OESF Portables Forum
Everything Else => Zaurus Distro Support and Discussion => Distros, Development, and Model Specific Forums => Archived Forums => Debian => Topic started by: flux on November 01, 2007, 06:54:18 pm
-
hello, excuse me to start a new topic ;
i installed debian on c3100 with the method without network (...)
http://www.bigbridgezau.sakura.ne.jp/dev/debian/ (http://www.bigbridgezau.sakura.ne.jp/dev/debian/)
i have 2.6.20-cpufreq kernel
dist-upgrade to sid
apt-get install xserver-xorg-video-fbdev laptop-detect xbase-clients fluxbox (...)
in root : startx works -> i arrive under fluxbox.
in user : startx
i get a library pb (from /opt/kdrive/libts-0.0.so.0 , it can't open this file)
and a /tmp write error.
i did : chmod 1777 /tmp -> no more /tmp error, but still the libts-0.0.so.0 pb
i tried to ln -s this library to /usr/bin -> doesnt work.
then :
LD_LIBRARY_PATH=/opt/kdrive startx -> gives me a ld-linux.so.3 pb (can't open the file)
so my question : is there another way to be able to startx in user ?
-
try to follow this guide :
http://inv2004.googlepages.com/z1000_debian.txt (http://inv2004.googlepages.com/z1000_debian.txt)
just after the line "couse standart Xorg is too hard for zaurus we'll try to install kdrive"
i remember i suceed but that was not easy. i remember i had to grab some files from angstrom then i was able to launch X with my default user.
-
i followed instructions, changing for my system ...
cp /opt/kdrive/libts-0.0.so.0.1.1 /usr/local/lib
ln -s
ldconfig
(same for /usr/local/bin/Xfbdev) (and for the $HOME/.xserverrc...)
after that, i still had the same problem, as if ldconfig didn't work
i did a : ln -s /usr/local/lib/libts.0.0.so.0.1.1 /usr/lib/libts.0.0.so.0
with it, i'm on a new error :
/usr/local/bin/Xfbdev: lib/libgcc_s.so.1: version 'GCC_3.5" NOT FOUND (required by /usr/local/bin/Xfbdev)
/usr/local/bin/Xfbdev: /lib/libdl.so.2: version "BLIBC_2.4" not found (required by /urs/lib/libts-0.0.so.0)
before posting here, i had tried to echo /opt/kdrive >> /etc/ld.so.conf && ldconfig
-> i had to reinstall, because library versions pb for all program....
-
and now, i :
ln -s both library i told in the precedent post...
that made me the same error as when i had /opt/kdrive in ldconfig....
don't have any more program workin on bash
reboot -> kernel panic
-> reinstall.
but... tell me guys... are you all working on X under root ? there is no easy way (automatic way) to startx as a user (even in xdm, i could not log while i was "user"...) i can't believe it...
-
hello, excuse me to start a new topic ;
i installed debian on c3100 with the method without network (...)
http://www.bigbridgezau.sakura.ne.jp/dev/debian/ (http://www.bigbridgezau.sakura.ne.jp/dev/debian/)
i have 2.6.20-cpufreq kernel
dist-upgrade to sid
apt-get install xserver-xorg-video-fbdev laptop-detect xbase-clients fluxbox (...)
in root : startx works -> i arrive under fluxbox.
in user : startx
i get a library pb (from /opt/kdrive/libts-0.0.so.0 , it can't open this file)
and a /tmp write error.
i did : chmod 1777 /tmp -> no more /tmp error, but still the libts-0.0.so.0 pb
i tried to ln -s this library to /usr/bin -> doesnt work.
then :
LD_LIBRARY_PATH=/opt/kdrive startx -> gives me a ld-linux.so.3 pb (can't open the file)
so my question : is there another way to be able to startx in user ?
I think the LD_LIBRARY_PATH=/opt/kdrive command makes that the _only_ place it looks for libraries.
