October 6, 2004

Alternatively, You Could Actually Check

Filed under: digital — mhoye @ 11:15 am

Subject: [Bug 88400] De-selected package installed

Bug #88400
Summary: De-selected package installed
Product: Red Hat Linux
Version: 9
Component: anaconda
Status: CLOSED

Additional Comment #3 From Jeremy Katz on 2004-10-04 23:09 ——-

This should be better in newer releases.

… and the circle of life continues.


  1. Shit, I wish I could close all my bugs that easily. Maybe I’m doing it wrong.

    Comment by Alistair Young — October 6, 2004 @ 12:10 pm

  2. That’s the beauty of that line. You can close any bug with it.

    Comment by Mike Hoye — October 6, 2004 @ 1:33 pm

  3. Damn, that sounds suspiciously like the “I’ll transfer you to the manager…” *click* ploy by helpdesks.

    I woulda figured they would have at least said “RH9 is EOL, you’re SOL. Upgrade.”

    Comment by kev — October 6, 2004 @ 3:57 pm

  4. Or at least “which newer” and “how much better”.

    Comment by janice — October 6, 2004 @ 4:28 pm

  5. kev,

    I’ve worked at an international-level help desk before. I’m not proud of my time there. One thing I did learn, however, is that the only person more irritated by the hangup-during-transfer procedure than the caller is the agent trying to complete the transfer. At least under the system with which I had to work, transfers were oftentimes silently dropped, sometimes connecting the agent to a black hole, another call (!), or some kind of head-cracking source of full-volume static noise. If an attempted transfer was simply never picked by the intended recipient, there was no way to drop back to the call – the only possible outcome was to disconnect.

    It was /awful/. I felt like a rat fink every single time, and there wasn’t a damned thing I was able to do about it.

    Comment by Gnomon — October 7, 2004 @ 12:44 am

  6. Hey Gnomon,

    I hear ya. Been there, and deal with it on a regular basis. I work for a company that has a couple rather large helpdesks, and it is never level 0 or level 1, it’s almost always the mgmt queues that do it.

    They’re not recorded like the regular queues in most cases, and they’re usually not on the clock/counters. I work with some excellent desks, and some really crappy ones. The excellent desks unsurprisingly have high call times that the beancounters hate, the crappy ones have the best “performance” in the company, and probably the worst satisfaction rate (but it’s almost impossible to tell because of how we aggregate metrics).

    I regularly call in with various problems (billing, tech support, unrelated services, mgmt escalations, etc.) and it’s amazing how often it takes more than 3 transfers to get to the right place, and the IVRs get so complex that sometimes queue numbers change, and don’t get updated, and then blooie.

    It’s upsetting. We’re working on it, but it’s a slow slog, and you still get the jackwads that play the game to get rid of an upset customer or keep their call volume down. It catches up with them eventually, but at a scary cost to customer sat. I love 97% of the call centre folks I work with on a daily basis. It’s the other 3% that spoils it for the rest – just like anything else.

    I take my shot in humour, and hope I didn’t offend. My aim was squarely on RedHat, and I was shooting helpdesks in an effort to show how little I think of them.

    I had bought 3 subscriptions 2 weeks before they decided to go corporate. I understand their desire to switch to a different biz model, but taking my money and then saying “oh, btw, the product you just bought a subscription for is officially EOL in 45 days, and fuck you, we’re not giving refunds” really cheesed me off. I don’t like ’em much any more, which is sad, because they have been my fave server distro for years, and I have installe at least a couple hundred of them in production environments and always paid for a seperate CD set because I valued the product.

    Oh well, all good things…

    Go SuSE!

    Comment by kev — October 7, 2004 @ 11:41 pm

  7. Because I am contrary: In defense of the bug-closer, if they’ve done a major revamp of the system since you filed the bug, it’s vanishingly unlikely that the exact same bug is there. So if they’ve got a dozen symptom bugs related to the same root problem, they probably just go through and mark them all closed.

    (Which’ll sometimes be wrong, which is why programmers shouldn’t be allowed to close bugs, only mark them as ready for testing by QA. But this is open-source software we’re talking about, so you are the QA.)

    Comment by Mike Kozlowski — October 11, 2004 @ 11:47 pm

  8. Because I am contrary: In defense of the bug-closer, if they’ve done a major revamp of the system since you filed the bug, it’s vanishingly unlikely that the exact same bug is there.

    Why is that?

    Comment by Mike Hoye — October 12, 2004 @ 7:31 pm

  9. Because the weird interactions that produce rare-condition bugs tend not to survive a wholescale rip-and-rewrite. Totally new bugs get introduced, of course…

    Comment by Mike Kozlowski — October 13, 2004 @ 6:53 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress