• RandomGuy (unregistered)

    I AM FRIST ENOUGH!

  • AV (unregistered)

    Ahm...

    If the new DEV server was so powerful, why not run the DB in a seperate VM on that same physical machine?

  • Paulie (unregistered)

    I'd gather evidence and shop this unprofessional idiot to senior management ASAP. Either that or punch him in the face :)

  • (cs) in reply to AV

    In that situation I would come into work late at night, grab a hard disk image of the machine hosting MITCH-DB (all 4.3GB of it, most likely), archive that, duplicate it, P2V it onto the shiny new server, then DBAN the original box to make sure it can't boot up and cause an IP address conflict with its virtual clone.

    Arseholes like Mitch are organizational damage that needs to be routed around.

  • Bob (unregistered)
    “Son of a…” Eric muttered
    Dammit, another unfinished article!

    Son of a Mitch?

    Son of a president's daughter? No, that couldn't be, no "pull".

    TRWTF is places where you need "pull" to get things done.

  • eVil (unregistered)

    I never understand it when people are angry about having to do their job. IT, or service delivery people who get annoyed at having to do slightly different work that they would otherwise be doing. I mean, its understandable if they get delayed from going home, but if you're just going to be there anyway, whats the issue?

    Lets say you're going home at 5:30, does it really matter if you're doing X for colleague A, or Y for colleague B? Either way, you're doing something tedious for some tedious cunt, until such time as you can do your own tedious thing. Why be angry about one and not the other?

  • Smouch (unregistered)

    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it MSSQL, Sybase, MySQL, or even Oracle (if you can get it installed).

    The reality is that most databases are simply designed incorrectly, and the apps which query them are also crap.

    So, the performance bottleneck, is, as usual, the user - or in this case the developer, and not the hardware.

  • eVil (unregistered)

    Also....

    I would email Mitch and be very courteous, making sure to CC in any relevant bosses, to ensure a consistently professional paper trail.

    At the same time, I would visit him personally, out of earshot of any colleagues, and ask him: "Oi cuntsocks, where is all that fucking hardware I demanded off you last week, you abject douche-master?"

  • Anom nom nom (unregistered) in reply to AV
    AV:
    Ahm...

    If the new DEV server was so powerful, why not run the DB in a seperate VM on that same physical machine?

    this ˆ

  • Cyclops (unregistered) in reply to eVil

    This is the correct answer.

    +10

  • Cyclops (unregistered) in reply to eVil
    eVil:
    Also....

    I would email Mitch and be very courteous, making sure to CC in any relevant bosses, to ensure a consistently professional paper trail.

    At the same time, I would visit him personally, out of earshot of any colleagues, and ask him: "Oi cuntsocks, where is all that fucking hardware I demanded off you last week, you abject douche-master?"

    No, dammit this is the correct answer +10

  • Tragedian (unregistered) in reply to Smouch
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it MSSQL, Sybase, MySQL, or even Oracle (if you can get it installed).

    The reality is that most databases are simply designed incorrectly, and the apps which query them are also crap.

    So, the performance bottleneck, is, as usual, the user - or in this case the developer, and not the hardware.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

  • faoileag (unregistered) in reply to eVil
    eVil:
    and ask him: "Oi cuntsocks, where is all that fucking hardware I demanded off you last week, you abject douche-master?"
    "When his colleagues heard the chilling scream, they did what they had done numerous times before: they dialled 999 and called an ambulance. When the paramedics arrived on the scene some 12 minutes later, the scream had stopped. 'Yuck!', one of them said. 'Na, this is nothin!', the older paramedic replied. 'You shoulda have seen what that hardware guru from <company_name_withheld> over in ...street did to one of the company's junior developer. Had it not been for the President's daughter stepping in, he wouldn't be the happy father he is today.' 'Still, a somewhat suspicious workplace accident, if you ask me.' The older paramedic felt something being stuffed into side pocket. 'But definitely just a workplace accident, fans occassionally break and the blades go wild. No need to get the cops in. See ya, man!', he nodded to Mitch, as they picked up the poor developer and left for the hospital. Meanwhile Paula in the Java development team..."
  • (cs) in reply to Paulie
    Paulie:
    I'd gather evidence and shop this unprofessional idiot to senior management ASAP. Either that or punch him in the face :)
    Wait. Why do you assume these are mutually exclusive alternative courses of action? Why can't I choose both?

