Hello!
I am running Debian 4.0 64bit on an Opteron.
My disks are two RAID5 on a IPC VOrtex controller (gdth).
My / is becoming a bit to small (Mostly stuff in /lib), so i extended it
with
lvextend -L +200M /dev/Debian/root
Now, to resize the fs i intend to boot from a knoppix-live CD and then
>Just a guess, but according to this old thread, it sounds like you're
running out >of memory?
>http://lists.debian.org/debian-user/2005/02/msg01603.html
Doesn't quite seem logical:
haven:/etc/exim4# free
total used free sharedbuffers cached
Mem: 1029136
I am trying to set up courier-imap and courier-imap-ssl (4.1.1-3) in
Etch, but I constantly run into vthe below when starting courier-imap
haven:/var/mail/Eriks Maskiner/erik/new# /etc/init.d/courier-imap start
Starting Courier IMAP server:/etc/init.d/courier-imap: xrealloc:
../bash/subst.c:512: c
I am trying to use mod_auth_mysql for Apache2, and even tho it is linked
from mods-enabled, I get
"Invalid command 'AuthMySQLHost', perhaps mis-spelled or defined by a
module not included in the server configuration" any hints?
(yes, I restarted Apache2)
Versions:
apache2 2.0.55-4.1
libapache
>Further to my previous, it appears that character set for the server is the
>issue. Since you've told the and & new servers to use different character
sets, >that might be part of the problem.
Killing off /etc/apache2/conf.d/charset fixed the issue. Thanks for the help
anyhow!
smime.p7s
Descri
>Further to my previous, it appears that character set for the server is the
>issue. Since you've told the and & new servers to use different character
sets, >that might be part of the problem.
>
Hello!
First, I'd like to say that I have not seen a previous post from you in the
thread. Could you
>You may try using a more standards-compliant charset, for example
ISO-8859-4, for >Northern Europe. Or try UTF-8, it should contain all your
special characters.
>This is the best way if there are a few pages only.
Well, yes. But I can't really change everyones pages, because ther prolly
is 50-60
Hello!
I have been running my old web server (with apache 1.3.x) for years, but now
it is time to build a new box.
But, I ran into a problem:
I tried moving my wifes page over, and I get encoding problems. Below is a
copy of the header.
Min Sida
But, when I serve this page with my old web s
If i want to migrate a system from a "plain ide" system (on board IDE, and
root-fs being /dev/hda2) to a new system where the root-fs would be on a
ICP-Vortex RAID-controller, managed by the gdth module. How would I do that
in an easy fashion?
I thought I would do the following:
1. amend gdth to
I am trying to compile ivtv_xdriver_0.10.6 on etch (trying to get support
for xv), but I get the below.
mythbox:~/Download/IVTV/ivtv_xdriver_0.10.6/ivtvdrv/xc/programs/Xserver/hw/x
free86/drivers/ivtv# xmkmf /usr/src/xorg-server-1.0.2/
mv -f Makefile Makefile.bak
imake -I/usr/src/xorg-server-1.0
>The -dev packages generally only have the static libraries and headers, not
the sources. If you need sources, >then just run `apt-get source
xserver-xorg` and it will download and unpack the correct source package in
the >current directory.
Thanks! Now to see if it is all I need :)
smime.p7s
D
>http://www.google.ca/search?q=partition+magic
Supports FAT, FAT32, NTFS, Ext2, and Ext3 file systems.. and the user
wanted:
>XFS and merging them into one >partion,
Sooo, your point was?
smime.p7s
Description: S/MIME cryptographic signature
Hello!
I want to compile the IVTV-drivers for Xfree, but I can't figure out which
packages I require to obtain the source for Xfree?
I have installed xserver-xorg-dev 1:1.0.2-9 but I do not have any source in
/usr/src
I am running Etch
smime.p7s
Description: S/MIME cryptographic signature
>I have some NTFS partions in my computer and I want now to format it into
XFS and merging them into one >partion,
>are there any tools is able to do this? thx a lot!
Without wiping your existing data? No.
smime.p7s
Description: S/MIME cryptographic signature
>hope this helps
>Kev
H, for some reason, it seem to only accept users with a UID >1000. I
changed the UID of mythtv from 108 -> 1003, and then all was well.
smime.p7s
Description: S/MIME cryptographic signature
>Hi Jan,
>I was looking into setting up a kiosk mode with gdm/gnome and I read
>about gdm having 2 options (not sure they are in the gdm in
>etch/testing) in Sid. It has autologin and timed autologin. You edit the
>/etc/gdm/gdm.conf.