The X script in /opt/kdrive has that, but in the context of a script which says:
#!/bin/sh
export TSLIB_TSDEVICE=/dev/input/event1
if [ ! -e /etc/ts.conf ]; then
cp /opt/kdrive/ts.conf /etc/ts.conf
LD_LIBRARY_PATH=/opt/kdrive /opt/kdrive/ts_calibrate
fi
export LD_LIBRARY_PATH=/opt/kdrive
exec /opt/kdrive/Xfbdev -screen 480x640@270 "$@"
Try that script, if you don't have it.
-
ok...
from beginning : login as user.
su -
chmod 1777 /tmp
exit
cp /opt/kdrive/X .xserverrc && chmod 755 .xserverrc
startx
--> Fatal server error : Server must be suid root
---------------------------------------------------------------------------------
su -
chmod 4755 /opt/kdrive/Xfbdev
exit
startx
---> /opt/kdrive/Xfbdev: error while loading shared libraries: libts-0.0.so.0: cannot open shared object file: No such file or directory ; giving up
xinit: Connection refused (errno 111) : unable to connect to X server
xinit : No such process (errno 3): server error
---------------------------------------------------------------------------------------------
chmod 4755 .xserverrc
startx
--> same as above
any idea ?
-
ok...
from beginning : login as user.
su -
chmod 1777 /tmp
exit
cp /opt/kdrive/X .xserverrc && chmod 755 .xserverrc
startx
--> Fatal server error : Server must be suid root
---------------------------------------------------------------------------------
su -
chmod 4755 /opt/kdrive/Xfbdev
exit
startx
---> /opt/kdrive/Xfbdev: error while loading shared libraries: libts-0.0.so.0: cannot open shared object file: No such file or directory ; giving up
xinit: Connection refused (errno 111) : unable to connect to X server
xinit : No such process (errno 3): server error
---------------------------------------------------------------------------------------------
chmod 4755 .xserverrc
startx
--> same as above
any idea ?
1) My server was not installed suid root.
2) When I kill my xdm, then exit X with a Ctrl-Alt-Backspace, I see the same behavior when running startx. When xdm starts, it starts as root (or suid root). So, it would seem that startx is not set up correctly to run from the console.
Try running xdm in startup... It's been a long time since I have run any computer without xdm or gdm. At first I hated it, but then I got used to it.
-
ok, it works with xdm (that's weird, before the reinstall, it wasn't).
i wish i'll understand how to "startx" because i never be able to love *dm
thx you, if you have a clue that's come up for "startx", i'm listening and test.
-
ok, it works with xdm (that's weird, before the reinstall, it wasn't).
i wish i'll understand how to "startx" because i never be able to love *dm
thx you, if you have a clue that's come up for "startx", i'm listening and test.
oooooh now i remember i had the same problem !!!! i could not login with my normal user without xdm !
launching xdm was the only solution i had...
-
a friend explained me that it was because the fb feature needs root privilege (i still don't understand why with chmod 4755 /opt/kdrive/Xfbdev, iwould'nt work... but .)
i wonder ... apt-get install xserver-xorg-video-fbdev xbase-clients xdm xserver-xorg-input-evtouch (...)
my pointer doesn't respond.
i can't launch : /opt/kdrive/ts_calibrate (same libts-0.0.so.0 error as above) this under root and user.
would you know what should i do to calibrate the pointer ?
-
an other way to ask this question would be :
in /opt/kdrive/ts.conf
which "module_raw" should i choose, as i'm using a zaurus sl-c3100 ?
i tried the "corgi" one... but doesn't work. and i still can't calibrate the screen when i "rm /opt/kdrive/ts.conf"
-
an other way to ask this question would be :
in /opt/kdrive/ts.conf
which "module_raw" should i choose, as i'm using a zaurus sl-c3100 ?
i tried the "corgi" one... but doesn't work. and i still can't calibrate the screen when i "rm /opt/kdrive/ts.conf"
leave /opt/kdrive/ts.conf as is (it comes correct for C3X00) and "rm /etc/ts.conf" instead, then restart xdm -> you should be asked to do calibration.
