psu problem would panic any time, cabling problem would panic only when something extremely broken happened, like the drive writing data to the wrong place on the computer

Kevin`: I think it’s a kernel related problem

so..
loh
log*

when kp occurs I can only turnoff or reset

which is consistant with kernel panics
and intentional, of course

Kevin`: a broken module?

dunno, LOG IT AND FIND OUT

hio all
hi

Kevin`: done! writing a data dvd with success, so it only happen after a dvd copy
Kevin`: or k3b mismatched configuration?

Mazingaro k3b itself shouldn’t be able to cause this

Kevin`: just tying out the copy now
Kevin`: could a read error throw a kp?

Mazingaro it shouldn’t.

Kevin`: now the copy is started, last time same source cd (windows buruned) kp before power calibration

Am I just blind or has there really been no announcement about the Debian Maintainer GR?

Kevin`: succesfully copied!
Kevin`: I mean

test
~sudo

failed

i guess sudo is (SUperuser DO) better than su, according to talon.It is able to give limited super user privileges to specific users, or can allow you to do silly things like run X apps with root perms, or good in scripts with “username ALL = NOPASSWD: /some/program”, or http://www.aplawrence.com/Basics/sudo.html, or good for ordering sandwiches, or not pseudo

please dont use at beginning special chars “~” – it activate the bot(s)

ok

!test

Test Failed.

good morning
I smiled as soon as I woke up today
I let e2fsck run last night, today my drive is clean of errors
yippy
now I am going to buy another 500gb disk, make my backup, and not worry

good!

hehe
man
I have to go to work in 40 minutes
I’m kinda sleepy, hopefully the boss won’t be there cuz it is sunday

to sell lingere?

no man

shoes?

for example yesterday
I fixed a simple outlook express mail could not be deleted problem and charged 200$
regular job, I manager a pc sales/tech/repair store
I boss people around, its fun

heh

basically the problem was that I had been playing the new oblivion expansion
so I was using XP, and all my media was on the ext3 500gb disk
so I mounted it in windows with ext2ifs driver
I had a few crashes, you know windows right
so then, I try to write to ext3 under windows and it says, can’t write, file or disk corrupted!
so I booted a 64bit ubuntu, (the only cd I had) and ran fsck, took all night a good few passes, but I fixed it

!beer stoned

so man dvs
what have you been upto

I spent last night at this city “block party” last night and it was a pain in the ass

heh
block party eh
Chemical Brother – Block Rockin’ Beats

well, that’s the best way to describe it.

I think I need to see a dentist
need to detail my mouth

with gold caps?

I mean everytime I try to clean the back of my toung with a toothbrush
it makes me wanna throw up
how the hell do dentists do it without having you throw up
gag reflex etc.

why are you trying to clean the back of your tongue? I don’t

cuz it gets dirty after a while
residue builds up and gives you bad breath

that’s what mouthwash is for

its like a offwhite creamy gunk that is all over your toung
dude…
mouthwash?
all the way back there
makes me wanna puke
I rinse my mouth with it, I don’t try to drink it

heh

get that tootbrush out and start scrubbing

I like the blue mind kind
mint
and the orange citrus

which reminds me i need to scrub that to get last nights bar taste out of my mouth

man I don’t trust mouthwash
I still brush my toungue for a good 1-2 minutes
as far back as I can

listerine

then I rinse with mouthwash
listerine

no wonder why you gag!

yes

as you should

no but I’m saying
how to clean way far back

i like plax also

ow
wow*. fixing my ext3 disk has now somehow made windows unbootable
somehow the c was changed from a primary(1) to a logical (5)

and the problem is….?

i get superblock last write time in the future every time i fsck an ext3 fs, even if i just fsck’d it

fuck dude

i call that a bug

I should just buy an xbox 360
and use nothing but linux and call it a day

only reason i put ext3 on my laptop is because i don’t know of any jfs windows reader things

then buy a converter so I can hook up my xbox to a kvm
ext is good

can’t tell

mount time is fast, fs is kinda slow
and its stable
plus it has journaling
drawback, no undelete

then why the superblock write time in future every time

slow
I don’t kow

me neither
it’s across all installs for like a year or more afaict

not here

i will give ext4 a shot though

I have not used it yet
maybe I should stop stoning
I don’t really care

I don’t like using two partitions to hold one set of data

what else am I gonna do at work
I’d rather just stone at home and then go to work

read manpages

6 days a week man
but its ok

i just worked 6 days and today i get to work on my cars

I am paying my dues
no I worked yesterday too
I had an onsite appointment
3 hours
damn, thats like almost 7 days

!pity stoned

hehe

I pity da fool stoned!

dang
aha!

I pity the fool too
but I also may suggest how he may better himself
- cleveland -
~~6-
*

heh i was on heinlein

morning, does anyone have a repository i can go to to get PyQT?

uh oh i still am

~you know

so are you going to be late again ?

I don’t want to go to work today
but I have to

so?

I’m not late on purpose man
wtf

find another if it’s so rough

I was only late twice

since the ‘talk’?

sicne always

so you were late once and your employer thought it necessary to have a chat…

I was late twice, and he had a chat
thats it
and there was a tornado in nj or something I think
so traffic was crazy

something tells me third time might be the charm with this guy then

nah
he needs me
I do a good job
and I bring his company a good name
everyone else who works there has a foriegn accent
so I deal with the custoemrs, plus I’m good at dealing with people

yoda, I dont know about anyone else but I am only on x86 and x86 _64…sorry

yoda?
hahaha

quit

no
irc is beginning to bore me :/

are there problems with s.d.o today?

what is s

security

!sdo
ah

alright man Im out
off to work, starved

bye
don’t be late

Hello everybody, I know this is an old news, but does anybody have more informations about amd and their promise of release free drivers ?

Hello. I’ve got a laptop with an extremely annoyingly sensitive touchpad. ksynaptics and such don’t seem to have an effect, neither when I try to adjust down the sensitivity, or when I try to shut the pad down alltogether. Can anyone help me out?

how do i set debian to pull one specific package from unstable tree
i want to install pyqt version from unstable tree

Go to packages.debian.org and manually download it
At least, that’s what I usually do

Dark_Apostrophe, what about dependencies ?

