Yes, it’s that time of year again … UPGRADE TIME!
With the release of Fedora 12 a few months ago, I’m taking the holiday break from work to get my servers up to a more current version of Linux.
The fact that Fedora 10, the version of Linux I am currently running, is no longer going to be updated added a little more incentive.
I just completed the 2nd of two upgrades of my Linux systems to Fedora 10.
I’m very worried.
What’s got me worried? Both upgrades when relatively smoothly.
Ginny has been bugging me about the problem she’s been having posting to her blog from the iPhone and other third party tools that she uses.
The problem has been narrowed down to a bug in LIBXML2 that causes leading angle brackets “<” to be stripped off of posts when processed through the XMLRPC support in PHP. I logged the issue in the WordPress bug tracker, and the consensus is that the problem started manifesting after an automatic update applied by Fedora 8.
Well, to solve the problem I’ve moved Ginny’s blog to a temporary server running in a base Fedora 8 install in a VMWare instance. Luckily WordPress stores most of it’s content in the MySQL database (not the images, unfortunately). A bit of NFS magic to mount the appropriate directory in the virtual system, and it’s working fine. The server itself only has Apache, PHP, MySQL client, SSH, and a few other bits & pieces to keep it running. No automatic updates, GUI, compilers, etc.
Once the bug is fixed (either WordPress, using a different XML processing mechanism, or LIBXML2) I’ll move her server back to the main server.
In the mean time, her blog is going to run in the virtual machine ‘Hobbittown’ (named because it’s small).
Yep, it happened again … about this time last year I was trying to upgrade my servers to Fedora Core 6 and ran into some problems.
Well, I decided it was time to upgrade to Fedora 8 … and, since I have time off, I figured this was a fine time to do it again.
Bad move.
Of course, in retrospect … there never would have been a good time to do the upgrade, based on the problems I encountered. At least I know my backup procedure is fairly good now.
I had been planning this upgrade for weeks … everything was set. In fact, the first half of the upgrade went smooth as silk. I upgraded the main web server (gondor) to Fedora 8 and it went pretty nicely. Only two issues, both of which were solved after a little research.
This gave me the confidence to proceed to upgrade Rivendell to Fedora 8.
I started the upgrade by booting from the CD so I could install Fedora from the DVD ISO image I had on a USB hard drive. Problem is, the system wouldn’t boot this way.
Well, I finally got my Linux systems upgraded … both Rivendell & Gondor got upgraded to Fedora Core 6 this weekend.
There were a few minor glitches … but nothing that couldn’t be handled.
One thing that was kind of annoying is the fact that the Apache config (httpd.conf) was replaced on both servers (the original was backed up to a ‘rpmsave’ file). This, in itself, wasn’t that big a deal … but I wasn’t expecting it.
I have an idea for a good open source project … that could be incorporated into new distribution releases … an “Upgrade Impact Analysis” tool. It would evaluate your existing configuration, compare to what is known about a new distribution, and tell you what config files could be used without modification and which files would have to be reworked.
Another problem I had, which had me worried, was the fact that I couldn’t get ClamAV to rebuild. After a bit of research, I found that my linker configuration file ‘ld.so.conf’ had a ‘/usr/i486-linux-libc5/lib’ listed. This has to be left over from one of my older Redhat installs. The files in that directory are timestamped April 15th 1999. To be honest, I’m shocked I was able to build anything with that old code in the linkers configuration. Luckily, simply commenting that line out fixed the problem.
So far everything seems to be working OK, so I’m cautiously optimistic.