>
>this autologsin the kiosk user
>-
>[daemon]
>Autom
>look in /etc/kde3/kdm/kderc (I think that's the file) you can add
>users in there
Will certainly do when I get home, thanks!
But, would GNOME use KDE3 config files?
Hello!
I am grying to set up autologin in GNOME 2.1.4 in my Etch-box.
But, the only user in the list I can chose from is the user I created
during initial install of the system. None of the users created after
install (more specifically, I would like to auto-logon the user mythtv)
is displayed i
>Are you losetup'ing and *then* running dd to create the file? From
reading
>your post, I get that impression. If you are, you should create the
file
>with dd, and *then* losetup it.
I am following:
http://deb.riseup.net/storage/encryption/loop-aes/
Here is a longer post I did on linux-crypt
(Not sure if this is a loop-aes problem or a "genral platform problem)
Hello!
I used the instructions on
http://deb.riseup.net/storage/encryption/loop-aes/ to create a encrypted
area. But when I get to the 'dd' stage (using /dev/loop1 and /dev/md2),
weird things starts to happen. (problem does no
>Perhaps some more information would help. The Debian version
>you are running and any relevant changes. "fdisk -l".
cookiemonster:~# cat /etc/debian_version
3.1
cookiemonster:~# uname -a
Linux cookiemonster 2.6.8-2-686 #1 Tue Aug 16 13:22:48 UTC 2005 i686
GNU/Linux
cookiemonster:~# fdisk -l
>because you forgot the missing drive
Lets change that theory. Lets say I want to growthe array as I move data
over. And I need to grow it in several stages. Then I can not use
missing, since I then I effect have >1 missing drives in a RAID-5.
So, is there tools to actually _add_ drives to an exi
>No RAIDs defined in /etc/mdadm/mdadm.conf. However, mdrun
>searches for all RAIDs (IIRC by partition type = 0xfd) and
>starts them.
>
>I wouldn't recommend doing this blindly but given what you've
>posted I think this will fix your config for you:
>
>mdadm --examine --scan >>/etc/mdadm/mdadm.conf
>No RAIDs defined in /etc/mdadm/mdadm.conf. However, mdrun
>searches for all RAIDs (IIRC by partition type = 0xfd) and
>starts them.
Well, I haved DEVICE partitions in there. Should not that do a scan?
>Are /dev/hdd1 and /dev/hdc1 both partitions of type (hex)fd?
>
>--Mike Bird
[Followup to previous mail]
I looked in /etc/init.d/mdadm-raid (enclosed below)
And saw that the script choses between mdadm and mdrun.
Funny thing is that if I execute "mdrun" it starts up my arrays.
Any clue to why
>Are /dev/hdd1 and /dev/hdc1 both partitions of type (hex)fd?
No, they were 83.
I have now chaged it, but still
cookiemonster:~# /etc/init.d/mdadm-raid start
Starting raid devices: mdadm: No arrays found in config file
done.
cookiemonster:~# cat /etc/mdadm/mdadm.conf
DEVICE partitions
cookiemons
I created a two disk array, created a file system. Mounted the file
system. Tested the file system, and all was well.
I then rebooted the system, and the array does not come back
automatically.
I have the following config.
cookiemonster:~# uname -a
Linux cookiemonster 2.6.8-2-686 #1 Tue Aug 16 1
>Raid5 on n-1 disks is more like raid0
Are you sure there? It seem to have come up as one 200GB device?
cookiemonster:~# cat /proc/mdstat
Personalities : [raid5]
md0 : active raid5 hdd1[1] hdc1[0]
195358336 blocks level 5, 64k chunk, algorithm 2 [2/2] [UU]
unused devices:
cookiemonste
>>We use mdadm to administer our software raid. For (certain) hardware
>>raid there other tools available. Just "apt-cache search raid"...
>
>Wait... does mdadm have that built in?
Let's rephrase that.
Lets say I just did
cookiemonster:/mnt/hde1# mdadm --create /dev/md0 --level=5
--raid-device
>We use mdadm to administer our software raid. For (certain) hardware
>raid there other tools available. Just "apt-cache search raid"...
Wait... does mdadm have that built in?
Is "RAID Reconfiguration Tool" ( http://unthought.net/raidreconf/ )
still the only way to reconfigure an array? I just thought since the
page is from 2001 maybe there is a new way? I would like to add another
200GB disk and go from RAID-1 -> RAID-5
>Erm, your local network address wouldn't happen to be a 192.168.0.x
>address, would it?