    And it should be something with a bit more authority than a fist, like the GAU-8 you keep in your back pocket. You do keep one there, don't you?

  • (cs)

    I once had to deal with a designer a bit like this. I was working in a large insurance corporation doing front-end stuff. Every time a task came in with design work, I would ask which designer to get involved, and hope that it wasn't John. It usually was.

    If I had to go to John with design work, he would always complain to me about the task, as though I had written the spec, and it was all my fault.

    I was once half listening to him complaining about a task which I had been asked to involve him in, which had come back to me because his design was not compliant with the spec (somehow this was my fault), while I was doing some work (probably updating Lotus Notes, something John never seemed to to) and gave a vague response along the lines of "Yes John, I'm sorry you're upset, John". His counter-response was to mimic me in a high-pitched voice.

    Rather than leap across the desk divider between us and throttle him, banging his head against the desk and ultimately wearing his bloody skull as a hat, yelling "I am the King!" and being escorted out by security, I got up and made myself a tea. I drank a lot of tea while working with John.

    While I never technically resorted to or even threatened violence, there was one case where he was using my computer to use a messenger program for work purposes (I guess his wasn't working, but I forget the why of this). He decided to enact the Machine Lock Fairy Protocol, which was put in place specifically to avoid people forgetting to lock their machines for security reasons. This protocol involved sending a message to others (usually something juvenile and unoriginal like "I like boys") from the unlocked computer so that others were aware that the person had left their machine unlocked.

    Having spotted him enacting this protocol, despite giving him - a (supposedly) trustworthy party - access to my machine, I decided to enact the Fairy Interrupt Ambush Protocol. My variation on this was to sneak up behind him and quietly reach under his chin for the other side of his jaw, with my other hand on the back of his head, and apply just a little pressure, giving the pretense that I was going to break his neck. Bear in mind, he was a rugby player and bigger than me, so I didn't expect him to complain of neck pains (he got better - I think the sudden contact made him tense his neck muscles).

    So that resulted in a conversation like this: John: "My neck hurts." Me: "My chat history hurts." John: "Yeah well that's what you get." Me: "For trusting you, apparently."

    Later he would say he didn't know what he had done to make me dislike him.

  • faoileag (unregistered) in reply to Tragedian
    Tragedian:
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it Sybase, MySQL, Oracle or even MSSQL.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

    There. FTFY.

  • (cs)

    I've found acting as though they say everything in a friendly way is a great way to do it since it confuses the hell out of a lot of people who think being idiots like Mitch is a good way to go.

  • corroded (unregistered) in reply to Paulie
    Paulie:
    I'd gather evidence and shop this unprofessional idiot to senior management ASAP. Either that or punch him in the face :)

    Why not do both, it's not mutually exclusive and far more satisfying.

  • Aris (unregistered) in reply to faoileag
    faoileag:
    Tragedian:
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it Sybase, MySQL, Oracle or even MSSQL.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

    There. FTFY.
    Most RDBMS fast enough, probably, but in line with 2000-ish performances and latency. As a exercice, just remember how fast google was in 2000 and how fast it is now. It's ridiculous to expect fair performances from a desktop that's probably 100x slower than a consummer desktop from today, especially in term of latency. And latency is what makes the application "look slow", not the throughput.

  • ZoomST (unregistered)

    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?

    I think lately the histories are becoming very obscure... like the origin of the sickness of the president's daughter. Please TDWTF, try to fix this in the future.

  • Smug Unix User (unregistered)

    Just point your app to use his current machine. Problem solved?

  • long johnson (unregistered)

    Hardware Mitchmatch

  • Pock Suppet (unregistered) in reply to faoileag
    faoileag:
    Tragedian:
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it Sybase, MySQL, Oracle or even MSSQL.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

    There. FTFY.
    I heard Hitler secretly funded the development of MSSQL. </Godwin>

  • Patrick (unregistered)

    If this is true, I feel bad for you son. I got 99 problems but Mitch ain't one.

