[cards]

I've read the post "Moving CATS to a new server..." http://www.opencats.org/forums/viewtopi ... lit=server, and it appears all that's needed is to backup the MySQL database and copy the attachments directory, get the new installation running, and then bring in the saved db and attachments.

I have an existing installation of CATS 0.8.0 that *was* working well. If I were to install 0.9.1 on a new server, would I be able to take a db backup and the attachments directory and bring them into the new installation, or does that not work across versions?

-Justin

[RussH]

Hi Justin,
I'd tend to copy across your existing installation, then run an upgrade on that to get it to 0.9.1 (or 0.9.1a... in the downloads section). If you imported your 0.8 data directly to 0.9.1x you may get unpredictable results.

Whilst you're here... any requests,comments for features etc?

[cards]

Sounds like good advice...pretty much what I expected to hear. So far, I've backed up the attachments folder and the db on the only server. The new server is running Fedora with apache/MySQL/PHP already up and running. Going through the INSTALL document in the cats-0.8.0/doc directory, I installed php-gd, Antiword, html2text, and UnRTF via yum. A yum search pdftotext gives no results. Any advice on getting that installed? I've also set up the cats symlink and chowned/chmodded as instructed. I haven't followed the MySQL instructions yet; will importing the db backup handle creating the cats db and user in MySQL? Once I get those issues figured out, I assume I should be able to run the web-based installer and things should be easy from there.

As far as requests/comments, I'm pretty much green to CATS all over again. I worked with it for a while about a year ago, but now haven't touched it in months. One idea that would probably help adoption though...why not make a CATS virtual appliance? Whenever I'm reviewing new tools to potentially use in our company, I love it when I can fire up a virtual appliance where it's already running so I can just play a bit. That really makes the decision a whole lot easier.

[cards]

To answer my own questions:

1. pdftotext was already available in Fedora, so no install required
   I indeed needed to create the cats database and user in MySQL before the old db could be imported

After I had the old installation (0.8.0) back up and running, I figured I may as well upgrade to 0.9.1. Everything seemed to be going well until I tried to access the login screen, and I get a request for my license key...ugh. Searching these forums gave me a key to use, found here: http://opencats.org/forums/viewtopic.ph ... e+key#p359

Now my new questions:

• I haven't done anything to enable Sphinx, but searches seem to be pretty quick. Is it auto-enabled somehow? How do I test if it's installed and working?
  What changes does 0.9.1a have? Should I upgrade from 0.9.1?

Thanks in advance...

-Justin

[RussH]

Sounds like good advice...pretty much what I expected to hear. So far, I've backed up the attachments folder and the db on the only server. The new server is running Fedora with apache/MySQL/PHP already up and running. Going through the INSTALL document in the cats-0.8.0/doc directory, I installed php-gd, Antiword, html2text, and UnRTF via yum. A yum search pdftotext gives no results. Any advice on getting that installed? I've also set up the cats symlink and chowned/chmodded as instructed. I haven't followed the MySQL instructions yet; will importing the db backup handle creating the cats db and user in MySQL? Once I get those issues figured out, I assume I should be able to run the web-based installer and things should be easy from there.

As far as requests/comments, I'm pretty much green to CATS all over again. I worked with it for a while about a year ago, but now haven't touched it in months. One idea that would probably help adoption though...why not make a CATS virtual appliance? Whenever I'm reviewing new tools to potentially use in our company, I love it when I can fire up a virtual appliance where it's already running so I can just play a bit. That really makes the decision a whole lot easier.

pdf2text - I've not had Fedora running for a while (currently running cats on centos) but I think it's included within the package "poppler-utils".
You'll need to create the db and the users in the new server before importing the backup. Personally, I just dumped the mysql tables from a command line rather than use the built-in facilities within cats.

Virtual appliances - yes, you'll find an attempt at one if you search rpath - but I could never get the dependencies ironed out!

You may or may not need to install sphinx as well - to accelerate the full-text indexing of any resume's (or C.V's) you've imported. There are instructions on the wiki - http://opencats.org/wiki/index.php?titl ... all_Sphinx
There are some advantages in installing the more recent version of sphinx rather than the default one - you can use boolean search operators etc, but it does take a little tweaking.

Anyway - welcome back!