I guess you could have even other problems :
xserver-xorg-video-fbdev and xserver-xorg-input-evtouch should be provided by installing zaurus-xserver from the titchy-repo ("neilandtheresa"). They are in /opt/kdrive. When installing them in /usr as well, I presume you could have created a mix that little Z doesn't like.
Just a tip : the way things are going right now, and since you installed through the rootfs-tarball, you could as well install the eabi version. Most users here seem to be going that way and you'd probably get more support. (some of the answers in this thread are for the eabi version).
Chero.
-
i finally get it worked. but at the price of a reinstall. for people who would follow this way... :
i installed debian on c3100 with the method without network (...), with the rootfs-tarball
http://www.bigbridgezau.sakura.ne.jp/dev/debian/ (http://www.bigbridgezau.sakura.ne.jp/dev/debian/)
change in /etc/apt/source.list to "sid" then,
login : root
apt-get update && apt-get dist-upgrade
apt-get install xorg # whole xorg... many non needed... but this way, it worked
apt-get install xdm fluxbox
chmod 1777 /tmp
su - user
echo -e "xterm&\n fluxbox\n" > .xsession && chmod 755 .xsession
exit
xdm
then, the calibrate screen appears... and then xdm to log. if you doesn't "chmod 1777 /tmp", you won't be able to log as a user.
QUESTION : why /tmp isn't world-writable ? usually, it is, or am i dreaming ?
if anyone has a/some question(s), please ask !
-
i finally get it worked. but at the price of a reinstall. for people who would follow this way... :
i installed debian on c3100 with the method without network (...), with the rootfs-tarball
http://www.bigbridgezau.sakura.ne.jp/dev/debian/ (http://www.bigbridgezau.sakura.ne.jp/dev/debian/)
change in /etc/apt/source.list to "sid" then,
login : root
apt-get update && apt-get dist-upgrade
apt-get install xorg # whole xorg... many non needed... but this way, it worked
apt-get install xdm fluxbox
chmod 1777 /tmp
su - user
echo -e "xterm&\n fluxbox\n" > .xsession && chmod 755 .xsession
exit
xdm
then, the calibrate screen appears... and then xdm to log. if you doesn't "chmod 1777 /tmp", you won't be able to log as a user.
QUESTION : why /tmp isn't world-writable ? usually, it is, or am i dreaming ?
if anyone has a/some question(s), please ask !
whow!
this means you're using xorg and not kdrive if i understand well
how is your memory ?
-
nop... the problem was since the beginning "kdrive" and still is... i think it is automatically installed with zaurus-debian-hardware.deb or an other stuff... i don't know. but it's /opt/kdrive/Xfbdev that is launched.
i know kdrive is only 2Mo instead of 8 for xorg... but in order to use "startx" as a user, i'm still thinking if i wouldn't remove the kdrive... it would be such a mistake... maybe, probably, certainly.
-
well ... chero... i was reading "debian titchy vs EABI" ... and as at third post, they forget totally about titchy, and say how EABI is so much wonderfull... i'll follow your advice you gave me before, and reinstall while i don't have to many things running on my new debian-zaurus.
i'm choosing the EABI install with 2.6.23.1 ... have to install UBOOT.
hope it's a good choice !
thx everybody
-
There is a new OESF wiki entry for the installation instructions ( https://www.oesf.org/index.php?title=Debian...eries_models%29 (https://www.oesf.org/index.php?title=Debian/Zaurus_installation_guide_%28for_all_C-series_models%29) ) . Hope it is useful here.
-
thx zdevil !
i already knew it, as i followed your (long) post beginning by it, and the other post named 'the wiki, where is it..."
it's right now "tar xpvf armel*" ... hope there will be no pb while installing uboot
BTW : the link for downloading uboot in the wiki can not work ... the url is bad because of a bad c/p ... maybe you should update that
thx for the work done on the doc, i'll tell you if there is other pb while following you "Q18" guide.
-
Thanks. The uboot link in Q18 is fixed now.
Good luck and enjoy!