Hmm… Not sure. Download them too, I guess. Not sure how else to do it.
Then again, I’m not exactly an APT expert.

xset -m? i dunno if it works for touchpads

Hello. How do I rer-run package configuration (for an installed packagee) so that it re-merges the config files (do you know what I mean?)

if you’re running testing, it might turn out ok. If you’re running stable, chances are you’ll need a lot more dependencies

you could check backports.org to see if they have a backport (or attempt to build one yourself ; /msg simple sid backport) Installing a sid package will often want to upgrade your libc which is not what you want.

i.e. the config file was upgraded and asked about re-merging my changes. I sayd No but I have now changed my mind

ok

I’ll give it a spin

reed[]: generally, there should be two copies of the config now: the original and the new one. You should be able to diff them for changes and merge manually.

well dont need libc6 upgrade, pyqt’s required libc6 version is already installed

you mean foo.dpkg-dist?
does that config file merge thing come from the postinst script?

reed[]: part of dpkg iirc. You can check /var/lib/dpkg/info/pkg.postinst if you want to see if it’s called from there though.

my sound is fixed

i.e. is there a command to just force a redo of everything that happened after the pkg was unpacked?
or force apt-get to redownload and reunpack and everything else?

reed[]: /msg dpkg confmiss (maybe)

reed[]: there’s dpkg-reconfigure, but it might not do what you expect it to

if you think bug 437361 is a ssecurity bug feel free to mark it so

i tried dpkg-reconfigure, but since the packages were configured, it doesn’t do anything. or at least it’s not doing the same things.

i’m not sure about this, but adding line APT:efault-Release “stable”; to /etc/apt/apt.conf and then just apt-get install -t unstable pyqt should do that… or at least it worked for me with sarge and unstable rails

hi
http://www.debian.org/mirror/webmirror#list
why isn’t the USA on here?
I know there are mirrors here…..
can someone tell me where the list is?
I can’t connect to the georgia tech mirror

the US mirror doesn’t have a country code

there could still be a link to it

I guess it wouldn’t be a mirror, would it?

or rather, I still need a link to it

that list is the list of www pages mirrors, you want www.debian.org/mirrors

well there are plenty of mirrors

reed[]: if you’re really desperate, you can run the script with same arguments as dpkg did: /var/lib/dpkg/info/foo.postinst configure “previous-package-version”

ah, thank you~
I mean thank you!
another question – apt-get is telling me a list of files that were automatically installed and are no longer needed; how do I select these for removal?

apt-get autoremove

rrrrreally
very cool feature

YA RLY

lol
ORLY

ORLY sounds like an airport in France.

NO WAI

BAJESUS

do you like chezburgerz?

i’m installing debian from floppies
but it can’t detect my partitions
no /dev/hda i mean
some suggestion?

is it an internal hard drive?

yeah, a laptop withouth cdrom too
hard drive is working because i can see the partition with other dos tools and i can change de mbr or format it

Hello all
is march a vaild option in apt-build.conf

what made you decide on floppies vs CDROM?

no cdrom avaiable on my laptop
no usb boot either

oh

trying__2 is it a sata drive?

mmm, some doubts there
how i can verify?

well do you have /dev/sda?

let me see, one minute
it’s not a sata
i used to have a hda

ok

some of the older kernels identified sata drives as hda and would change on kernel upgrade

well i’m installing etch… i guess it would have a new kernel
any other idea?
well another aproach… any idea to install ubuntu with floppies and inet?

hi any one got any experience with openwrt ?

the folks in #openwrt probably do

oh i missed that
ty
!ty

somebody said ty was thank you, or cobb — a baseball player, or the hosting company that creates Beanie Babies(TM)

s/thank you/I’d thank you but am too lazy to type in the required nine characters/

heh, if he did i prolly would have replied with np anyway

!moc

well, moc is the QT meta object compiler

.4.4-8+etch4

re

not sure why it even wants etch3 instead of etch4 (php4) in the first place

dos000, you can pin it in one of the confs but if it’s a specific source you can force aptitude to use that source over others via -t othersource I believe
dos000, other deps may be newer

eps, the source is the same .. they are coming from debian security
eps, what i dont understand even is why it wants to still use etch3

i would like an automated method for fixing lintian’s ‘executable-not-elf-or-script’ warning

dos000, aptitude show package on both of them and see the required deps; there is usually a reason for it

is there one?

do you have a security.debian.org line in your sources.list?

jelly-home, yes i do

i could hack some find kung-fu, but wtf, someone did this before for sure

did an update recently?

jelly-home, yes .. multiple times !
jelly-home, this has been going on for 2-3 days now

lintina
!lintian

rumour has it, lintian is the Debian package checker, and has pages at http://www.debian.org/lintian/

I’m getting seg faults when dpkg is rebuilding the package list – I’ve killed the .bins in /etc/apt/ and blown away /var/*/apt/* data and done a clean

i came her yesterday .. and no one could figure what was going on .

aptitude install foo/6:4.4.4-8+etch4

got the dpkg update from this last week

ah
no pinning I presume?

phogg, foo is what in this case ?

the package name

phogg, he still wants to use etch3 !
jelly-home, no i have no pinning so far

sorry, = now /
s/now/not/

can you pastebin apt-cache policy …?

why does the media source have etch3 and etch4 under the same branch?
that;s just dumb
or is one stable and the other testing?

jelly-home, yes .. one sec
phogg, it still gives Err http://security.debian.org stable/updates/main php4 6:4.4.4-8+etch3

dos000, do you want etch4 from testing?

no, testing’s php4 is at 6:4.4.4-9+lenny1

i was told etch3 was supperceded by etch3 .. in stable
eps, i mean etch4

jelly, ah I see

eps, i have another system and policy returns etch4

ahh boy im hungry

I have no used stable for many years now so I’m behind on how it is done

I dunno why

jelly-home, http://pastebin.ca/654511

well, you don’t have the …etch4 version avaialbe at all.

jelly-home, what do you mean ? etch4 is in the security repositories. the question is why i am being froced to fetch ethc3 which does not exist

on my system it lists etch4 in security

phogg, yes .. that is the normal way. but for me it is stuck at etch3. cant understand why

it’s not in the repositories you have available. check your proxy if you’re using one. delete /var/lib/apt/lists/security.debian.org* and redo the update

the seg fault coming from apt-get/aptitude/synaptic is caused by the following; /usr/lib/dpkg/methods/apt/update: line 42: 3075 Segmentation fault $APTGET $OPTS “$APT_OPT0″ “$APT_OPT1″ update
is any one familiar with it?
can’t find any known bugs

phogg, 1) i have no proxy 2) i pointed fingers at the data center and they said we have no proxies 3) i removed those so many times !