NOO i copied the
_wrong_ conf file from my backup! Thanks.. Talk about being blind.
But I still get this:
Feb 26 03:30:59 cookiemonster net.agent[780
Title: OpenVPN makes system loose all network conectivity?
I have a wierd phenomenon. I have a new box, that can't seem to run openvpn.
Whenever i start Openvpn, the system looses all network connectivity. I have never seen anything like this before, and i am stumped. This is the first box i
>However, the resultant home.tar file appears to be OK (I can extract it
>without evident errors).
>
>Google shows a number of references to this problem but none seems
>particularly relevant.
>
>I don't much like relying on files that have been generated with this
>error. Any ideas? Can I ignore
Is it possible to spin-down drives that is connected to a 3Ware
controller? I am pondering adding an old 7000-series board to one of my
boxes, but it would be nice to retain power save (Or rather Noise-save).
I am about to set up a new VMWare host since VMWare Server is now a free
application.
I have understood that in order to complie the various modules, you "must"
use the exact same GCC for the kernel as for the module.
Now, if I want to grab linux-image-2.6.15-686-smp and make sure I can build
th
> Now I am wondering if it is hardware-related; I am
> using a motherboard with integrated ethernet: SiS900
> PCI Fast Ethernet (rev 83)
Yeap, I have the same situation on a system with that chip. What happens
is that eth0 and eth1 simply "swaps places" on some boots (integrated
SiS900 and a Intel
> Has anybody experienced a similar problem or have any idea what the
> problem could be?
Had that problem in a system using DEC/Tulip cards, never worked out
what the cause was, I just replaced the cards with Intel Etherexpress
and haven't seen the problem since ;)
> what happens if you use "-M /usr/share/snmp/mibs" or "-m
> /usr/share/snmp/mibs/printer.txt" with snmpwalk? If this works, there
is a
> configuration problem and the mibdir is not getting picked up. If it
> doesn't, it's likely you have the wrong mib, or netsnmp doesn't
understand
> the mib.
Ok
> what happens if you use "-M /usr/share/snmp/mibs" or "-m
> /usr/share/snmp/mibs/printer.txt" with snmpwalk? If this works, there
is a
> configuration problem and the mibdir is not getting picked up. If it
> doesn't, it's likely you have the wrong mib, or netsnmp doesn't
understand
> the mib.
Sam
I have the following components on a debian system
hooch:/usr/share/snmp/mibs# dpkg -l | grep snmp
ii libnet-snmp-pe 4.0.3-1Script SNMP connections
ii libsnmp-base 5.1.1-2NET SNMP (Simple Network Management
Protocol
ii libsnmp-perl 5.1.1-2NET SNMP (Simple Network Man
Is that supported, or is there a hack-around-guide available somewhere?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>Is there an easier way to do this even? Thanks for any help!
dpkg --get-selections > somefile
then
dpkg --set-selections < somfile
apt-get dist-upgrade
should do what you want.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTE
>Does anybody else experience similar log events and if so, is there
>anything you did to stop this from happening?
>Thanks,
>Andre
Replace the drive. The error means that the ECC (Error Correcting Code)
of the Barracuda is kicking in to correct corruption in the drives
on-board cache RAM.
--
>You picked the wrong time to be in testing. This is a documented bug,
>read this for more details:
>http://lists.debian.org/debian-devel-announce/2003/debian-devel-announc
e->200303/msg6.html
Well, I am glad that it wasn't just me then. Thankyou.
--
To UNSUBSCRIBE, email to [EMAIL PROTECT
What does this mean? All the dependencies _are_ installed. But why wont
php still install? (Debian/Testin)
scooter:~# apt-get install php4
Reading Package Lists... Done
Building Dependency Tree... Done
Some packages could not be installed. This may mean that you have
requested an impossible situat
Uhm, I used dselect to set a package on hold. And then I got the below
error, which I definitely do not want to do. How do I tell apt/dselect
to Look, I want to keep the packages I have on here, update your
database to be in sync with real-life please? (Debian / Testing)
Reading Package Lists..
Is there someway to find out what caused a system reboot? I have the
following in my "last"
argus:~# last -20
root pts/0192.168.221.207 Mon Mar 10 08:11 still logged
in
reboot system boot 2.4.18-bf2.4 Sun Mar 9 17:49 (16:06)
root pts/0192.168.221.207
>What is it called when you set up foo.mydomain.com
If you are talking "apache" it is called "VirtualHost" or "VHOST".
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>I'm getting notifications of emails having failed which were allegedly
>sent by me to addresses I never heard of. I presume this is some sort
of
>attack but I can't figure out where it is coming from or what is
>happening. Does anyone else see anything of the kind?
Nah, some worm infected windows
>I have debian distro installed on hda and I would like to install a
Slackware on hdb. Now my problem is, When the slackware installer
>executes it wan't to use both swap partitions. One on hda and one on
hdb. I don't want it to do this. Has anyone worked with Slackware >and
if so did you have th
>Thanks for any suggestions!
Most would opt for NIS / NIS+ for this.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>dpkg --force-depends --purge exim
>Use this only when you KNOW for CERTAIN this is safe. In the case of
exim
>your machine will not fall over and die because it is missing a mail
>server.
Warning heeded.. But that, along with apt-get install exim, did the
trick just nicely, thankyou.
--
To
>dpkg --purge package
>
>If this is failing, give us more info.
argus:/etc/exim# dpkg --purge exim
dpkg: dependency problems prevent removal of exim:
at depends on mail-transport-agent; however:
Package mail-transport-agent is not installed.
Package exim which provides mail-transport-agent is
Tried a "customized" deb of a package (Exim), decided that it wasn't
"all that I wanted" so I want to replace it with a stock Exim.. But I
cant seem to make apt-get and/or dpkg to understand that?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact
> ssh-keygen -N ""
>
> That skips the passphrase question... :)
THANK you. :) I tried single quotes.. but never thought of doubles *bah*
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> I think I just hit the enter key when it prompted for password.
And that makes it _un_attended just how? :)
>
> Note however, that since X runs under ssh-agent on
> Debian, you don't need
> passwordless keys. Just use ssh-add to add the passcode(s) to
> your ssh-agent
> and you can
Doing some unintended installations, and would like to generate SSH2 DSA/RSA
host-keys. But i cant figure out how to set "no password" for ssh-keygen. Help?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>Any ideas on how I
> can reschedule
> this job to the middle of the night so I'm not left staring
> at the screen
> and waiting every morning?
Edit /etc/crontab and change the time for the "daily" batch.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble?
> I'd like to implement RAID 1 or RAID 0+1 in my Debian woody.
> Please recommend the product or chipset which supports Linux well and
> proven stable in productive environment. Thanks.
My vote goes to the Escalade product range.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "u
Not strictly a Debian question, more a general hardware issue, i know.. but i
gotta start somewhere.
I have a BP6 with 2xCeleron 433. And it seems that justabout any heavy disk
activity towards the disks on the HPT366 controller freezez the system up
completely to the point of cold boot. Has an
Installing the .26 apache removed my apache-ssl, no worries since it was a lab
system, but uhm.. how do i proceed if i _want_ apache and apache-ssl on the
same rig?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Installing the .23 apache removed my apache-ssl, no worries since it was a lab
system, but uhm.. how do i proceed if i _want_ apache and apache-ssl on the
same rig?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> Is there any way I can make the script work without having to install
> sendmail? It calls sendmail like this: "/usr/lib/sendmail -i -t"
Exim should be compatible with sendmail in that respect. Just symlink exim to
sendmail. Else i am sure the exim user list will be able to fix-you-up in no
t
> so contractual, however inresponsive, support from a lame-ass
> linux distro
> means more to you than actually securing the system?
Nope. Read my last paragraph. A system provider which can not also offer a
_legally binding_ support contract is simply not allowed on any production /
mission
> you've got to be new around here. there isn't enough salt in
> the world to
> make your hat tasty enough to retract the last sentence
> above. go directly to
> jail. do not pass go. do not, under any circumstances,
> attempt to collect
> anything at all. bye-bye.
Well, there is a valid poi
> How can i start and stop the zope server? In SuSE and FreeBSD there
> are the start and stop script, and in Debian/GNU Linux? I have stopped
> with kill, and can't restart
/etc/init.d/zope restart (or stop, or start or...)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "
> indeed _not_ a good idea.
> It might help to convert RE: into Re: by hand?
That i hadnt noticed actually.. Thanks, ill actually try to do that.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> I believe
> I have correct driver.
> Thanks for any advice.
Start by telling uss what card it is, and what chipset it uses. If it is a PCI
device, you should with almost complete certanty _not_ add anything to
lilo.conf.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubs
> Your mail setup is also rather warped - it doesn't insert
> attributions,
> and it keeps starting new threads which I have to manually
> link together
> in mutt if I want to follow the conversation ...
