Prev: Wherefore art thou Google repos?
Next: Scanner problem with Epson Perfection 2450 on 11.1
From: Marty on 26 Jul 2010 06:21
houghi wrote:

> mjt wrote:
>> So a person has wasted their time doing the upgrade
>> that failed - better to do a clean install the
>> first time and be done with it ... it takes about
>> the same amount of time
>
> What about the non-wasted time of the ones that worked? It takes me
> about a weekend to have a system up and running the same as the previous
> one and then I still notice programs I forgot and settings I did not
> change correctly.
>
> I now know where to look, but do you think everybody knows where to get
> the crontab information? Or what the exact settings where for his
> Apache? Moving his MySQL database? Reconfigure NFS?
>
> houghi

I agree, backups should work and a new installation once a year (or even
every 2 years) is a big ask. No question, new, clean installations are
better but they are not hassle free either and a good configuration takes
the best of one day or more depending on hardware issues. I done 2 upgrades
and 3 new installations on 5 machines, there was not problem with the
upgrades (except the one I am having now, which may not be related). The
upgrades were fast and everything worked afterwards, all server
configurations worked, just the way it should be. I was impressed. The new
installations incl configurations took much longer. All of the 3 new
installations had some hardware related issue. On 2 of the 5 machines, for
example nvidia driver didn't work at all, that cost a few hours to discover.

Meanwhile, in the absence of a quick fix I am moving all the data of the
drives to then format.
Thanks, cheers Marty

From: Marty on 26 Jul 2010 10:41
houghi wrote:

> Marty wrote:
>> Meanwhile, in the absence of a quick fix I am moving all the data of the
>> drives to then format.
>
> As you apparently can now correctly quote, here is what I did at a
> moment like yours.
> 1) Edit /etc/fstab so only / is mounted (hope that is not an affected
> partition) by adding a # (Octothorpe) as first character
> 2) boot into the system with `init 1`
> 3) Log in as root
> 4) fsck all extra partitions
> 5) edit /etc/fstab by removing the #
> 6) mount -a to mount everything
> 7) init 5 && exit
> 8) A reboot to see if all works as expected
>
> houghi

Did that and it worked, thanks a lot. Before I had deleted a number of
corrupt files I found while copying files to other locations. I am sure
those files caused the problem.

Anyway, life is worth living again and I even know now where to quote. Good
stuff.
Cheers Marty
From: JT on 26 Jul 2010 12:13
On 26/07/10 16:55, houghi wrote:
> Marty wrote:
>
>> Did that and it worked, thanks a lot. Before I had deleted a number of
>> corrupt files I found while copying files to other locations. I am sure
>> those files caused the problem.
>>
> This happend most likely with a (or several) hard reboots.
>
> houghi
>
Since a while (some years now), I use journaled filesystems (in my case
ext3). Since then: hard reboots never caused me any problems (like the
above, the annoyance stays the same ;-) )

--
Kind regards, JT

First  |  Prev  | 
Pages: 1 2
Prev: Wherefore art thou Google repos?
Next: Scanner problem with Epson Perfection 2450 on 11.1