Skip to content

Some days you just want to do nothing

Today I finally pushed out a new binary release of my slaughter server-automation tool. (Think "CFEngine-lite", with perl. full documetnation is available. Though nobody ever reads it.)

Otherwise the weekend is being quiet; we spent last night mostly drinking vodka, until midnight rolled over, and along with some messing around with a camera ("Wow, your arms are getting bigger!")

Today has consisted of a Turkish breakfast, an Indonesian dinner, and an ice-cream based tea.

I could write more, but I'm hung-over. A rare thing for me.

Comments On This Entry

  1. [gravitar] ThorstenS

    Hi Steve,
    I like slaughter very much and read your docu several times from beginning to end!

    Is there a diff to a previous guide?
    I have seen a new backup primivite - is there more I should know?

    Perhaps you could put the guide also into github…

    as always: thanks for your work!
    /thorsten

  2. [author] Steve Kemp

    The source for that guide is stored on github, so you can use standard git to look at revisions.

    The backup primitive? I think you might be confused - there is a FetchFile primitive which allows you to pull a file from your central store to the local machine - in the past if it updated the local file it would save a copy.

    For example pretend you wanted to pull /etc/sudoers from the central location. If the remote version matched the local one then there would be no change but if the central version differed from the local copy two things would happen:

    • The remote version would be moved to the local machine.
    • The previous version of that file would be saved with a .bak suffix. (e.g. In our case the new version would become /etc/sudoers as expected, but rather than being overwritten the previous version would be saved as /etc/sudoers.bak.

    The recent backup change? Allows to to prevent saving the .bak version. Which is handy if you're fetching a file that would have a problem with a backup version being present in the same location. (e.g. /etc/cron.d, or similar).

    As for more you should know? Hard to say. But if you have problems or suggestions I'm here to help.