Well, thats what happens when the Mail admin thinks policy-locking Outlook to
his own prefe
>I've
> seen this on PCs ranging from 386s so maybe one of these
> should work for
> you.
The first PS/2 to have a "internal" BIOS config utility was the 95XL series.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> Does any one know how to access the BIOS setup of an old IBM PS/2?
> It seems that hitting "del" while booting doesn't work...
You spamassassin setup is severly warped by the way.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> Does any one know how to access the BIOS setup of an old IBM PS/2?
> It seems that hitting "del" while booting doesn't work...
Not possible. If you by "older" mean models 8xxx (386 and 486 systems) or
older, there is no BIOS config utility, you have to install their utility
software on a dedi
> I try to configure mailman on my gateway for internet, using
> the exim mail
> server
http://www.exim.org/howto/mailman.html
follow that one, and then you will not need to alter your aliases at all. In
fact, altering aliases will break the config.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTE
> I remember reading about a data recovery team that recovered
> files after
> the data had supposedly been removed with a 'dd' command like above.
www.ibas.no they claim to be able to see under 3 layers of overwrites. And
since they are a public contractor.. wonder what the not-so-public people
> Perhaps you're right, and I'm doing it wrong, But on my box, I've
> configured my user to have full access via sudo. That way,
> when I need
> to do root stuff, I can just "sudo rootstuff.foo" instead of becoming
> root and then forgetting that I'm still root and doing
> something nasty
> t
> USB too?
> I remember having heard that USB modems were basically winmodems
No, almost all USB modems are "real" modems. But they still require drivers.
But there are drivers for quite a few.
http://www.idir.net/~gromitkc/usblist.html
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subj
>Can I
> safely uninstall this service/program without affecting my p.c.?
Yes.
> Also they recomend disabling nfs...I have "nfs-common" and
> nfs-kernel-server"
> installed. Can I safely disable these too?
Yes.
>I dont require
> them for apt-get
> updates etc?
No.
--
To UNSUBSCRIBE, e
> append="idebus=66"
Good luck recovering that drive.
Lets understand one thing here. The IDE controller resides on the PCI bus. The
PCI bus (in affordable hardware) is 33MHz. ATA/66 / ATA/100 / ATA/133 has
_NOTHING_ to do with the bus speed of the system.
--
To UNSUBSCRIBE, email to [EMA
> well... if it dont work... define it ...
It is never read or in any other way used by samba.
> ( i always define it... and dont have any samba problems ..
Neither do i with that config file on a RH system.
> i dont like these "shortened" default smb.conf files...
Works on other systems, henc
> what is the permissions of /home/j2
drwxr-xr-x 18 j2 j2 4096 May 28 10:18 j2
>
> what is it exports... homeserver:/etc/exports
> #
> # ip# and masks
> # some apps wants to see mask instead of just an ip#
> #
> /home/j2 192.168.1.0/255
[EMAIL PROTECTED]:~$ smbstatus
Samba version 2.2.3a-6 for Debian
Service uid gid pid machine
--
j2 j2 j28207 rizzo(192.168.0.239) Thu May 30
02:19:16 2002
Yet any write operation on the homedirectory fa
> now my question is: how do i set this IRQ parameter since i
> know for a
> fact it is 12 on my system
You never pass IRQ to a tulip board (you should never ened to pass IRQ to a PCI
board at all). Something else must be wrong.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of
> i have installed debian 2.2 in my server now i want to
> activete ipchains & ip masqurading on it ...
> how should i do this ?
Personally i would install shorewall.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> What's the value of the TERM environment variable?
> env|grep TERM should tell you, at least under bash.
TERM=vt220
BitchX gives me color tho. but MC, ls and stuff does not.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> None. Most of them use a hybrid software/hardware RAID which depends
> on a (usually Windows-only) driver to make them work.
I get the feeling that the only IDE RAID controller you have heard of is
HighPoint and/or Promise? True they have issues. But there are also _real_
controllers, such
> Thanks for your time!!
Escalades controllers have served me fine. I havent had a chance to play with
the new generation controllers. But the old one has run flalwlessly with
hotplug/hot rebuild and all.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? C
I have been over the archives, but i cant find an answer to this one. I am
running Testing, and when logging on to the system(s) via SSH (Using Secure CRT
3.4/VT100 or VT200 with ANSI color) as the client, i get no colors. if i use ls
--color, or ls --color=auto it works, but still no colors in
88 matches
Mail list logo
