• (cs) in reply to Jim
    Jim:
    I can well believe that some of these turkeys would shortcut to the point of hardcoding a pid because they just assume the same program will always get the same pid. Trust me, these people do exist (most of the ones I've met have grown up around C# - or at least windows)
    Even on Windows there's no link between a program (non-running) and the process ID you get when you run it. Which is not to say that there aren't people who believe that there is such a link, but there really isn't one. Not for C#, nor for Windows in general.

    There is a fixed pseudo-handle that can be used in process-related Win32 API calls to refer to the current process, but you shouldn't normally be calling them directly from C#, and it isn't a process ID either.

  • D (unregistered) in reply to Captain Oblivious
    Captain Oblivious:

    Curry isn't an Indian word. Curry is the Anglicized word for 'kari', which means 'sauce' in Tamil.

    and Tamil is Indian .

    Unless of course , you are a rebellious Tamil nationalist .

  • Martin (unregistered)

    The REAL WTF is the missing "process monitoring" script.

    There is monit, God, you can use upstart.

    But monitoring key business process by a boss and wait for his e-mail to manually start it? This is TRWTF!!

  • Daniel (unregistered) in reply to xaade
    xaade:
    You're assuming PIDs are randomly handed out. I've debugged enough to notice that given a regular startup, the processes I start in order tend to be within a somewhat predictable range of PIDs. Keep restarting to debug and I get PIDs that are rather close to each other. Is this not the case with the particular OS in the OP?
    PIDs are handed out sequentially on most (all? I have a nagging memory of some OS giving random (free) PIDs, but I can't recall it, and can't find it by trivial googling) OS types. See http://stackoverflow.com/questions/3446727/how-does-linux-determine-the-next-pid for the specific Linux implementation (though we don't know the exact system in this case).

    Over the course of a day, if the process is respawning somewhat often, the PID at a given time will for all intents and purposes be random. Over years as in this case: definitely. This is one of the fundamental pillars of probability theory, enabling us to use theoretically perfect random ("stochastic", really) variables to describe real life processes. So the "assumed random" limitation is not really a limitation.

  • Sam (unregistered)

    How much will management pay for a project to "stabilise" the platform? This is just the quality equivalent of the well known speed-up-loop, the "sleep" you take out to convince others that important development just happened.

  • Paul Neumann (unregistered) in reply to Daniel
    Daniel:
    PIDs are handed out sequentially on most (all? I have a nagging memory of some OS giving random (free) PIDs, but I can't recall it, and can't find it by trivial googling) OS types.
    Running a hardened kernel or in SE-Linux will cause the PIDS to be pulled from a pool (p)randomly. It also does fun things with memory allocation.

    Right or wrong, akismet thinks my comment is spam. Verto.

  • Captain Oblivious (unregistered) in reply to D
    D:
    Captain Oblivious:

    Curry isn't an Indian word. Curry is the Anglicized word for 'kari', which means 'sauce' in Tamil.

    and Tamil is Indian .

    Unless of course , you are a rebellious Tamil nationalist .

    Since when does 'kari' == 'curry'?

    It is exactly as I said.

  • (cs) in reply to Captain Oblivious
    Captain Oblivious:
    D:
    Captain Oblivious:

    Curry isn't an Indian word. Curry is the Anglicized word for 'kari', which means 'sauce' in Tamil.

    and Tamil is Indian .

    Unless of course , you are a rebellious Tamil nationalist .

    Since when does 'kari' == 'curry'?

    It is exactly as I said.

    I don't think that he was trying to refute he, he was trying to add something on without going into the history of the Tamil Tigers (not the ones from Detroit).

  • Essex Kitten (unregistered) in reply to shinobu

    Except the default is SIGTERM (that's what the kill that happens at 22:12 in the article sends).

    Captcha: wisi. I'm a wisi guy today.

  • foxyshadis (unregistered)

    I've worked with server platforms that spawn and recycle thousands of children a day. Haven't you guys ever worked with Apache httpd 1.3? It was only the dominant web server for fifteen years.

  • Jim (unregistered) in reply to da Doctah
    da Doctah:
    ohnao:
    Chad Garrett:
    Watch out for the difference between "turkey" and "Turkey"...

