hal gone mad! [Solved]

Mr Green · 2006-01-30 09:28:19

I updated hal a couple of days ago now system refuses to start....

pacman.log

[01/27/06 08:59] upgraded hal (0.5.5.1-5 -> 0.5.6-1)
[01/28/06 22:59] synchronizing package lists
[01/29/06 02:30] removed hal (0.5.6-1)
[01/29/06 02:31] could not extract etc/rc.d/: Permission denied
[01/29/06 02:31] could not extract etc/rc.d/hal: Permission denied
[01/29/06 02:31] errors occurred while installing hal
[01/29/06 02:31] installed hal (0.5.5.1-5)
[01/29/06 02:32] could not extract etc/rc.d/: Permission denied
[01/29/06 02:32] could not extract etc/rc.d/hal: Permission denied
[01/29/06 02:32] errors occurred while installing hal
[01/29/06 02:32] installed hal (0.5.5.1-5)

As you can see rc.d shows permission denied, I have tried to change permissions remove move but nothing works :?

Does anyone know how to solve this very annoying problem :?:

/me in Knoppix atm

vacant · 2006-01-30 11:04:44

Checked your file system recently? "reiserfsck --check" or equivalent.

scarecrow · 2006-01-30 11:07:54

Surely enough you don't have a corrupt hal 0.5.6-1 package?
And you may have before upgrading doing some /etc/rc.d/hal stop ?

Mr Green · 2006-01-30 11:50:30

vacant wrote:

Checked your file system recently? "reiserfsck --check" or equivalent.

Well.... I did that & reiserfsck --rebuild-tree /dev/hda1rc.d permissions are now ok .... (I can see see rc.d & its files now)

Guess I need to edit my fstab now ;-)

Thanks for your help

Back in Arch now very happy, thank you for your help!!!

T-Dawg · 2006-01-30 12:59:02

reiser for the loss.

Mr Green · 2006-01-30 13:19:36

pacman -Syu
error: /var/lib/pacman/local/nvidia-1.0.8178-8/desc: No such file or directory
:: Synchronizing package databases...

:cry:

T-Dawg · 2006-01-30 13:34:02

ROFL!

just delete /var/lib/pacman/local/nvidia-1.0.8178-8 and reinstall it using the force switch.

Mr Green · 2006-01-30 13:49:57

yeah I like done that .... updated system seems ok now

might be a good time to backup lol

T-Dawg · 2006-01-30 13:51:29

furthermore, you may want to check your hard drive -it sounds like it could be failing. Most hard drive manufactures supply free diagnostic software usually in the form of a boot up disk to test the drive. Seagate has an excellant utility that seems to work on all drives...

Mr Green · 2006-01-30 14:01:55

I think stopping reiserfs from checking was a bad idea ....

Will keep an eye on system for a while (/home /etc are backed up already)

might dump hda1 on hdb be safe.....

I'm always messing around installing removing stuff it was just a matter of time before system broke ....

New drive emmm /me starts saving....

I was confused, I could not fix system under chroot (the normal way!) rc.d rpc motd were showing permission denied

using reiser tools is not always a good idea but I'm back in Arch atm

Maybe its time to go back to ext3 (don't open that can o worms!!!)

elasticdog · 2006-01-30 14:40:15

I can't seem to find the thread, but I thought I read someone here this past week mention that you don't have to explicitly tell ReiserFS partitions to <code>fsck</code> in your <code>fstab</code> since ReiserFS automatically does it when you mount the partition. I did notice that when I was doing it explicitly, the process showed up twice on my bootchart output. Is that right?

Mr Green · 2006-01-30 14:46:44

well I changed fstab to read 0 0 ... I think you are right it does check anyway ...

I could have done something to filesystem by mistake ...

system may not have shutdown correctly ... no idea ....

I have changed kernel so many times over the past week its not surprising that stuff gonna break,,,

Just have to keep an eye on system for a few days .... may reinstall Arch if I still have got problems (would clean up drive!)

may have done more damage than I first thought

/me has Knoppix at the ready ;-)

scarecrow · 2006-02-02 11:20:07

Hijacker here... having current problems writing without hardlocks in a large ( 250GB ) ReiserFS 3.6 partition, even after a "reiserfsck --rebuild-tree". Used to work without any problems a couple of kernel revisions ago, and a thorough check with hddregenerator failed to find any bad blocks.
Coincidence?

iphitus · 2006-02-02 11:50:57

Another good idea is to check dmesg. If the kernel has any issues reading or writing -- a common sign of a failing/broken/damaged drive, they will appear there.

Watch out for curly brackets especially

iphitus

Mr Green · 2006-02-02 13:23:29

Well I'm running ok atm .....

emmm maybe a script to scan for {} in dmesg give me some sort of warning?

Arch Linux

#1 2006-01-30 09:28:19

hal gone mad! [Solved]

#2 2006-01-30 11:04:44

Re: hal gone mad! [Solved]

#3 2006-01-30 11:07:54

Re: hal gone mad! [Solved]

#4 2006-01-30 11:50:30

Re: hal gone mad! [Solved]

#5 2006-01-30 12:59:02

Re: hal gone mad! [Solved]

#6 2006-01-30 13:19:36

Re: hal gone mad! [Solved]

#7 2006-01-30 13:34:02

Re: hal gone mad! [Solved]

#8 2006-01-30 13:49:57

Re: hal gone mad! [Solved]

#9 2006-01-30 13:51:29

Re: hal gone mad! [Solved]

#10 2006-01-30 14:01:55

Re: hal gone mad! [Solved]

#11 2006-01-30 14:40:15

Re: hal gone mad! [Solved]

#12 2006-01-30 14:46:44

Re: hal gone mad! [Solved]

#13 2006-02-02 11:20:07

Re: hal gone mad! [Solved]

#14 2006-02-02 11:50:57

Re: hal gone mad! [Solved]

#15 2006-02-02 13:23:29

Re: hal gone mad! [Solved]

Board footer