  • faoileag (unregistered) in reply to Aris
    Aris:
    faoileag:
    Tragedian:
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it Sybase, MySQL, Oracle or even MSSQL.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

    There. FTFY.
    Most RDBMS fast enough, probably, but in line with 2000-ish performances and latency. As a exercice, just remember how fast google was in 2000 and how fast it is now. It's ridiculous to expect fair performances from a desktop that's probably 100x slower than a consummer desktop from today, especially in term of latency. And latency is what makes the application "look slow", not the throughput.
    Wrong reply - you failed to start a flamewar ;-)

  • faoileag (unregistered) in reply to Pock Suppet
    Pock Suppet:
    faoileag:
    Tragedian:
    Smouch:
    Fact is most hardware from the year 2000 should be plenty fast enough to serve most RDBMSs, be it Sybase, MySQL, Oracle or even MSSQL.

    Good trolling attempt, but you need to find some way to work a religious debate in there.

    There. FTFY.
    I heard Hitler secretly funded the development of MSSQL. </Godwin>
    That's cheating, you did it on purpose! ;-)

  • ¯\(°_o)/¯ I DUNNO LOL (unregistered) in reply to flabdablet
    flabdablet:
    In that situation I would come into work late at night, grab a hard disk image of the machine hosting MITCH-DB (all 4.3GB of it, most likely), archive that, duplicate it, P2V it onto the shiny new server, then DBAN the original box to make sure it can't boot up and cause an IP address conflict with its virtual clone.
    You're not thinking far enough ahead here. What you need to do is, after duplicating its hard drive, find some way to cause the hard drive to fail in such a way that it appears to have died of natural causes. After all, I think we can be sure there isn't likely to be a good backup of it. If there is, it's from a tape drive that can't read back anything it's written, but nobody has ever read-tested the backups to find that out.

    In fact, given the age of the machine, it's possible that you could find a dead click click hard drive of the same vintage in the scrap pile, and Mitch wouldn't even notice that it was a different brand.

  • Rob (unregistered) in reply to Smouch

    I agree with what you say, that a properly designed database should be fast enough on older hardware.

    But the wall that developers hit is when the database schema is poorly done and the application queries are also poorly done. The developers can spend considerable time re-factoring the application or they can request a faster server. In most situations, it is more effective going with the faster hardware solution.

  • ZoomST (unregistered) in reply to long johnson
    long johnson:
    Hardware Mitchmatch
    +10 Make this^^^ featured comment pleeeeaaaseee!

    Captcha: saluto, as in: "Io saluto il tuo talento" (Google translatable!)

  • Chad (unregistered)

    You gotta love those types of people. Especially the ones that can't speak english and get mad at you. I laughed in a chinese guys face when he was stuttering and trying to spit out angry responses that made sense when I asked him to DO HIS JOB.

  • (cs) in reply to ZoomST
    ZoomST:
    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?

    I think lately the histories are becoming very obscure... like the origin of the sickness of the president's daughter. Please TDWTF, try to fix this in the future.

    You don't think that having to deal with a half-wit who has screamy-shouty psychotic episodes when you have hardware issues is not sufficient WTF?

    At one workplace a group of us had a length of two-by-four that wandered between our desks. It was generally referred to as "The Persuader" or "The Enforcer". Later on, someone who had access to a lathe turned a handle-sized length of one end to make it easy to hold. My comment about a need for something like nails (sticking out, of course) was heeded, although short lengths of 5mm pine rod were used instead of actual nails for workplace safety reasons. It was usually left lying suggestively on a desk, or it appeared in conversations like, "Where's The Persuader? I need to talk to ${IDIOT_CTO}."

  • Chad Garrett (unregistered)

    Seems likely that this Mitch got caught red-handed with his used hardware sales business on the side. All decommissioned servers went straight to "recycling." Except that apparently no one actually figured it out when he was exposed.

  • faoileag (unregistered) in reply to Steve The Cynic
    Steve The Cynic:
    ZoomST:
    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?

