of idling. Once booted, the Linux system seemed to behave normally, as far as I have been able to investigate: Mysqld, OpenOffice, Firefox etc... seemed to behave as they should. A few things were abnormal though, may be more many things than those I have discovered so far. For once, the CDROM drivers did not operate at all on Debian environment ; under MSWIN one of the two drivers did operate ; none of the two drives would boot a bootable CD (whatever settings in the BIOS SETUP), so there is no way that I could possibly launch a SystemRescueCD bootdisk. I could try re-partitioning and re-installing, but I am not sure that it is the best thing to do, since it could end up that some piece of hardware is disfunctioning or simply disconnected. I would hate having to return to the repair shop, since I doubt if the tech man is trustworthy. What I expect, is that someone could help me understand, from my log files, which piece of hardware is possibly disfunctioning or what driver has possibly been corrupt because of the graphic card collapse. For instance, my /var/log/syslog shows some lines that say: hdc LITE-ON ATAPI CD/DVD hdc : lost interrupt ... scsi0 : sata-promise scsi1: sata-promise scsi2: sata-promise ... ata1 SATA max.. irq 23 ata2 " " irq 23 ata3 " " irq 23 . ata1: SATA link down ata2: " " ata3 " " .. .. then, when you get to hda : pci :00:06.0 device not available because of BAR 0 [fccf000 collisions] I could send my log files etc... for your advices. Thanks in advance for your help Bernard -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f47ad41.2010...@teaser.fr
Re: Help needed to repair a damaged dual boot Debian System
Bernard wrote: > was fitted with MSWIN XP ; I repartitioned the HD to have a dual boot > (GRUB) with Debian Lenny and the original OS ; it worked very well in > such config for more than 2 years or so ; I had a mysql server on it, > with mysql databases. Here you say Lenny. But then... > just waited indefinitely without doing nothing. Error messages appeared > after 4 or 5 long minutes of idling, they mostly said that /dev/sda3 did > not (no longer) exist, and that, perhaps, I should try /dev/hda3. I did And /dev/sda would be right for Lenny. In Lenny's 2.6.26 kernel all device names work under the scsi naming. But then you say /dev/hda and the hda names are the old kernel names. This leads me to believe that you are booting an older kernel. The previous Etch kernel used hd names. I think that is part of your problem. The first thing I would check is to see which kernels are available and which one you are booting. You can do this from the grub command line. I usually use TAB for filename completion expansion and use it like an ls command. (The new version of grub includes a real ls command but older ones did not.) Find out if you are booting the right kernel. If you find that you are not booting the right kernel then modify the grub boot on the fly as you have already been doing successfully to specify both the kernel and the associated initrd for it. If that works it should get you to a good booting system. But then you will need to report that problem and fix up any other modifications setting things back to hd* that you may have made such as /etc/fstab. Everything should use /dev/sd* names now. > discovered so far. For once, the CDROM drivers did not operate at all on > Debian environment ; under MSWIN one of the two drivers did operate ; If my guess is right about the kernel then that would explain this part not working too. > none of the two drives would boot a bootable CD (whatever settings in > the BIOS SETUP), so there is no way that I could possibly launch a > SystemRescueCD bootdisk. Hopefully you would be able to boot a rescue cdrom. The debian installer disk has a rescue mode. That can be very useful to repair a system such as yours. If the above grub selection of a newer kernel does not work then I recommend using the debian-install disk as a rescue disk. Bob signature.asc Description: Digital signature
Re: Help needed to repair a damaged dual boot Debian System
On 24/02/12 17:35, Bernard wrote: It is a 3-4 yrs old machine that belongs to my Genealogical Association. (Intel Pentium 4 - 3.2 GHz - RAM 2048 MB). At the time of purchase, it was fitted with MSWIN XP ; I repartitioned the HD to have a dual boot (GRUB) with Debian Lenny and the original OS ; it worked very well in such config for more than 2 years or so ; I had a mysql server on it, with mysql databases. Two weeks ago, while trying to boot the system, the monitor screen had remained black. Same thing when trying another monitor. So, we took the machine to a repair shop, where our Graphic card was being replaced ; First thing to do is make sure all your files are backed up, (databases etc). Did you install the Debian to its own partition, with a separate partition for your data, (i.e. a separate /home)? If you have a separate /home, you can re-install quite safely, after you have your backup data. As long as you have your backup data; I would repartition so you had a /, /home, & swap partition, in addition to your Windows partition; then re-install & add your data. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f47db9b.9040...@gmail.com
Help needed to repair a damaged dual boot Debian System
It is a 3-4 yrs old machine that belongs to my Genealogical Association. (Intel Pentium 4 - 3.2 GHz - RAM 2048 MB). At the time of purchase, it was fitted with MSWIN XP ; I repartitioned the HD to have a dual boot (GRUB) with Debian Lenny and the original OS ; it worked very well in such config for more than 2 years or so ; I had a mysql server on it, with mysql databases. Two weeks ago, while trying to boot the system, the monitor screen had remained black. Same thing when trying another monitor. So, we took the machine to a repair shop, where our Graphic card was being replaced ; it costed close to 200 bucks. Once tested, the machine booted OK on the MSWIN partition, but it no longer booted on Debian Linux: the system just waited indefinitely without doing nothing. Error messages appeared after 4 or 5 long minutes of idling, they mostly said that /dev/sda3 did not (no longer) exist, and that, perhaps, I should try /dev/hda3. I did so, editing the GRUB boot command. Debian Linux then booted all-right, after about 3 minutes of idling. Once booted, the Linux system seemed to behave normally, as far as I have been able to investigate: Mysqld, OpenOffice, Firefox etc... seemed to behave as they should. A few things were abnormal though, may be more many things than those I have discovered so far. For once, the CDROM drivers did not operate at all on Debian environment ; under MSWIN one of the two drivers did operate ; none of the two drives would boot a bootable CD (whatever settings in the BIOS SETUP), so there is no way that I could possibly launch a SystemRescueCD bootdisk. I could try re-partitioning and re-installing, but I am not sure that it is the best thing to do, since it could end up that some piece of hardware is disfunctioning or simply disconnected. I would hate having to return to the repair shop, since I doubt if the tech man is trustworthy. What I expect, is that someone could help me understand, from my log files, which piece of hardware is possibly disfunctioning or what driver has possibly been corrupt because of the graphic card collapse. For instance, my /var/log/syslog shows some lines that say: hdc LITE-ON ATAPI CD/DVD hdc : lost interrupt ... scsi0 : sata-promise scsi1: sata-promise scsi2: sata-promise ... ata1 SATA max.. irq 23 ata2 " " irq 23 ata3 " " irq 23 . ata1: SATA link down ata2: " " ata3 " " .. .. then, when you get to hda : pci :00:06.0 device not available because of BAR 0 [fccf000 collisions] I could send my log files etc... for your advices. Thanks in advance for your help Bernard -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f47ca7a.8070...@teaser.fr
