Ste Packaging

It took a day longer than I thought, but it’s all working now, and better than before. Enjoy!

-ste

FDCServers just emailed me and said:

“we`ll be adding new additional carriers to our network mix tonight around midnight CST. While we do not expect any downtime however there might be some packet loss and routing issues”

-ste

I realized, after yesterday’s failed attempt to move the log processing to a Xen Virtual Machine with an image file as its “disk”, that I needed to rebuild the base system that the virtual servers will run on. I’m working on that and will have the Cacti monitoring back up and running soon.

As for package updates, there will be more  this evening. I was up until 4:00 AM working on the above and didn’t get up until Noon. Since then I’ve been working on updating my resume, since I really do need to find a job ASAP. It occurred to me that managing this blog and repository is a job and needed to be added to my resume. Not only will it show I’m not sitting around idly, but that I am also learning new skills, and thus making myself more marketable (one hopes), in the process. Plus there was a bit of consulting I did for someone, earlier this month, that needed to be added as well.

-ste

I am moving the database and the scripts to another system and rebuilding the data. When I’m done, I’ll make the stats page visible again. It should only take a day or so.

Update: Scratch that idea. Apparently, using an image file for a Xen Virtual Machine’s disk is okay for certain things, but not for a database that you want to load 102 million log entries into. Talk about being insanely slow … I’ll have to allot a real disk partition to it, but not right now. Don’t mind me. I’m having fun. Sometimes you learn the most about something when you screw it up and have to redo it. :)

-ste

I have implemented server monitoring with Cacti. It’s pretty cool. When you go to the link below and login, you can see various graphs of information about the two servers that make up the repository. Clicking an item in the tree on the left will show you the corresponding graphs for that item, for both servers. You can then, if you wish, drill down a level by selecting one of the graphs, to see longer term views of the data. At the main page, you can also select other options from the bar at the top of the page.  There’s not a lot of data showing right now, because I just reset everything after my testing. Give it a day or two and there’ll be a lot more data and trends will show themselves. As time goes on, I’ll add other graphs as I think of other things that might be useful to know. It can even be easily viewed from your iPhone or iPod Touch, for when you just have to know if the site is up or down. Enjoy! I’ll add a permanent link to this, somewhere on the blog, shortly.

URL: http://65.78.67.213:8080/cacti/

Username: guest
Password: account

For the technically minded, this Cacti installation is running on a 64-bit CentOS 5.1 Virtual Machine, itself running on a 64-bit CentOS 5.1 host system, in my den. My next project will be to add another virtual machine to the box and migrate the processing of the lighttpd logs to it. That will let me shut down the 32-bit system that is currently processing them and move its 2GB RAM to the 64-bit system, giving it a total of 4GB of RAM.

-ste

P.S.: I’ll try to get a few more package updates out tomorrow morning and early afternoon, but then I’ll be gone the rest of the day. Monday morning I’ll be back at them, full swing.