Excuses for scarce updates

I haven’t been updating much in the past week, so I’ll do a “blog” entry as an explanation.

Last Friday I flew from Newark to Columbus, and worked all day from the office there in central Ohio. We had a Poweredge 1955 blade’s motherboard die before it ever got to be in production, so that will be getting replaced next week.

I didn’t go back just to talk to tech support, though. On Saturday, we boxed up all of the equipment from the server room, put it in a U-Haul, along with everything else that wasn’t nailed down, and on Sunday my boss and I drove it up to NJ.

Since then, it’s been nonstop fun around the office. The HQ office in NJ (the one I work from) used to have a bare four post rack with just a couple of machines in it. Now it houses our backup stack, which I’m soon going to be converting to AMANDA. A blog entry will be forthcoming once it’s in place and I’ve played with it some more. I’ve found a lot more documentation about backing up with it than recovering from it. If anyone has any killer links to recovery tips, I’d love to see them.

Continuing, I’ve now got the entire contents of our old office to stuff into the already crowded server room.

I feel like I’ve worked a ton, even though yesterday was really the only late night. I think it was the not-having-a-weekend bit that did me in. This weekend I’m planning on traveling to the city to check out a museum. I haven’t decided between The Met or MOMA. My wife and I did the American Museum of Natural History last time we were in and loved it.

Also, just so there’s some sort of sysadmin-y thing in this post, for those of you using Reddit, there is a Sysadmin subreddit. It’s actually been in existence for a few months, but had 0 articles (and 4 members) until the beginning of last week. I’ve submitted 99% of the articles on there as I’ve come across them, so feel free to join it and start submitting.

How has your week been?

  • Bill

    MOMA’s…..interesting. Something you should definitely check out at least once.

  • augmentedfourth

    You sure you don’t want to use rdiff-backup? It’s working great for me, especially with the custom Perl wrapper script I wrote…

  • Matt


    It sounds great, but at this point in time, I’m so short on time that I’ve just got to get my implementation done. If you’ve got any links for rdiff, I’m open to new ideas, but the implementation has to be magic*

    * – magic is defined as fast, easy, and reliable

  • augmentedfourth

    Hey Matt,

    Sorry it took me so long to get back to you. If you’re doing tape backups, amanda is probably best. rdiff-backup is a disk-to-disk solution which uses librsync to keep reverse diffs of your data. The backup server functions like an rsync mirror, except that you can also use rdiff-backup tools to restore files from any previous backup state.

    It can be as easy as a cron job of “rdiff-backup /source server::/destination” (which, by default, uses SSH for transport), but I’ve built some custom configuration, including automatic rdiff removal after a custom amount of days for each backed-up directory, in my script.