-
hum... it's not in "5 min" that i'll enjoy i think...
i made THE mistake... after booting in emergency mode, at the question "do you want to proceed with pdaxrom installation or boot emergency mode [y-n] , i don't know why i pressed "y"
it "Rebooting" , but since it's halted, my zaurus don't respond to anything. the power -> no light,
press OK + ac -> nothing
presse D+B + insert battery -> nothing
i wonder where i'll find the way to fix it... in oesf forum probably... but f**k i had "somethings" else to do !
let's googleing to find the trick
-
Try removing all power sources for a couple of minutes.
Then plug in AC and boot while pressing [OK].
If it boots into the emergency mode, then you can proceed with the rest.
If nothing works, the last resort: NAND restore -> reinstall uboot -> reinstall kernel (i.e. from Step 6 onwards)
And pay attention to Steps 7.1 and 7.2. You need a different set of files for kernel installation.
The good news is your rootfs is still intact. So you don't need to reformat the disk and extract the rootfs tarball again.
-
hum... fn + d + m -> i can arrive at "nand restore" ... but for that, i don't need to reinstall the stuff from trisoft.de and then, redo all ?
with what you said before, it seems not... i 'll try, and cross my finger
http://forum.zaurusfr.org/viewtopic.php?=&p=2559 (http://forum.zaurusfr.org/viewtopic.php?=&p=2559)
^^ post 'im reading about this right at this moment
-
remove + ac + plug worked ... i'm in emergency mode... i was afraid for nothing. cool
-
That's the tricky thing about the current uboot (from pdaXrom).
But for Debian, if you have alreaady extract the rootfs tarball, the kernel module tarball and the after installation tarball, then the rootfs is more or less done.
What's left is to get uboot and the kernel installed properly so that the machine can boot into the rootfs on the internal drive...
Good luck.
-
There is another typo in the Q18 (just corrected ): download the yonggun kernel from http://yonggun.tistory.com/57 (http://yonggun.tistory.com/57) , not http://yonggun.tistory.com/53 (http://yonggun.tistory.com/53)
-
http://forum.zaurusfr.org/viewtopic.php?=&p=2559 (http://forum.zaurusfr.org/viewtopic.php?=&p=2559)
^^ post 'im reading about this right at this moment
are you french ?
bon courage!
-
@ ZDevil : in fact, the page you gave first had a link for the good page... i forgot to tell you that... with all this korean characters, it wasn't "easy" to find... but ctrl-f 2.6.23.1 found it well
@ jpmatrix : et ouais, j'suis francais
sorry, i didn't notice i posted a french link for no-francophone users
-
Yes, the yonggun kernel is korean.
-
ZDevil :
i booted in emergency mode (u-boot emergency mode) because of console pb while booting
according to " df " /dev/hda1 is already mounted on /mnt/cf
BUT, i can't "ls" or "cd" into /mnt/cf ... it makes the zaurus freeze = i have to remove all power and reboot.
i tried to mount or umount /dev/hda1, it doesn't want to (device busy).
what do you think i should do ?
-
Have you already extracted everything (rootfs, kernel modules, after installation) *before* installing uboot?
If yes, then you don't need to touch /mnt/cf at all. If you choose "no" to the question,
simply do this (suppose the yonggun kernel (renamed to kernel.img) and the autoboot.sh are on SD; actually they should be there):
cd /mnt/card
./autoboot.sh
The flashing goes really fast (just a couple of seconds!) and then it will tell you it starts rebooting. Press [2] all along until you see things are booting properly (like mounting the rootfs from /dev/hda1 to /).
And yes, the current uboot has this strange problem. I got it too. And nobody told me why and there seems nothing has been said in the posts. That's why I emphasized that we have to install the rootfs before installing everything else.
-
hum... don't you need to update what you just said here in the 7.4 of the Q18 wiki ?
-
Actually it doesn't matter. If you say yes then the system will look for the autoboot.sh on SD. If you say no then you can run the autoboot.sh manually. Either way you don't need to touch the CF.