    And FYI, "turkey" is actually named after Turkey: http://en.wikipedia.org/wiki/Turkey_(bird)

    I thought it were the other way around
    Nope, it's a law of nature. Name the place first, then name the bird after it. See also "Canary Islands".
    oh dear, methinks someone didn;'t get a joke....does that happen here often?

  • Jim (unregistered) in reply to Steve The Cynic
    Steve The Cynic:
    Jim:
    I can well believe that some of these turkeys would shortcut to the point of hardcoding a pid because they just assume the same program will always get the same pid. Trust me, these people do exist (most of the ones I've met have grown up around C# - or at least windows)
    Even on Windows there's no link between a program (non-running) and the process ID you get when you run it. Which is not to say that there aren't people who believe that there is such a link, but there really isn't one. Not for C#, nor for Windows in general.

    There is a fixed pseudo-handle that can be used in process-related Win32 API calls to refer to the current process, but you shouldn't normally be calling them directly from C#, and it isn't a process ID either.

    Yah, sorry - doidn't mean to suggest that Windoze does correlate process to PID - just that most of the people that think that way seem to come from hack and slash backgrounds where they sort of work out that doing A causes B, but can;'t really grasp what is actually going on....

  • J (unregistered) in reply to eViLegion

    I figure he's calling it Turkish pizza because that's what the places that sell it call it, so you'll probably have to blame some Turkish-American person from the latter half of the 20th century rather than the submitter or Remy Porter.

  • nomen (unregistered) in reply to Steve The Cynic
    Steve The Cynic:
    Neil:
    Who uses 82×24?
    Remy Porter?
    Steve Orkaktani, Sysadmin of Taiwan Production, apparently does.

    TRWTF is, other than that incorrect cron, why the core application hanged, instead of shutdown, if the cause was a cron, like other mentioned, was probably not the culprit.

    Image-captcha-so-2003: nimis : Nimis (Slovene: Neme) is a town and comune of 2,817 located in the Italian province of Udine, about near to the border with Slovenia. It is situated at the foot of Mount Bernadia, home to a World War I Italian fort and a sweet white wine, Ramandolo.

  • Tzafrir Cohen (unregistered) in reply to Cronboy

    The author also forgot that there's a username field. Thus what this does is: at 22:12 each night run '21342' as the user called 'kill'.

    Also, as noted by others, given that the service was not interrupted, its PID could not have stayed 21342 for all of the kills (though killall or pkill, or whatever could have worked).

  • (cs) in reply to EuroGuy

    Maybe they're in the UK, which would make UTC local.

  • asdasczxasd (unregistered)

    12 22 * * * kill 21342

    That particular cron line will run a job at 22:12 (e.g. 10 in the evening, not at midday)

  • ion freeman (unregistered) in reply to Geoff

    Clearly, it was not the best solution for whatever problem it was solving, but I would suspect it was a developer who could drop unreviewed scripts into production, but was not allowed to touch production itself, and was trying to correct an earlier error that had resulted in a process that he was motivated to clean up quietly. I'm a developer myself, and I know we can't be trusted.

  • Norman Diamond (unregistered) in reply to Seahen
    Seahen:
    Maybe they're in the UK, which would make UTC local.
    Except when British Summer Time is in effect.
  • daily (unregistered)

    The real WTF is "Steve restarted the service before investigating"

  • Insensitive Claude (unregistered) in reply to daily
    daily:
    The real WTF is "Steve restarted the service before investigating"
    No. If it's a critical service, restore service to users first, worry about root cause second. The users are paying for service, not for you to do root cause analysis.
  • Felix (unregistered)

    Late to this party, but I feel compelled to point out that for some reason, nobody jumped on the other RWTF:

    When looking for the cause of a crash, one would scrutinize the syslog, which (among other things), keeps tabs on cron jobs. A job issuing a random kill should have stood out.

  • Tux "Tuxedo" Penguin (unregistered) in reply to Mr. DOS
    Mr. DOS:
    TRWTF is the screenshot of Mac OS X: a platform where cron has been deprecated in favour of launchd.

    Or he just ssh'd from mac into Linux server. Didn't think about that?

Leave a comment on “A Crony Joke”

Log In or post as a guest

Replying to comment #:

Log Out

« Return to Article