I am searching for beta-testers for Pingus (lemmings-like game), anybody interested? /join #pingus

grumbel, my sister loves that game

what does security.debian.org resolve to?

jelly-home, 128.31.0.36
jelly-home, 128.31.0.36 212.211.132.32 212.211.132.250

force it to any of the other two IPs in /etc/hosts and try updating again?

hi

Hi. I added a line into /etc/apt/sources.list to install a package and when I run apt-get update it says “The following signatures couldn’t be verified because the public key is not available: NO_PUBKEY” How should I proceed?

jelly-home, still says etch3

how can i extand my resolution list in the gnome setting
?

how do i apt-get install a new kernel ?

giz_bmz, I’m not sure but it probably reads from /etc/X11/xorg.conf

garotosopa, if you go to the home page of that source there should be some keys they give you that will get rid of that
else ignore it
unsolo, apt-get install newkernel, restart when done

eps, thanks, there was a apt-key add command line there.

which architecture?

jelly-home, i am on x86

awsome, snatch is on telly

liable, not a bad movie

jelly-home, is there a way i can definitely detect if the guys at the data center are using silent proxies ?

one of my favourites

wget -nd http://security.debian.org/dists/etch/updates/main/binary-i386/Packages.gz

thanks
and listing possible kernels ?

what’s the timestamp on that file?

unsolo, if you are using grub it will automagically list the new kernel for you at boot (if it is newer it will be default)

4″ (I’m in UTC

just a note you may have to recompile extra drivers like nvidia again under the new kernel

jelly-home, it says 2007-08-11 10:44

is there a way to prevent the kernel to get upgraded?

Good morning everyone

and, do you see the correct php4 web services version in it? Search for “Package: php4″

morning Daveman

i was more thinking of listing ones i can install actualy

Hi
i am using RS232 Serial Port

jelly-home, yes its there !

oh sorry unsolo; grep is your friend; aptitude search linux|grep kernel

thanks

you may have some sort of explicit proxy in your apt.conf

with a GPS i can make “cat /dev/lp0″ and i see the output

jelly-home, where is the actual packages i am using ?

i have another device, but when i use “cat /dev/lp0″ tehre is no output

/msg dpkg hold

but if i use windows and Hyperterminal, i can see output

!paste

Please do not paste anything at all to this channel. Instead, use: http://channels.debian.net/paste/ or http://rafb.net/paste/ or http://picpaste.com for pics.

jelly-home, thanks … soooooooooooo much ! there is something in there !

linux-image-amd64 == linux-image-2.6-amd64 ?

http://paste.debian.net/34477

how can i change Baud Rate, Stopt Bit and so
?

this is my xorg.conf

asg, thanks. do you think it could cause any further problems?

ahh iam getting better at fix pc

so how can i increase my resolution ??

eh, you shot yourself in the foot

contral pannel

it will only prevent that package getting upgraded automatically. That includes any security fixes, etc.

jelly-home, its the guys from the dc who did this ! when they auto install etch

i read setserial documentation, but there is no option to change that

asg, ok. thanks again.

how would i get a list of all installed packages in a format that would enable me to feed them to aptitude on another machine?

jelly-home, what do i do whith this file ?

probably just comment out the Proxy related lines and redo the update
// is used for comments (or just remove them)

is drbd the only disk clustering over net available ?
and is it still maintained ?
and it seems that installing a new kernel blew up the system .. ouch

jelly-home, i am your slave forever !

sound cards what is dif between 4.1 and newer

myselections

thank you

so what do i do when apt-get install newkernel fails ..

you’re welcome

lol

using http://paste.debian.net/34477

darn thing blows up because of some mdadm incompatibility

how can i change the resolution to be higher ??

you boot the previous kernel and fix the problem

I want to flash my bios on a Asus mainboard I downloaded update from Asuswebsite…the downloaded file is a .ROM file what I have to do with that? Burn ob a CD as a image?

you need a flash software

there’s flashrom.. never tried it but

Can you recommend some software for windows?

##windows ?

hm yes sorry…

!info flashrom

(Universal flash programming utility), section electronics, is extra. Version: 0.0+r2739-1 (sid), Packaged size: 32 kB, Installed size: 124 kB

doesn’t the motherboard manufacturer provide the flashing utility as well? awdflash if the BIOS is Award’s

hm but flashrom only works on linux? My friend has only windows
mhh jelly-home I will take a look onto the website thx.

you need to get the asus bios flash utility
and perhaps a flopy
maybe

ok thx

I use grml for flashing bios’s (but from the dos that it includes) as its on a usb I just put the files in (I have no floppy)

look on your motherboard cd
my gigabyte mobo cd is a bootable one

hehe he lost it

I’ve recently seen a commercial linux fs that has some redundancy capabilities: http://wiki.digitalbazaar.com/en/Starfish_Distributed_Filesystem

you can download the flash rom from the site, same place you should find utilitiesasus flash thing

Yap I found it thx

/msg dpkg etch

Which bazaar package is the, well, canonical one? (canonical not as in “the company”

http://support.asus.com/technicaldocuments/technicaldocuments.aspx?root=198&SLanguage=en-us
^^
oh ok

hey thx stoned

737 nicks!

how can i share my desctop in two screens ??

jelly-home: interesting..

anyone manage to get apparmour running under debian?
uh, i mean. can i get apparmour running under debian without going insane
buh, i mean. can i get apparmour running under debian without going insane/b

hello,i’ve just installed the mysql-server meta-package on a debian etch “stable” with aptitude, but i can’t find it under the misc “installed” packages view(but the mysql server is effectively installed): the other packages i’ve installed are showed right….do you have any suggestions?
tnx

dpkg -l|grep php mysql web hosting will list *mysql* pkgs you’ve installed

does anyone fancy logging into my box and installing the 2.6.21 kernel with the nvidia driver

sure, i wouldn’t mind another box

:-S
what does that mean? :-S

it means i could run the torrent tracker on your box and not have the RIAA on my back

eeek

so why do you want .21

i run mythtv on that box what uses a Hauppauge Nova-T 500 TV card, and there is a bug in the driver but it has been resolved in the 2.6.21 kernel

ahh
!bpo

rumour has it, backports.org is an unofficial repository of backports for the current stable distribution, prepared by Debian developers. Read http://backports.org/dokuwiki/doku.php?id=instructions ask me about
backports

yes it shows up that effectively the packages for mysql jboss server hosting are installed, but i can’t figure out why aptitude does not show mysql-server installed under the misc installed packages view….i mean i found aptitude functions
and views really helpful,so…i wonder why

use aptitudes search function to see if you can find it that way

ive installed the backport 2.6.21 kernel no problem, im just having difficulty installing the nvidia driver, ive been trying to follow these instructions without success: http://grizach.sc18.info/nvpatch/

i believe it is /searchterm
ahh yeah, forgot about that nvidia fun

installed and not installed, do you know some way to search only in the installed view under aptitude ?

i do not

There is some problem doing upgrade on etch …klogd seems to hang the upgrade
sorry not etch but lenny

anyone here with mysql-server meta-package could control under the mi sc installed packages view in aptitude if is it showed up ?

blah anyone tried to use sierra 595u modem in linux?

is that a 56k dial up modem?

nope its a evdo modem

ok

hmmm, does aptitude search ~imysql help?

