• (cs)

    I'm a network dummy, so I hope nothing is lost in the translation...

    I came in early one morning and couldn't get anywhere on the network. I went into the computer room and our network guy was finishing up his change control scheduled for that morning. I told him the network was broken. He spent the next hour trying to fix/workaround the issue. He changed something and got the network functioning very, very slowly. I asked him what the issue was. He showed me the results of "show cdp neighbor" (which basically shows all the switches that are connected to the current switch). He said, "This should only be connected to one other switch, but show cdp neighbor is showing two switches." I notice the MAC address for the mystery switch is the same as the MAC address plastered on the switch with a big label. I thought he just taught me something, so I said "Ohhhh I get it, so the switch is plugged into itself and that's why it's showing it's own MAC address in the show cdp neighbor." He smacks himself in the head and unplugs a neon green cable that is plugged into two ports on the switch.

  • (cs) in reply to Zylon
    Zylon:
    WTF is a "frizbee"? A bee with really bad hair?

    "'Worse than failure' is a 'frizbee'?" ??

    speak sense, man! slap

    :P

  • (cs) in reply to -
    -:
    vertagano:
    Be thankful the tech was careless enough to trust a kid who had just caused serious network problems to have access to every machine on the network once again to fix them, and you only lost privileges to one folder.
    It probably was the same tech which previously agreed to installing the comunication program on all the PCs. So it's easy to know he was careless...

    Oh, he didn't agree...

    Just for some reason the students weren't locked out of adding programs to run at startup, and the public shared drive had execute permissions...

  • (cs) in reply to Ubersoldat

    I hope he doesn't get beer farmed!

    First one, he did get it to work. It just did goofy things.

    Second, why didn't he, instead, get a whole new big shiny server with back-ups and a security team for a screen saver?

    Third, there was documentation, but by the looks of it, he left out the part about the jpegs.

    Fourth, the Firefox speel checker can't do anything about "words" like jajajajaja (I doubt there will ever be any support for that in the future).

    Also, that is the longest captcha I've seen. I'm impressed that you actually deciphered it!

  • AC (unregistered) in reply to Carnildo
    Carnildo:
    AC:
    The real WTF here is that the developer WAY over-engineered his little screensaver application. Why he did make it so configurable when all the customer wanted was to display a static set of given images? He should have compiled the images into the executable and this never would have happened.

    It's one of those cases where over-engineering is easier. It takes all of about four lines of code to list the contents of a directory in Windows, while compiling non-bitmap images into an executable takes real work.

    So your excuse for doing it the wrong way is that it was easier? Figures.

  • (cs) in reply to AC
    AC:
    Carnildo:
    AC:
    The real WTF here is that the developer WAY over-engineered his little screensaver application. Why he did make it so configurable when all the customer wanted was to display a static set of given images? He should have compiled the images into the executable and this never would have happened.

    It's one of those cases where over-engineering is easier. It takes all of about four lines of code to list the contents of a directory in Windows, while compiling non-bitmap images into an executable takes real work.

    So your excuse for doing it the wrong way is that it was easier? Figures.

    What, you've never done anything the shortcut way ever? Do you fully comment every script you write, write a manual for it, get it signed off by your manager, check both the script and the manual into source/document control, and revise every six months for potential updates? You must be mighty slow responding when asked to do even the most minor administrative task.

  • (cs) in reply to amandahugginkiss
    What, you've never done anything the shortcut way ever? Do you fully comment every script you write, write a manual for it, get it signed off by your manager, check both the script and the manual into source/document control, and revise every six months for potential updates? You must be mighty slow responding when asked to do even the most minor administrative task.

    Is that the company's policy, to completely document and get management's approval of every script, no matter how insignificant? If so, then that's what I would be doing. I've never worked at a place, however, where that was the policy. Generally, there are guidelines as to how to scale requirements. There's a difference between slacking off and taking shortcuts, too. What we have here is someone pointing out that "It was easier that way" does not excuse bad work. It might explain it, but certainly does not excuse it.

  • (cs) in reply to SilverDirk
    SilverDirk:
    Er.... 1000 workstations on a token ring? like... a single token ring? Like, the token would be passed through 999 workstations before a comp would get to speak again?

    I have not worked woth a TR network for a long time, but I am under the impression that the maximum number of nodes on a single ring is limited to much less than 1000 - something closer to 250? The maximum number of MAUs is also limited per-ring.

    I do not remember how you would bridge two or more separate TR networks...

    Anyway...

  • (cs) in reply to dkf
    dkf:
    ... it was normal for the machines to have a load in the 5-15 range. The loop changed that. I believe the highest confirmed value for the load was on the order of 750, but sampling was so poor at that stage that who knows what the true load was?

    I remember when doing comp sci assignments on the sun cluster we had (4 servers, 30+ students across many rooms of xterms simultaneously doing work)... during the worst part of the times, it would be in the 80+ load range. Remarkably, the machines were (barely) usable - interactive commands were possible, though compiles and such naturally took forever.

    Seems under high loads, suns still remain fairly usable... I suspect around 750 or so it might be a bit sluggish. Heck, when I have fun occaisonally to forkbomb on Linux, the load usually ends up around 500 or so (after killing and doing an immediate uptime to find the load).

    BTW in bash - :(){:|:&);:

  • Arioch (unregistered) in reply to ebs2002
    ebs2002:
    So, is this because the deployment instructions were poorly written,

    Erghm... Deployment ? You mean, they should have deployed jpeg-pak with the program as single bundle? Then they just would DOS their LAN sooner :-)

    However why them did rebooted the servers ?

    PS: 3 mb... was it per-jpeg or was it all-jpegs.tar?

  • Arioch (unregistered) in reply to ebs2002
    ebs2002:
    So, is this because the deployment instructions were poorly written,

    Erghm... Deployment ? You mean, they should have deployed jpeg-pak with the program as single bundle? Then they just would DOS their LAN a bit sooner :-)

    However why them did rebooted the servers ?

    PS: 3 mb... was it per-jpeg or was it all-jpegs.tar?

  • Carl de Billy (unregistered)

    There's another WTF in this Olympic story... Celine Dion is from Canada (borned in Charlemagne, in the Québec province), not United States. So why did she contribute to the "American proudness" ?

  • steve uurtamo (unregistered) in reply to AC

    back in 1994 (it was circa 1994 or thereabouts, right?) (3MB * # of images) of of raw data compiled into a binary would have crippled the end-user's operating system. i remember when 8MB ram chips were expensive.

    you need those images on local disk, sure, but not in the binary.

    s.

Leave a comment on “slammer.SCR”

Log In or post as a guest

Replying to comment #:

Log Out

« Return to Article