    I think lately the histories are becoming very obscure... like the origin of the sickness of the president's daughter. Please TDWTF, try to fix this in the future.

    At one workplace a group of us had a length of two-by-four that wandered between our desks. It was generally referred to as "The Persuader" or "The Enforcer".
    There is nothing really new in that idea, it has been around a long time. In spouse-parlance it's called a "rolling pin".

  • ZoomST (unregistered) in reply to Steve The Cynic
    Steve The Cynic:
    ZoomST:
    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?
    You don't think that having to deal with a half-wit who has screamy-shouty psychotic episodes when you have hardware issues is not sufficient WTF? [...]
    I meant the technical WTF, if there is one. Anyway, MITCH-DB sounds like a 1987's did-it-by-myself "database engine" (pay attention to the quotes), done by one of these reinvent-the-wheel stubborn guys. Like creating BOBX as better than PHP.
  • (cs) in reply to ¯\(°_o)/¯ I DUNNO LOL
    ¯\(°_o)/¯ I DUNNO LOL:
    <snip>You're not thinking far enough ahead here. What you need to do is, after duplicating its hard drive, find some way to cause the hard drive to fail in such a way that it appears to have died of natural causes. <snip>
    The proper tool to use in this situation is a rubber mallet, preferably while the drive is running.
  • faoileag (unregistered) in reply to RichP
    RichP:
    ¯\(°_o)/¯ I DUNNO LOL:
    <snip>find some way to cause the hard drive to fail in such a way that it appears to have died of natural causes. <snip>
    The proper tool to use in this situation is a rubber mallet, preferably while the drive is running.
    *sigh* Another of those proven, good ole tech tips that will become obsolete once solid-state disks are ubiquitous...
  • (cs)

    Seems like they missed an opportunity with the name of this particular article

  • dpm (unregistered) in reply to faoileag
    faoileag:
    Steve The Cynic:
    ZoomST:
    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?

    I think lately the histories are becoming very obscure... like the origin of the sickness of the president's daughter. Please TDWTF, try to fix this in the future.

    At one workplace a group of us had a length of two-by-four that wandered between our desks. It was generally referred to as "The Persuader" or "The Enforcer".
    There is nothing really new in that idea, it has been around a long time. In spouse-parlance it's called a "rolling pin".
    In the Scary Devil Monastery, it's called a LART. The meaning of the acronym is left as an exercise for the admin.

  • (cs) in reply to ZoomST
    ZoomST:
    I meant the technical WTF, if there is one. Anyway, MITCH-DB sounds like a 1987's did-it-by-myself "database engine" (pay attention to the quotes), done by one of these reinvent-the-wheel stubborn guys. Like creating BOBX as better than PHP.
    No particular need for a technical WTF. In fact, interpersonal WTFs have a long history of, um, success on TDWTF.

    And MITCH-DB could just as easily be the otherwise normal database instance that the borderline psycho created back in 2000 as a scratch experimenting zone that just sort of "growed", as it were.

  • (cs) in reply to Bob
    Bob:
    “Son of a…” Eric muttered
    Dammit, another unfinished article!

    Son of a Mitch?

    Son of a president's daughter? No, that couldn't be, no "pull".

    TRWTF is places where you need "pull" to get things done.

    <lily eriksen>Youuuu SONOFAMITCH!</lily eriksen>

  • (cs) in reply to faoileag
    faoileag:
    There is nothing really new in that idea, it has been around a long time. In spouse-parlance it's called a "rolling pin".
    Of course there's nothing new about the idea. Even when we had it (and that workplace died the late-casualty-of-dot-com death in 2003) it wasn't new.

    (I say "late casualty of dot-com", but in reality it was a victim of internal politics and repeated mass sackings of the sales/marketing side of the business. Oddly enough, after you sack all the salesmen and marketeers, you don't sell much product. It is another instance in my career of a company that would/could not sell the product, even though we, the engineers, built something good/better.)