im following the instructions on http://www.sierrawireless.com/faq/ShowFAQ.aspx?ID=602 but i keep getting no carrier

how can i share my desctop in two screens ??

vnc?
user01 I havent a clue..

i use one monitor on laptop, another lcd on desktop

no not vnc
i have two screens
i want to make one desktop with them
so i’ll have a one big desktop

you have two physical display devices?

yes

How do I prevent system freezeups while I am in KDE? Sometimes the programs (Opera, keditbookmarks, …) freeze the screen and the computer becomes useless. Suggestions?
How do I prevent system freezeups while I am in KDE? Sometimes the programs (Opera, keditbookmarks, …) freeze the screen and the computer becomes useless. Suggestions?

does ctrl-alt-backspace still work?

run memtest86

I will try it next time, what does ctrl+alt+backspace do?

which graphic card(s)? Are you using free drivers?

to check if my memory is valid?

aye

ati

kills x

Alright, that’s something good to try. Are there any other things I should do? Maybe some special configuration? “Don’t let programs hijack me?” heh’

run sshd and when it freezes see if yo ucan login and kill the process thats the issue

the point here is that aptitude does NOT show this package under its “installed packages” view, even if it is regularly installed!

ah, but I don’t have a machine nearby
wouldn’t it be a bug with the KDE interface if all user interaction is denied?

whaat is the kernel version of debian?
latest version

2.6.22

looking all over the documentation for it
thanks

but does aptitude recognise it as installed..

I am partitioning the drivee manually… should I turn on ANY bootable flags?/

what do you mean, “freezes”? Holds the keyboard/mouse focus? Does the music player (if any) stop playing? Do you get disk activity?

if aptitude says its installed, then i’d be inclined to wonder if you just haven’t missed it in the list
i don’t think you need to

jelly-home: music stops playing too, yes, and I see no disk activity IIRC

the latest stable version of fedora linux for debian is 2.6.18

stable Debian uses 2.6.18

what? You and I were not talking about aptitude

thats what they said in linux

uhuh, mavior i mean

ahh good 2.6.18 is more stable the 2.22 from what I have seen… though I had problems with suse’s Samba on Linux…

you can use that as the search term in the aptitude ncurses interface, btw…

you may experience better interactivity with a _very_ new kernel with CFS patches, or try the (now unmaintained) ck patchset

On another note, where’s gtk-config found?

!MOC

well, moc is the QT meta object compiler

why do you need that ?

well linus’s epic plan is that the distros create stable kernels, and he releases not quite stable kernels, iirc

s/not quite stable/@#$%^&/

that epic plan has costed me 2 months of my life

well, use something otgher than linux then

nothing supports expandable arrays

solaris/zfs?

openNAS seems AWSOME… but no expandable arrays
solaris supports expandable arrays?
how is it with new hardware

dare i ask whats an expandable array

probably not very good

google for zfs. it’s da shit, supposedly.

mdadm /dev/md0 -a /dev/sde

or better, ask in #solaris about it

mdadm /dev/md0 ….

#opensolaris
i use ZFS, you can add to a pool online

oh, there’s a separate channel, right

it means you can add a drive to a RAID 5 array without loosing the data on it

it’s better than raid5!

it is raid 5.

here’s the thing

(I mean, zfs)

and you can use raidz

Solaris doesn’t support all the wild block-device-level stuff that Linux does.

it can be raid 1 or 3 or 55

So that kind of thing ends up implemented in the filesystem

ok, aptitude shows the mysql-server meta-package under the “new packages” view….don’t know why….thank you anyway

mdadm, in linux, today, will allow you to reshape RAID online
go from RAID5 to RAID6

hello.

eh, I don’t care about mixing up the levels as long as it works

damnit
i thought debian kernel swapping was safe
seems i was wrong
system broke.

jelly-home: But that’s what GreenJelly was describing: dynamic ability to reshape arrays.

kernel swapping as in…

ZFS is pretty great, don’t get me wrong, but we already have that on linux

you swapped linux based hosting services for kfreebsd and it broke?

nope, he just wanted to add disks.

(shared storage pools, we don’t have. )

we don’t have the checksummage though

jelly-home: … You can reshape from RAID5 to RAID5
checksummage?

raid5 / 6 still have write penalties

I have 6*500mb drives I want in Raid 5 (this is paart of my problem)…

checksumming of data read back off the filesystem, to make sure you get back what you wrote in the first place

jelly-home: humbug. Welcome to the modern era. We have this great processor, the pentium II!

tomorrow, I run out of space

That’s built-in at the filesystem level, not at the block device level.
XFS does that.
Ext3 might, don’t know.

uhh ohh… how do I back up 2.5 TB of data (sorry, I ment 7)…

XFS does? did not know that

and the reason I mentioned it is because the mentioned being stuffed with Linux’ no stable kernel at all policy

Yup. That’s how I discovered problems with sx8
jelly-home: Yeah, well, that’s just fucking stupid. But fucking stupid is the design philosophy over at kernel.org. Go figure.

so I add a drive to the RAID–5 aarray as a spare… then I convert that spare right into the array… meaning I can add a drive without loosing my data

hence, solaris.

jelly-home: For all of my Solaris partisanry, I use Linux on my home desktop. And I have fewer and fewer Solaris systems at work. Go figure.

yeah well, me too
but someone has to play devil’s advocate!

no offense, but static device names suck

No, you need to use the “–grow” option. Don’t add a spare.

pipeline you first add a spare
then you grow
weird, I know
but thats how you do it…

grrr where is that picture of a girl with a cat on her bubby, representing firefox
I cant google it

You add the spare in order to have someplace to store data while you grow to another device.
The spare plays no role in the finished array.

‘It is not possible to add a disk to a RAID-Z or RAID-Z2 vdev. This feature appears very difficult to implement.’ hmm, there be a bummer

pipeline the spare becomes the extraa drive

‘fraid not.

yeah, got to do it as raid-z2 from the start

It’s just swap space so that you can fuck up safely.

here… this is EXACTLY the steps I use…
and have used successfully

btw do you really use 500mb drives

o reshape, add drives to your raid set in the future, you first add the new drive to the existing raid set. mdadm –add /dev/md0 /dev/sdf1 Next use the –grow switch to reshape it to the new number of drives you want. The command below assumes you had 3 devices in the raid set before you added
sdf1 and now after you added sdf1 to be available to the set, you want the raid-devices to be 4. …
…Next time you add a device, it would be 5 … mdadm –grow /dev/md0 –raid-devices=4

aren’t we nitpicky

sorry cast… I ment 500gb or what they call aa GB… 467

no i upgraded the kernel and it blew.

what kernel should I pick?

ahh, ok. some people really do have arrays of 500mb drives is all..

nah.. I have around 2tb
and am waiting on a DOA drive to come back

a single filesystem?