-
wait, sorry ... when i saw "unable to open an initial console", i thought it would freeze, and i reboot immediatly to try the MAKEDEV /dev/console...
in fact, it keeped going on the boot process ... i'm so stupid ! sorry to make you loose time
-
One more note: remove the CF card from the CF slot (i don't mean the internal drive) when installing the kernel.
-
wait, sorry ... when i saw "unable to open an initial console", i thought it would freeze, and i reboot immediatly to try the MAKEDEV /dev/console...
in fact, it keeped going on the boot process ... i'm so stupid ! sorry to make you loose time
"Slow down, you move too fast, you've got to make the morning last
Just kickin' down the cobble-stones, lookin' for fun and feelin' groovy."
Read the lines right under "FIRST BOOT AND SYSTEM CONFIGURATIONS".
-
yeah, you're right.
but i think this time, there is a problem.
step 10 ... you ask me to "cd /afterinstall " and then , to ../afterinstall/1st-cfg.sh
but ... it fails (i told "y" at the 1st question) .
in fact .. you make us do, in 5b what the 1st-cfg.sh do later .... and it doesn't find the package.
the script should begin at line " cd package " (sorry, vi (not M) doesn't tell me the line... )
-
Bingo! You found it....
You are right. I should take off the first few lines in the 1st-cfg.sh.
Please let me know how the rest goes.
-
and ... in 1st-cfg.sh
while reading into .... the "start-stopdeamon.REAL" line est false ...
-
and ... in 1st-cfg.sh
while reading into .... the "start-stopdeamon.REAL" line est false ...
Can you see if there is start-stop-deamon.REAL in /sbin ? It should be there, according to 2or0's instructions. I got that too on mine.
If not, then i can upload it here for you.
-
then, as the "loadkey" cmd isn't on the system, the keymap still wrong...
-
nop, there isn't in /sbin .... that's why i told you this line was false
-
then, as the "loadkey" cmd isn't on the system, the keymap still wrong...
That's not expected if console-tools is installed (loadkeys is in the console-tools package). Do you see it installed?
-
with the wrong keymap... i can't go up ... and all was made very quickly ... maybe, in plus of the echo, you should add some :
read nothing
in order we can see errors and what append (i know because i read the script... but others, in the future...)
question : what is the difference between start-stop.FAKE et start-stop.REAL ?
i can't "more" it ... i wonder...
-
I checked the console-tools package (in afterinstall/packages) again. The "loadkeys" command is there, and the script executes this... so it should work.
Again do you need the binary?
-
with the wrong keymap... i can't go up ... and all was made very quickly ... maybe, in plus of the echo, you should add some :
read nothing
in order we can see errors and what append (i know because i read the script... but others, in the future...)
question : what is the difference between start-stop.FAKE et start-stop.REAL ?
i can't "more" it ... i wonder...
How about some waiting commands? (but that's not crucial for now)
The start-stop-daemon thing is 2or0's. I put it in the tarball and we both checked it.
-
libconsole + console_tools wasn't installed ... libiwi + wirelesstools i don't know (i didn't check), i installed them manually (sorry not to have checked :/ )
but i'm pretty sure they wasn't
in fact, all the script failed the 1st time. why ? i don't know
in a second time it failed after pb with start-stop.REAL ...
but we don't really see it fail, because of all "done" and no "error"
-
so ... there still isn't an hostname, a hwclock...
all has to be done manually maybe
i'm configuring network, to get the start-stop.REAL and i'll retry the script, slowly with some "read nothing" in order to see what really appends
i keep you in touch
-
libconsole + console_tools wasn't installed ... libiwi + wirelesstools i don't know (i didn't check), i installed them manually (sorry not to have checked :/ )
but i'm pretty sure they wasn't
in fact, all the script failed the 1st time. why ? i don't know
in a second time it failed after pb with start-stop.REAL ...
but we don't really see it fail, because of all "done" and no "error"
1st time failure: I guess it may have to do with the beginning "#!/bin/sh" line.
The second time failure has to do with missing start-stop-daemon.REAL file in your system. (But I had it and 2or0 wrote it in the instructions here: http://yonggun.tistory.com/54 (http://yonggun.tistory.com/54) )
The thing is the 1st-cfg.sh will cd into the packages dir to dpkg -i the packages. Others (jpmatrix and tux) confirmed that the first script ran fine.
