• Leo T. (unregistered)

    Idiots will be idiots...

  • Anon (unregistered)

    This is never a problem in IT. I have tens of thousands of email and they're all read. That's why we have the "read all" button in Outlook.

    Problem solved

  • Brian Boorman (google)

    Were the "500 Server Error" messages when trying to post comments earlier today because thedailywtf ran out of disk space? Was Alex P fervently in the background deleting log files?

  • paulb (unregistered)

    Oddly, I have heard that there are other choices of things to do with logs outside of "script to auto delete them" and "delete them manually"...

  • skylarmt (github)

    TRWTF is not using logrotate or something similar.

  • Steven S. (unregistered)

    It must have been over a year since I submitted this story to vent my frustration. Too bad that it has been altered so much that it doesn't really come across all that well.

    The problem was not that the people in charge of the API kept too many logs or didn't read their email. The reason the platform went down was because they didn't have any server monitoring set up. Mind you that this is a professional web development company we're talking about. Every time it went down I had to find out during my own testing or because some demo or pilot project ground to a halt. And every time it was because disk space ran out. Whatever the reason may have been for it filling up in the first place.

    Now this in itself probably wouldn't have been a big deal for the app being described in the article, but we build something quite different. Our app was meant to aid in training people with mental disabilities (such as Down syndrome) to travel more independent. So we built a tracker that would compare the taken route to a predefined one and send alarms to caretakers if any possible danger presented itself. The app would just die if their API went down and they might get stranded somewhere without anyone knowing.

    I ended up just duplicating their entire platform. We just had to keep a connection with it for legal reasons.

  • Sole Purpose of VIsit (unregistered) in reply to Steven S.

    On the one hand, I agree with you. Your story has been distorted beyond all recognition, which is a pity. On the other hand, you have now joined the proud cohorts of IT guys who have been "Hanzoed" by Eric.

    Hey, it's not much. But we live in debased times.

  • Douglas Muth (google)

    On the subject of logging and log retention, isn't this what Splunk is for? :-)

    Ship the logs off to a Splunk cluster and then security can analyze the logs to their hearts' content.

  • Greg (unregistered) in reply to Steven S.

    I'm guessing it didn't happen in Belgium either as I very much doubt there is an app for us, Belgian citizens, to access our medical information. (And even if there were, it would take the administration months to react to a problem, see e.g. the current fipronil egg crisis.

  • Appalled (unregistered) in reply to Steven S.

    Ah. So I must remember to never submit an article unless I want to see it PBAR'd, Plagiarized Beyond All Recognition. Good to know, should I ever be tempted.

  • Steven S. (unregistered) in reply to Appalled

    Is it still considered plagiarism at that point? :p I do agree with you though. If I had known the story would be reduced to a bunch of placeholders in Eriks fantasy I wouldn't have bothered either. Still, it gave me a good laugh about the whole thing and it was a nice reminder of how much better my life became after I left that soul crushing place.

  • Derp (unregistered)

    Gern fantasy, didn't happen so I didn't read it.

  • I dunno LOL ¯\(°_o)/¯ (unregistered)

    I was expecting that the new app (or at least heavier usage resulting from it) would be responsible for the "attempts to break into the HCP".

    Disappointed.

  • Erik Story (unregistered)

    Poor Gern... Bring back Hanzo, Hanzo was great, trwtf is steven

  • Steven S. (unregistered) in reply to Erik Story

    The one in the story or actual me?

  • Erik Gern (unregistered)

    Steve S. is a liar. Erik Gern was the actual protagonist of this awesome story.

    And Hanzo was the shit.

  • Erik Story (unregistered) in reply to Steven S.

    the actual Steven S is the wtf because his wtf is sooo boring they needed to send the Big Guns Erik Gern so we all don't fall asleep

  • Zenith (unregistered)

    In Camille's defense, some places have a virtual torrent of e-mail flowing through every day. Hook up server monitoring tools to ServiceNow, for instance. Every time a server has to be rebooted, there's three e-mails for the reboot (ticket generated, ticket assigned, ticket resolved). But there's also three for processor/memory spikes as it tries to drive a stake through McAfee's heart, three for ping not responding, three for WMI not responding, and three for processor/memory spikes as McAfee rises from its grave to feast on the living once more. When you have so many routing rules that Outlook just won't take another, something is wrong. Server monitoring isn't a silver bullet; in fact, it can be so overbearing that actual issues are drowned out by noise.

  • Some Guy (unregistered)

    But logs switch over to new files regularly, like say, every day, right? (ISO-date + "_" + applicaton name + ".log" for example. You can now sort and binary search the files to get the one you want.) So why couldn't they just, you know... Move the files to a secure server? Security could then go over as many logfiles as their hearts would desire, and the disk space of the other servers wouldn't run out.

    No logfiles would be deleted, disk space would be available... Isn't this how you're supposed to do it, or am I the next guy who have made a stupid decision and will end up on TDWTF next to Paula Bean?

Leave a comment on “Disk Administrations”

Log In or post as a guest

Replying to comment #484246:

« Return to Article