??

i downloaded lirc 0.8.2 from lirc.org, I ran configure and now Makefile is empty and I cannot make the source – does someone know why or what the reason could be?

is all of your 2TB used for a single fs?

yess
its all one raid array

Hey
ttf-opensymbol seems broken

from what to what?

I keep getting “failed to write to cache” in regards to it

2.6.18-4-686?

Every single time I use apt-get
Anyone seen the problem before?

2.6.8 to 2.6.28-4
err 2.6.18-4

28?

which debian release?

what cpu? use 2.6.18-4, but perhaps -k7 or whatever you run

celeron D
64 bit
on a 965 ICH8 Chipset

jelly-home: 64 bit
amd
etch
but old kernel came from sarge i think
and sarge-amd64 was experimental

how did it break exactly?
did you follow the upgrade procedure from etch release notes?

I want to install a rsbac kernel. what’s the best web hosting way to do this on debian?

I think I screwed this up… it never asked me about the raid arraay… doh… should of set format to one of the drives
– doesnt want to manually build another raid array
I dont know how… but I went from not knowing what mdadm is, to knowing its small behaviors on different systems with different devices

jelly-home: its back up now..
jelly-home: seems it wasnt the kernel..
only the broken raid1
had to get someone to take a look
upgrading from remote is allways a risk

unless there’s remote console access (and even that doesn’t help sometimes)

yea..

raid is kinda foolish… if the software screws up the file or the array, then there is no surviving… and if the drive corrupts, yet doesnt repport a bad sector the repair has to make a 50/50 guess….

raid1 can be mounted as a /dev
lets say you raid1 sda and sdb
and sda fails you can mount sdb directly as well
raid0 however is a lot more unsafe.
and i would say foolish to use..

silent write errors are a good reason why raid5 is bad

unless you need a insane cache speed storage..

raid0 is stupid under any circumstances

jelly-home: im running hardware raid6 these days..

you will always be better served by just upgrading your storage.
RAID0 isn’t going to be half the win that a RAID card with a large cache gets you

Jelly-home I totally agree… but Raid 6 doesnt help much if its a software couruption… I am storing video files, so if one of my video files gets corrupted, I can live

thats why im running hardware..

um, checksum ‘em?

RAID will never save you from your own fuckups.

I’ve got this error after a apt-get update : http://phpfi.com/255868
What can I do ?

therefor it makes sence to use RAID5 because it A) provides a single place to put my data… B) I need a large number of drives, and C) an extra disk + SATA card + extra power does not make it feasable

now the challange is to find a HA filesystem–.

pipeline who says it has to be your mistake…

that doesn’t cost 6 figures?

jelly-home: its cheap

Use LVM now, or you will regret it later.
OCFS2

pipeline linux isnt 100% perfect, nor are the clients your using, or the hard drive controllers, or the motherboards, cpu, etc…

ocfs2 is distributed?

jelly-home: areca 1210 + 4 500GB drives isnt much..

that’s not HA, that’s just a RAID.

jelly-home: For some value of distributed, yes. You need three nodes, and you really need gigabit ethernet levels of latency and bandwidth.

ahh yea

6*500
soon to be 7

jelly-home: But it’s rock solid, well-tested, availble today, and performs well.

hehe.. you will get the LBA64 problem there GreenJelly

Another fun option is parallel NFS. Unfortunately, not available on Linux yet
Definitely worth reading about. It’s a beautiful system.

drbd

aaasolo?

drbd is pathetic.

you have to adress LBA64 when the volume exeeds 2TB

nice, hopefully any on three nodes may fail?

hmm what other options are there.

jelly-home: Yup.

aasolo were does tthis problem exist, and how do you solve it

Buy a SAN or a mainframe, use your vendor’s faculties.

my mouse cursor has stopped responding to movements I make on the synaptic touchpad, although synclient -m shows that the movements are being seen at some level

Peecees are not the solution to all problems

anyone know the secret of amavis with debian? there’s no amavis.conf, what’s up with the conf.d directory?

any idea how I can ‘unfreeze’ the mouse cursor?

!tell Xanthor[aw] about dynamic mmap

unsolo from what I read 2TB was easily extended with ext333
ext3

jelly-home: The catch is that when you get down to two nodes, you can run into a race condition

you have any idea how long can a fsck of that huge an ext3 can take?

jelly-home: Here’s the problem. Linux has no way to GUARANTEE that nothing writes to a block device. When an OCFS2 node “fences” itself (removes itself from the cluster) it kernel panics
jelly-home: It’s the only way to guarantee data integrity

jelly-home YEP! about 6 hours

that’s nice

jelly-home: When you get down to two nodes, and one of them loses network connectivity but not disk connectivity, both nodes will see two heartbeats but only be in contact with one node. And one or both nodes could fence themselves
jelly-home: It’s a race.
jelly-home: with three nodes, a voting process works even when the network is down.

eh, who cares, let the frontend reboot itself. I don’t run the stock market or a bank

hi, I have apt-geted kernel-image-2.6-686 which installed kernel 2.6.18+6 but when I run `uname -a` I get “2.6.18-4-686″
how is that?

unsolo what is this 2 TB problem… I really need to know more

jelly-home: well that kinda makes it less than HA, doesn’t it?

hush

am I missing something?

renewbie yes

?

single block devices are limited to 2TB.
At least, that’s the way it used to be.

whats that?

at least on my hardware controller i have to change adressing to LBA64 when exeeding 2TB..
which means i will have to reformat the volume as well

aasolo I am not using a RAID controller
unsolo purely software raid

there’s an option for large block devices in the kernel since about 2.6.20

well i would imaging these some of the same problems there.

20 disk raid controller costs around 1K

hi all. I cant go hibernate second time on my dell e1505 laptop running etch. if I boot usual way I can press Fn-F2 and It would work , but when I reboot again and the system is restored from hibernate image – Fn-F2 button doesnt work. does anybody know why? thanks.

http://www.webconnexxion.com/raid/product_info.php?cPath=60&products_id=414

any idea how to install RSBAC on debian?

anyways san isnt an option yet.
so what choices do i have

unsolo thats the device

people who build raid5 shit me in general, but people who build raid5 without making sure that the array stripe width is sane shit me greatly.

unsolo but they make a 20
unsolo I cant afford that

and adding a single spindle to an existing raid5 is a good way to get there

are you running 20 drives in raid ?

tried running hibernate script as root? Maybe just the key combo doesn’t work

in a software raid ?

anyone know about installing RSBAC on debian etch?

jelly-home: but it works if I reboot usual way… do I need to add something to hibernate.conf

unsolo not yet

any good options for a fallback file system ?

that’s not what I asked

…. OCFS2 …

ok

it might be only the key-combo is broken after the resume

I didn’t run any scripts manually.

unsolo I will wait till the drives prices come down and as my need for space increases, hence my requirement for an expandable raid system

parallel NFS is pretty sweet too

I’m asking you to try it, instead

You have a whole lot of nodes that serve as block servers, with a small number of nodes that actually do locking / transactions / etc
All backed by your choice of HA shared storage. No local disk crap.

thx jelly-home, it works

unsolo I WISH openBSD supported expandable raid arrays, cause openNAS is an amazing system. openfilter has not gotten the attention it needs

ccd(4)

jelly-home: where is this script?

areca is hot expandable btw..