-
so ... there still isn't an hostname, a hwclock...
all has to be done manually maybe
i'm configuring network, to get the start-stop.REAL and i'll retry the script, slowly with some "read nothing" in order to see what really appends
i keep you in touch
Then I think something may go wrong with the download. I will try the scripts on my C860 this evening to see how they fare.
-
1st, you should had to the script a line which tails :
echo -ne "auto lo\n\tiface lo inet loopback" >> /etc/network/interfaces
for hostname, that's fine in fact
PB ! do you think i can "apt-get install ssh"
or it will kill the "afterinstall" normal process ? it is for the "start-stop-deamon
-
now that i put 'start-stop-deamon.REAL" at the right place, the script goes well untill it fails on :
* mv cannot stat /etc/atp/sources.list .... but it seems normal, it already mv it before, without cp /afterinstall/settings/sources.list (but you'll check tonight on your cxxx
* cp : cannot stat hwclock.rules : no such file /dir : correct in 1cfg.sh cp settings/hwclock.rules ....
-
The thing is the 1st-cfg.sh will cd into the packages dir to dpkg -i the packages. Others (jpmatrix and tux) confirmed that the first script ran fine.
hmm just a word : i didn't launch the 1st-cfg script as all seems yet done with the 0827 rootfs ! i only had to launch the 2nd-cfg script.
-
pb : when, after i cp settings/hwclok.rules /etc/udev/rules.d
hwclock -w
Cannot access the Hardware Clock via any known method. Use the --debug option to see the details of our search for an access method
i did the "hwclock -w --debug" ... it tails me it doesn 't find /dev/rtc
so I :
cd /dev
ln -s rtc0 rtc
cd -
hwclock -w --debug
it seems it work, but i don't really know what it is for.
let's go to step 2, as i have a working connection (and no ssh... i didn't installed it before)
-
The thing is the 1st-cfg.sh will cd into the packages dir to dpkg -i the packages. Others (jpmatrix and tux) confirmed that the first script ran fine.
hmm just a word : i didn't launch the 1st-cfg script as all seems yet done with the 0827 rootfs ! i only had to launch the 2nd-cfg script.
what is "0827 rootfs" ?
-
hum ZDevil does'nt speak anymore... i bet he's test right now the install, not want to wait for tonight
-
what is "0827 rootfs" ?
it is the last version of the armel rootfs from ads:
http://www.applieddata.net/forums/topic.asp?TOPIC_ID=2443 (http://www.applieddata.net/forums/topic.asp?TOPIC_ID=2443)
-
for the 2-cfg.sh ... i think as i'm reading it that i won't start it, because it will install me stuff i don't want/need (icewm...) for example, i like better fluxbox (guess where's my nickname from )
but, ZDevil, as jpmatrix said before he tested the 2nd (not the 1st), i guess the 2nd runs without problem.
-
i don't have this one... i've got the one which is pointed by the wiki.... i got it on the same site, but not same exact url ... no matters i hope
-
i don't have this one... i've got the one which is pointed by the wiki.... i got it on the same site, but not same exact url ... no matters i hope
yeah we think both leads to the same as soon as the first apt-get dist-upgrade is done
-
jpmatrix : your /etc/fstab is virgin, too ? like the /etc/network/interfaces ?
i'm fearing to look at other critical /etc .... network/interfaces.... it's fine to be virgin (but iface lo ... should be there !)
but fstab ...
-
side note:
Have you already extracted everything (rootfs, kernel modules, after installation) *before* installing uboot?
If yes, then you don't need to touch /mnt/cf at all. If you choose "no" to the question,
simply do this (suppose the yonggun kernel (renamed to kernel.img) and the autoboot.sh are on SD; actually they should be there):
cd /mnt/card
./autoboot.sh
The flashing goes really fast (just a couple of seconds!) and then it will tell you it starts rebooting. Press [2] all along until you see things are booting properly (like mounting the rootfs from /dev/hda1 to /).
