Ubuntu upgrading: Sometimes reinstallation is necessary

Posted on

I got a awful vaccination against Ubuntu 11.04. I attempted an upgrade from Ubuntu 10.10 on an Acer Aspire One netbook. In the middle of the process I suspect the screen saver may have crashed (sundancer was that you?) and some other things also happened. I had to do a hard reset. This left me in a state were the system wouldn’t complete the boot. After some fsck.ext4 I could get to “mount -o remount,rw” and “sudo dpkg –configure -a” which seemed to go through alright. However, there was still something fishy. The smart harddisk characterization showed some bad things. It wasn’t clear to me whether this was a bad harddisk issue or something Linux and BIOS related, e.g., with ACPI. With some parameters changed for the booting (noacpi?) I got different error messages in my log file.

Among the error message I got was “failed command: READ FPDMA QUEUED” and “ata1.00 … status: { DRDY ERR } … error { UNC }”. From what I could gather my central package database (/var/lib/dpkg/status) was affected. “sudo cp /var/lib/dpkg/status-old /var/lib/dpkg/status” could not help as the two files seemed to be the same…

Perhaps it wasn’t an Ubuntu 11.04 fault but a harddisk error trigged by the upgrade. After several hours I came to the conclusion that it was a harddisk issue. At one point I bought a wrong harddisk, so I actually had one spare harddisk and could change the bad one. With that one in place I reinstalled Ubuntu 10.10.

Setting up the computer I ran into my classic multiple workspace/edge flip problem. I cannot find the configuration for Compiz and Metacity might have a “row” bug in the workspace switcher. Brightside seems to work for Metacity for edge flipping alright.

Do the frequency of computer errors ever convergence towards zero? It is interesting to watch the plot of Debian’s release-critical bugs through time. The Debian folks can squeeze the relevant release-critical bugs fairly low around the time of release, but otherwise the frequency of bugs tends to increase…

For my comfort Open Source blogger Peter Toft reports his sentiment about an Ubuntu 11.04 upgrade. He too has run into troubles, and the people commenting on his blog share sentiment.

An occational harddisk crash is good. All the rubish you accumulate gets wiped out and you are able to start on a fresh. Unfortunately, I was able to make a backup and did. So now I have the problem of restoring the backup to the new harddisk.


(Update 5 May 2011: For ppl searching on error messages here are a couple: “Current Pending Sector Count” was bad in the smart detector that I could run with a Live CD. “The disk drive for / is not ready or not present.” and “init: udevtrigger main process (453) terminated with status 1” were among the first error messages I could see.)

(Correction: 12 July: Added a fairly important ‘not’ in the sentence “sudo cp /var/lib/dpkg/status-old /var/lib/dpkg/status” could not help as the two files seemed to be the same)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s