OpenBSD is ten years behind the times and trailing. It’s never going to modernize. Don’t hold out hope

er… in the “hibernate” package

pipeline wouldnt doubt it… but damn is it secure

but parallel nfs isnt available for linux hosting yet is it

An undeserved reputation.
It’s UNIX. UNIX is insecure by definition. Otherwise it would no longer be UNIX
uIt’s UNIX. UNIX is insecure by definition. Otherwise it would no longer be UNIX /u

(not nexenta)

jelly-home: It’s going to be a long wait.
I don’t believe so.

so ofcs2 is an option..

unsolo:

thx jelly-home, it works

unsolo I WISH openBSD supported expandable raid arrays, cause openNAS is an amazing system. openfilter has not gotten the attention it needs

what is the IRC command for thoose emotes

d’oh
http://developer.osdl.org/dev/nfsv4/site/index.php
I doubt this includes pNFS, but I guess you have to start somewhere

wow, I did screw up the install

does anyone know a repository that has nvidia-kernel-100.14.11?

hopefull a raid build will fix it
hehe

nvidia-glx needs it but it’s not avaivable

crap darn server booted the wrong kernel.. guess i have to travel to it to fix it.

You have to build it yourself.

thanks for the input..

argh _

why do people use ocfs2 with drbd if drbd stinks.. any other ways to do it ?

ocfs2 doesn’t really give me wood I have to say; roll on dynamic striping and variable block sizes at the fs layer

Because they’re cheap shits
a decent SAN costs a lot more than a peecee crammed with disks
You’re looking at 40k just to dip your toe in the water.
a nice peecee costs all of 10k

i know..

Yes, well, there’s the cheap way to do it, and there’s the good way to do it. They’re mutually exclusive.

haha
well is the cheap way stable is what im interested in

Yeah, well, OCFS2 + drbd is a shitty solution.
But it sure is inexpensive

unsolo is the 2TB problem you have is with linux having a single drive that is larger then 2 TB, or is it with the Raid controller??

you’re talking to the wrong people; www.infortrend.com for ex.

its with how you refer adresses

should I choose a different FS?

Hi
When I want to move a directory to another filesystem, sometimes this happens
I move, then break using ctrl+c to check somethin
g

think it would affect the block device somehow..

then when I want to continue move, it gives this error
inter-device move failed: `xxx’ to `xxx’; unable to remove target: Is a directory
I don’t want to remove it, I want to move everything that’s not there yet

all i know is that some say i wuld have to reformat if i extend my Raid volume beyond 2TB..
that i do not want to do.

so what do I do to make sure the data is indeed moved (if I remove the target directory, I lose data)

so if you start by making your volume use lba64 it wouldnt matter basically.

….LVM dude

but if you don’t restripe and your filesystem is too useless to support dynamic striping then you will end up with wobbly access patterns

unsolo I will have an answer for you within a week if you send me a email at big_green_jelly_beanatyahoo.com

average service time? wossat.

unsolo that is if I get it to work

what for.
i like having 2TB large single partition drives.
@home

home are you?

it makes everything so much simpler

ok i can’t find the sources for nvidia-kernel-100.14.11 either _

(dropping and) restriping the lv underneath a filesystem that does not support dynamic striping is a feature, not a bug

I hate having to backup filesystems larger than, say 200-300GB

I can’t for the life of me understand why people try so hard to end up with *crap* IO access patterns.

what does +6 mean in this kernel version 2.6.18+6?

i know but i dont need it at home.

and what does -4 means in this kernel version 2.6.18+6?
oops

this drbd system is ofc not at home.

and what does -4 means in this kernel version 2.6.18-4-686 ?

jelly-home: solution: no backups at home because I can’t afford a tape drive of appropriate capabilities
jelly-home: it annoys me that tapes are so cheap and drives are so spendy

hehe
tape media sure is cheap… the good drives are a different story

jelly-home: thats an advantage with raid6
you can have multiple failures..

Debian’s fourth take on the 2.6.18 kernel series.

so you dont care that much for drive quality

raidz2

the advantage of raid6 is that you hit the garbage read throughput of a degraded raid5 with lower frequency
it won’t help you if you botch your stride size maths though
and it’s always going to suck latency-wise for writes
god help you if some tinker adds a spindle dynamically to it

im using stock drives.. Timing buffered disk reads: 434 MB in 3.01 seconds = 144.04 MB/sec . Timing cached reads: 2488 MB in 2.00 seconds = 1244.08 MB/sec

now fire up 2.5 the number of benchmark threads as you have spindles

hi

and come back and talk to us about average service times

what can i do when aterm segfaults when i try to run it?

well thats on my home server
and its by my terms more than fast enough..
but its not a insane speed.

use urxvt?

latency latency latency

thats what you have 1(0) for.

a late IO may as well not exist

and im talking of the cheap sollutions.

$®%’ã®FLoD’SaLdiri+$®%’ã®FLoD’SaLdiri$®%’ã®FLoD’SaLdiri+$®%’ã®FLoD’SaLdiri$®%’ã®FLoD’SaLdiri+$®%’ã®FLoD’SaLdiri$®%’ã®FLoD’SaLdiri+$®%’ã®FLoD’SaLdiri I’m Using’SpiDeR
ScRipT+$® H t t p : / / W w W . S p i D e R S c R i p T . C j b . N e T +$®%’ã®

the zfs guys have hit the bong hard and come back with an interesting design
cruel and unusual layering violation tricks, sure
but an interesting design and it works well for most mixes

fuck you grecee

thanks

wtf

same ip
^^ how about =B?
+b

I have a somewhat older debian install, using kernel 2.6.8-2-386. I also have a Broadcom 4306 wlan pci card I need to get working. I’ve studied the bcm43xx project, but I’m a bit lost. Does anyone have a link to a good howto on how I can get this card working?

Yeah

if I have filesystem trouble, and files show up in /lost+found, are they guaranteed to be intact?

ha! i found the rotten smell. was coming from this mostly empty tea cup

Permission denied (publickey,password). any clue?
there is no one here dude, get lost

Hi
I have question, how can I schedule a command to run after 2 hour?