And yes, the current uboot has this strange problem. I got it too. And nobody told me why and there seems nothing has been said in the posts. That's why I emphasized that we have to install the rootfs before installing everything else.
I guess the problem is you can't read an ext3 file system from the emergency system.
BTW : if nothing works after trying to install uboot or flashing the kernel, there is absolutely no need to do a nand-restore !! Just try to install uboot again and then to flash the kernel again. A nand-restore before installing uboot is a terrible waste of time.
-
jpmatrix : your /etc/fstab is virgin, too ? like the /etc/network/interfaces ?
i'm fearing to look at other critical /etc .... network/interfaces.... it's fine to be virgin (but iface lo ... should be there !)
but fstab ...
yes,
don't worry, your system can boot without it
now you can customize it with something like :
/dev/hda1 / ext3 rw 1 1
/dev/hda2 /none swap sw 0 0
proc /proc proc defaults 0 0
sysfs /sys sysfs defaults 0 0
/dev/mmcblk0p1 /mnt/card noauto defaults 0 0
/dev/hdc1 /mnt/cf noauto defaults 0 0
-
that's what i was about to do...
question : in your /etc/apt/sources.list, you have "deb http://armel-debs.appliedata.net/debian (http://armel-debs.appliedata.net/debian) sid main
if yes, does it failed when you apt-get update because it doesn't find "Package.gz" ? me yes, got a 404 error.
the other, ftp.gnuab.org works.
-
that's what i was about to do...
question : in your /etc/apt/sources.list, you have "deb http://armel-debs.appliedata.net/debian (http://armel-debs.appliedata.net/debian) sid main
if yes, does it failed when you apt-get update because it doesn't find "Package.gz" ? me yes, got a 404 error.
the other, ftp.gnuab.org works.
I'm using these :
deb http://ftp.easynet.be/ftp/gnuab/debian (http://ftp.easynet.be/ftp/gnuab/debian) sid main
deb http://ftp.easynet.be/ftp/gnuab/debian (http://ftp.easynet.be/ftp/gnuab/debian) unreleased main
and they didn't let me down in the last couple of days ...
http://armel-debs.appliedata.net/debian (http://armel-debs.appliedata.net/debian) seems to be down every now and again.
Chero.
-
hum ZDevil does'nt speak anymore... i bet he's test right now the install, not want to wait for tonight
nope. I just did my shopping...
But jpmatrix you said you ran the first script the other night... and tux... well, perhaps he's drunk while posting ...
-
that's what i was about to do...
question : in your /etc/apt/sources.list, you have "deb http://armel-debs.appliedata.net/debian (http://armel-debs.appliedata.net/debian) sid main
if yes, does it failed when you apt-get update because it doesn't find "Package.gz" ? me yes, got a 404 error.
the other, ftp.gnuab.org works.
Don't panic. I've put down TWO locations in sources.list. And they point to exactly the same set of files. Either one will work. If one fails, apt-get install is smart enough to pull things from the other.
-
Now I am switching my C860 from OABI to EABI, and am going to run both scripts. So I can tell what works and what doesn't.
-
i've got a problem, after apt-get dist-upgrade + reboot
it seems it comes from start-stop-deamon (i have put the st-st-daemon.REAL in /sbin/ )
while booting :
starting crond: /etc/rc2.d/S89cron: line 24: 985 Segmentation fault start-stop-deamon --start --quit --pidfile /var/run/crond.pid --name conr --starbtas /us/sin/cron --$LSBNAMES failed
AND : is there a way to avoid the automatic startup on a wrong partition that gives me a VFS error while kernel panic, such as debian become the "n°1" ... and so, not have to press "2" every time (cause every time, i forget, have to pull out battery, and now with uboot, when i plug battery, no wait for the battery cover and clip-button ... the zaurus starts immediatly. that's awful...)
-
ZDevil ... i don't know at which stage you are...
but according to the second script :
ts_calibrate
ts_open : no such file or directory
tell me where you'll there