Main Stories
Slash Boxes

Slash Open Source Project

Slashcode Log In

Log In

[ Create a new account ]

Article Poll

Poll I found this article to be
Very Helpful
Helpful
Not Helpful
Not Very Helpful
[ Results | Polls ]
Comments:0 | Votes:0
· Slash
· slashcode.com
· Slash
· cronolog
· apache
· slash
· More on Slash
· Also by Krow

Handling logging issues

posted by Krow on 01:33 AM April 22nd, 2003   Printer-friendly   Email story
Slash
I'm using cronolog for my apache logs, and I really, really like it. I'd like to be able to use it on the slash logs as well, which become large and cumbersome over time with many sites running on a server. How do you all handle your logs? What do you use for log rotation? How long do you keep logs? Is anyone using cronolog, or something like it with slash?
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

Handling logging issues 12 Comments More | Login /

 Full
 Abbreviated
 Hidden
More | Login
Loading... please wait.
  • I flush the logs every now and then using webalizer. My slash sites generate a lot of junk in the logs, like stupid messages about perl commands not terminating correctly and such. So, looking for any good information in the logs is like trying to find a needle in a haystack.

    I like webalizer cause it's free, and gives you some stats that slash software doesn't do.

    --

    --
    It's either on the beat or off the beat, it's that easy.
    • Because cronolog seperates the logs by date and time. This makes it much easier to find and archive things.
      --


      "How about you interface with my ass? By biting it!" --Bender
  • I just use logrotate. It comes standard with lots of distributions. Here's a snippet:

    /var/log/http/www.lottadot.com/access_ log {
        daily
        errors webmaster@lottadot.com
        postrotate
            /usr/bin/killall -HUP httpd 2> /dev/null || true
            DAY=$(date -d 'yesterday' '+%m_%d_%y')
            mv /var/log/http/www.lottadot.com/access_log.1 /var/log/http/www.lottado
    t.com/access_$DAY.log
        endscript

    }
    --
    lottadot [lottadot.com]
  • On a slightly different note, I've been checking my logs lately and in the past several weeks I've had to keep restarting slashd as it has been hanging during freshenup.pl (usually the last line in slashd.log is something like "Updating 03/04/22/1551213". Anyone else have this happen? It happens at least once a day now, I have had to schedule a cron job to restart slashd every hour. Slash 2.2.6

    I've been running the same code for about a year, the only thing I did recently was upgrade the kernel (RedHat):
    2.
    • What version of slash? We're not seeing that, but we're using the most recient T tag (I think it's the most recient anyway - T2_3_0_85).
      --


      "How about you interface with my ass? By biting it!" --Bender
      • slash 2.2.6 ... I haven't touched the code in about a year so that's why it's a bit mysterious. It's probably a sign I should start the move to the T tags.
        • I honnestly wouldn't reccomend it. We had no real choice, but if I had to do it all over again, I'd have waited for the upgrade stuff to be done. Seriously. It took us 6 days to migrate 15 sites, and much longer to iron out all the major bugs that resulted from the process - we're still killing minor ones.
          --


          "How about you interface with my ass? By biting it!" --Bender
          • Thanks for the tip, I'll definitely hold off, since it's not urgent. 2.2.6 has been extremely stable for the most part.

            I heard about this upgrade script a few weeks ago I think. I'll be looking for it!
            • I for one can't wait. :-) After what we went thru last time, I'll never do that again. The only reason we did it this time was because a new customer needed features only available in current CVS, so we upgraded. Going from CVS to the latest T tag wasn't so bad - it look less than an hour, and basically no one noticed. We still have some run away perl problems, Jamie was going to take a look at them, don't know if he ever did.
              --


              "How about you interface with my ass? By biting it!" --Bender
        • I know vlad's not recommending it, they had all *sorts* of problems doing it.

          But there are quite a few people who didn't have such problems that vlad and his crew did.

          I did my 4 or 5 sites, in roughly a week, tinkering with it in the evenings. Most of my time was spent meshing changes that I'd done to both the slash src and my plugins. Templates were a joy to do, but for each site it only took about a half hour to diff the template changes and reimplement them into the newly upgraded sites.

          Setup a second
          --
          lottadot [lottadot.com]
          • Thanks for your perspective and good to hear it went well for you. Approx. 1 week for you and Vlad's approx. 1 week really isn't too bad. I think it took me several months to initially get Slash running smoothly. This was version 1.0.4 and most major problems have been worked out now it seems.

            The transitions from 1.0x to 2.0 and also from 2.0 to 2.2 went well, thanks to the Slash crew's upgrade scripts. I definitely want to move up to the latest, it is just a question of when to pull the trigger and start