!at

extra, extra, read all about it, at is a Delayed job execution and batch processing program or an old IBM computer. austria

sechedule jobs

sleep 2h; cmd

!schedule jobs

that “at” factoid was meant for you
man at arms

what was the issue with at a long time ago?

jelly-home: but dpkg did not say what I want

man at

jelly-home: ok

wtf… “at teatime”

how can i install agetty ?

what do you need it for?

i need to chage Boudrate, Stop Bit, etc of RS232 /dev/ttyS0
jelly-home: any idea ?

hm. baud rate can be changed even with the normal /bin/getty

humm
let me try

there should be a line in default inittab
3:respawn:/sbin/getty -L ttyS0 19200

so… yeap, timezone can’t be chosen in the debian installer properly, as far as I can see, I have tried in both netinstall and regular install

345:respawn:/sbin/getty 38400

hmm can you do backup with preserve over scp ?

can i change that value?

Look up rsync over SSH- that may be more suited for backups.

don’t. scroll down.

I think the problem is that ‘choose language’ is taken as ‘choose country’ when they are two very separate things…

probably it is

add a line like I pasted, only uncommented.

http://www.pastebin.ca/654668

and since this is within a closed network i guess i can just rsync directly right

so uncomment your T0: … line and edit the baud rate properly

ok
then?

reload inittab
“telinit q”

hi

is there any documentation about run linux from a computer wothout HD
like console

what do you mean ‘like console’, what does thta have to do with no HD?

loads – search for netboot / nfsroo

reiser2 use a live cd

no

usbstick

i mean no live cd
is to use a computer like a terminal

the only thing you can do is search if there is a way to boot from network
but then you need a network card with bootrom
google for network boot linux or something like that

jelly-home: telinit q reload the runlevels?

it makes init reload inittab

this is outright funky, it can’t find packages grub nor lilo during the install

jelly-home: but is there any message outpu after use telinit q ?

hello there. Can someone gives me a list or a name for a linux game for two ?

no

you want multiplayer games? of what kind?

nethack! anything from the unreal tournament series will run, as will quake 3 and enemy territory

but there should be a getty working on your serial port 1 now

man

!games

i am using a device that send a string

koth, netris, quake, xevil, bzflag or the only piece of software that is allowed to be non-free, because they are more like an art form than a program, or http://www.uniball-central.com, or #gametome for newer games. WikiWikiWeb:LinuxGames, or
http://www.mtp-target.org, jumpnbump, supertux, gtetrinet and pathogen, or nethack, or freeciv, or dopewars, or http://icculus.org/lgfaq/gamelist.php

nethack is not multiplayer, unfortunately

well, it sort of is
you can find corpses
and share a scoreboard

the bones levels don’t make it multiplayer

was that a requir..oh, right. it was

telnet nethack.alt.org

I can’t believe people quote bzflag as a linux game, it makes me ashamed

just because it remembers some state, doesn’t mean it’s multiplayer

yeah, been there, but frankly I perfer playing on my pc with all the colours and such

eh? bzflag is a classic. And it runs on linux. And it sure beats the crap out of the original bz.

no, *nethack* is a classic, everything else is crud

Is there support for CORE2 QUAD cpu in stable kernel?

bz kinda sucked because it was single-server

jelly-home: i just put a GPS that is configured to send text string using 9600 baud

Surely.

is working
but i have another device that send another string, but doesn`t work
but in windows it send ok

nethack is not a linux game anyway

Any idea where to check if kernel can really make use of additional cores? Eg in multitasking

I’m having some trouble installing, it says it can’t find packages grub and lilo (I tried both), I’m using the regular CD

run htop, look

You’ll know immediately after boot.
cat /proc/cpuinfo
If there was a problem, it probably wouldn’t boot

I’m looking for the info atm, so I could decided which cpu to buy

Back

does it work with minicom for example?

minicom?

yes
!info minicom

(friendly menu driven serial communication program), section comm, is optional. Version: 2.2-5 (sid), Packaged size: 283 kB, Installed size: 1112 kB

With the price/performance on those suckers, I can’t imagine NOT buying a C2Q or the Xeon counterpart.

what’s Q? quad core?

what
what’s a good way to check and see if the PHP5 package was compiled with DB2 support

phpinfo();

I have a feeling I’m going to have to recompile the .deb to support db2

DB2 as in IBM’s database?

where is minicom config file?

yes
I have to connect to a DB2 database on an IS/400
er AS/400 for a customer

if you need nonfree libraries or headers to build that support, it probably isn’t built

i know PHP has the capabilities, but lots of the threads on the web point to needing a driver from IBM — but all these threads are 2 years old

you can set up serial port options while you’re running minicom

but is the debian src even going to have the true PHP source? Or should I scrap the package and just grab it from php.net?

ctrl a and then o

no response
CTRL-A Z for help |115200 8N1 | NOR | Minicom 2.2 | VT102 | Disconected

sure it has php source

I’ll see if I can figure it out

can I unzip multi-part files?

you’ll want to do something like apt-get build-dep php5; apt-get source php5; then edit php*/debian/rules, or wherever the ./configure is called

!bz2

from memory, bz2 is the extension used for file created by bzip2, a program like gzip that gives better compression(particularly on text files) but takes much more memory to run see bzip2. or use the j switch to tar, or install and use unp (ask me about unp)

you’ll need deb-src lines in your sources.list for that

jelly-home: yup — I have those

I even tried installing WinZip via wine, but I still cannot properly unzip multi-part files

concatenate them

unzip doesn’t work?

use JDBC

somehow I don’t think they’re just split

that won’t work, I tried that and then ran zip -F (or tried also zip -FF) to fix the archive, but no succes

as funny as that might be to you I’d use anything — although I hear JDBC setup is a pain

jelly-home: correct, no version of unzip properly unzips multi-part files

bugger. who is the fool that used multipart zip?

I was more poking fun at your choice of platform. I don’t know that anyone in history has ever tried to make PHP talk to a DB2 database

you really think he had a choice?

jelly-home: if it doesn’t work under Linux, it’s not by definition something or someone stupid; everybody uses winzip and multi-part files

jelly-home: He must have a lot of flexibility if he’s able to get as far as selecting a language that isn’t typically used and not even know whether it will work

I’d use rar for multipart

it’s not my choice
it’s a large mfg company that has all of their operations stored on a DB2 server

actually, I’d never heard of zip multi-part files. I’d only ever seen RAR and split tarballs.

iSeries IBM AS/400

I meant the PHP side.
Most DB2 development, regardles of iSeries vs. pSeries, is done with J2EE.

that’s because you’re a Debian user

well I can use anything we want, but it has to be inserted into a MySQL DB on a daily basis

zip multpart format is kind of crap. you _must_ have the last file to make an extraction

Windows users will use WinZip, not rar

you’re killing me, smalls.
RAR was windows-exclusive for many years o_0

Will they ? Most windows users I know use winrar

it’s the multi-part format loved and used by OMGTHESCENE all over the world for ten plus years

warez doodz use rar!

does someone read me when I say I have trouble installing or just when I talk about nethack?

well do you have a way to just pull a simple SELECT from a table in a DB2 database using Debian?