  • (cs) in reply to faoileag
    faoileag:
    RichP:
    ¯\(°_o)/¯ I DUNNO LOL:
    <snip>find some way to cause the hard drive to fail in such a way that it appears to have died of natural causes. <snip>
    The proper tool to use in this situation is a rubber mallet, preferably while the drive is running.
    *sigh* Another of those proven, good ole tech tips that will become obsolete once solid-state disks are ubiquitous...
    ... replaced by the firmware hack to turn off load-leveling.
  • (cs)

    The end of this story would have been even better if the "new" server actually was their db server... at least that was what I expected to happen.

    Oh! And TRWTF is BizTalk.

  • Ozz (unregistered)

    I used to have to deal with people like that at a former employer. There was one in particular, the Finance Director, who got all bent out of shape when I told him the servers weren't powerful enough to do what he wanted.
    FD: "I thought these were shit-hot high-performance machines!"
    Me: "They were, when we purchased them 12 years ago. Today the combined processing power of all our servers across the country is less than your laptop." FD: "..."

  • Anon (unregistered) in reply to Shoreline
    Shoreline:
    (probably updating Lotus Notes, something John never seemed to to)

    Well, there's your problem right there. Using Lotus Notes is enough to make anybody grouchy.

  • Alex (unregistered) in reply to eVil

    It's about control and perception of roles.

    Mitch thinks he's in a management position, in charge of designing infrastructure and setting policy. He gets frustrated as new requests for hardware come in because they remind him that he's really just providing a "gofer" service; people want to just ask him for something and have him deliver.

    I felt this pain myself while "running" a corporate intranet. I wasn't in charge of anything, any design decision I would try to make was immediately thwarted by any random employee who asked for something different, no justification required. I would have to hop and all my plans would stall.

    Here I am, brain the size of a planet...

    CAPTCHA: conventio: Roman god of wearing blazers with nametags, carrying laptops and shaking hands.

  • Anonymoose (unregistered) in reply to faoileag

    Easy enough to work around: carry around a 4-pin molex to SATA power adapter that has the 5V and 12V lines reversed.

    -OR-

    Break off the contact plate on the hard drive's SATA power port.

    -OR-

    Jump the red (5V) and yellow (12V) wires while the drive is on using a bent paper clip.

    Low tech "hardware failure" solutions are never in short supply.

  • Sherman (unregistered) in reply to dpm
    dpm:
    faoileag:
    Steve The Cynic:
    ZoomST:
    Even reading it several times, I really don't get the WTF. Somebody please make a synopsis of this history... a clear one, I mean. Is... could it be that Mitch decommissioned the working development server to assign it to production, doing it on-the-spot?

    I think lately the histories are becoming very obscure... like the origin of the sickness of the president's daughter. Please TDWTF, try to fix this in the future.

    At one workplace a group of us had a length of two-by-four that wandered between our desks. It was generally referred to as "The Persuader" or "The Enforcer".
    There is nothing really new in that idea, it has been around a long time. In spouse-parlance it's called a "rolling pin".
    In the Scary Devil Monastery, it's called a LART. The meaning of the acronym is left as an exercise for the admin.
    Lumpy Fart? I hate those.

  • ¯\(°_o)/¯ I DUNNO LOL (unregistered) in reply to operagost
    operagost:
    faoileag:
    RichP:
    ¯\(°_o)/¯ I DUNNO LOL:
    <snip>find some way to cause the hard drive to fail in such a way that it appears to have died of natural causes. <snip>
    The proper tool to use in this situation is a rubber mallet, preferably while the drive is running.
    *sigh* Another of those proven, good ole tech tips that will become obsolete once solid-state disks are ubiquitous...
    ... replaced by the firmware hack to turn off load-leveling.
    Or a SATA version of one of these.
  • El Ka-Ben (unregistered)

    It's senseless to run a bunch of old desktops for development in an environment where you have have access to recently decommissioned proper hardware. Not just for speed, although waiting around for things to complete can sap productivity, but also because it is preferable to test on something resembling actual production hardware.

    I don't understand the pound foolish approach to servers (and network hardware) companies seem to have. It doesn't take long before a $10,000 server pays for itself in man-hours if 6,000 employees are waiting for it to respond dozens of times every day.

Leave a comment on “Hardware Mismatch”

Log In or post as a guest

Replying to comment #:

« Return to Article