… or so I’m told

rar and those goddern sfv files ;/

okay, so you all like rar for multipart, but meanwhile no Linux user in the world can unzip multipart zip files

you’re the first!

indeed

can the linux version of 7zip do it ?

to actually complain here, yes
other people just think “Linux sucks” and the reinstall Windows

and they’re right

Does the LIVE CD contain crypto-modules cryptsetup-udeb ?
and is it suitable for rescueing crypted fs?

/usr/bin/p7zip: video.zip: unknown suffix — ignored

I’m mildly surprised that WinZip under wine can’t do it, though

p7zip is not winzip

care to upload an example

p.s. there’s no such thing as a multipart zip file. You want to interpret the specific-to-winzip format that very few people use

there is.

jelly-home: no standard

you’re in the wrong here, even PKZip supports it
while we’re at it, is there an actual standardized spec of .zip format?

how can i get the nvidia-kernel package?

jelly-home: Nope, just a common subset shared between implementations
jelly-home: pkzip’s multi-file is specific to pkzip IIRC

so there’s no “standard” to refer to

is there an easy way to find reverse dependencies in debian?

deborphan package has some tools to help with that

well let me ask this
is there any good reason to run the port mapper if I’m not using nfs?

NFS is the only RPC service most people ever use
so if you don’t know, then no, there isn’t.

I could’ve sworn that one time I was installing an everyday application that had portmap in its dependencies, recommends, or suggests
but grepping through /var/lib/dpkg/available, I don’t see any such app

gamin or fam, I think

gamin and fam run their own daemons
no RPC necessary
they use unix sockets

I believe one of the alternatives uses RPC
fam does.

I see nothing about portmap in the fam control info
that might have been where I saw it before, though

I have two box and I would like to backup all content of /home/ onto the second box every 24h, what is the best way ?

in retrospect, it does seem like it was fam

Perhaps take a look at rsync.

rsync for sure

I thought about doing a simple copy but then I am sure they are other optimised way that copy only the modifications
ok

aptitude search ‘~Dportmap’

i’m trying to create a 1TB volume group – vgcreate -M2 -s 32M vg01 /dev/sda1 but I end up with a VG Size of 280.25 GB how to get vg size larger ?

how can i get the nvidia-kernel package?

how can I list the users on my system ?
installed a package that created a user… I need to know what username it created

less /etc/passwd

ta

general protection fault: 0000 [1] what could cause this ? it crashes whole my system full error: http://pastebin.com/m72fac8ce

“general protection fault”??

rpm -q –scripts packagename is probably better.

Does the LIVE CD contain crypto-modules cryptsetup-udeb ?
and is it suitable for rescueing crypted fs?

Hmm.. or is it…

does anyone here use nagios ?

your kernel is tainted, can you reproduce this without nasty modules?

jelly-home: no, as i really need that module for my raid card :§

www.mail35.de

spammer…

jelly-home: where is it getting that information from? how come when I do apt-cache show fam, it doesn’t mention portmap?

ah, fakeraid with a binary module? I’m not sure you’ll get much help

www.mail35.de

.1.1-12), libstdc++6 (= 4.1.1-12), lsb-base (= 3.0-6),
!ops tommy15 spamming

Hydroxide, bob2, caphuso, dondelelcaro, doogie, eeyore-, ElectricElf, ):, helix, ljlane, LoRez, RichiH, mentor, Netsnipe, TML, walters, xk, gravity, azeem, Maulkin, stew, peterS, Alife, Myon, Ganneff, Maulkin, weasel, zobel: jelly-home complains about: tommy15 spamming

oh wait, you’re right

www.mail35.de

how can i get the nvidia-kernel package?

Hello all , Is there any way to check if my ssh server is working fine?

ssh into it?

ssh to it?
DUH

fam isn’t even in /var/lib/dpkg/available

yea jelly-home i know its fucking crap, but controller doesn’t work without the driver

www.mail35.de

please stop that

I’m not even running fam, but for some reason, nautilus seems to be notified immediately when a file or directory’s name changes
or when one is deleted

maybe you have the other one
gamin

I don’t have gamin either

www.mail35.de

nautilus can’t use gamin
other file browsers can
it’s either polling or it’s using some built-in kernel mechanism
and I doubt it’s polling

[how] can apt-cache show me what repository a package will/can come from?

maybe Gnome uses something else to set up dnotify/inotify

isn’t there some sort of kernel driver to replace fam now? I seem to recall hearing something about this

apt-cache policy foo might help

jelly-home: u think its caused by the driver ? or u think it might be something else

how in the hell do you pay attention to all of the channels you’re in?

no idea

I don’t think I’ve ever seen anyone in that many channels in all the years I’ve been on IRC

I don’t pay attention to them all, but I do have my client monitor for various things (like people calling for ops, etc)

how large is your logical device?

what iss the grub command to turn off ACHI?

I heard about problems with XFS on large block device

how can i get the nvidia-kernel package? …

or people asking about the number of channels you’re in? :p

jelly-home: i’m running xfs on a 5TB block device
working perfect :s so :s

!nvidia-kernel

nvidia-kernel packages do not exist any more – they were too much of a pain to keep in synch with kernel-images. To install an nvidia-kernel package, apt-get install module-assistant && m-a a-i nvidia

I see, it’s working exceedingly perfectly

thanks xand

if you can get your fakeraid working with free drivers, try asking on lkml or xfs lists

argh
“Bad luck, the kernel headers for the target kernel version could not be found and you did not specify other valid kernel headers to use “

jelly-home: it doesn’t work with free drivers
stopid crap controller

sucks to be you?

m-a prepare

!kernel headers

http://uwsg.iu.edu/hypermail/linux/kernel/0007.3/0587.html

I try not to buy hardware that works with unsupported drivers

i know i’m learned also now
:p

buying hardware for linux can be tricky

doesn’t help

is there a way to ‘mask’ a package from apt? make it think it is installed so that other packages that depend on it will not be blocked?

note I didn’t exclude closed or 3rd party drivers, only “unsupported”

its supported

why is deborphan a binary? it seems like it would be a relatively simple script in any number of languages

by highpoint drivers
…;

we’re running RHEL instead of the preferred Debian on some machines just to have support for SAN