This discussion is archived
1 7 8 9 10 11 Previous Next 163 Replies Latest reply: Oct 22, 2009 3:23 PM by Hoek Go to original post RSS
  • 150. Re: Index rebuild
    6363 Guru
    Currently Being Moderated
    If you want to believe the majority of Oracle behavior is random and cannot be reproduced and if you face issues the best you can do is pay someone who says "Trust me. I am an expert" without question then fine.

    Most of us can do better.
  • 151. Re: Index rebuild
    706417 Explorer
    Currently Being Moderated
    I don't want anyone to believe that, silly! I would welcome the acceptance by most that Oracle is just software. Is it so hard to divorce the reality of being a DBA/Developer from the fanatsy of being a scientist? No, it isn't. So chin up, shoulders back, swallow those hot, stinging tears of disappointment and rebuild that index! There, feels better already, doesn't it?

    Love - Don Lewis
  • 152. Re: Index rebuild
    6363 Guru
    Currently Being Moderated
    Don Lewis wrote:
    I don't want anyone to believe that, silly! I would welcome the acceptance by most that Oracle is just software.
    That isn't what I said.
    If you want to believe the majority of Oracle behavior is random and cannot be reproduced
    Whereas I know the majority of Oracle behavior is consistent and can be reproduced.
    Is it so hard to divorce the reality of being a DBA/Developer from the fanatsy of being a scientist?
    I really don't care how you want to twist and label it, the majority of Oracle behavior is consistent and can be reproduced.
    No, it isn't. So chin up, shoulders back, swallow those hot, stinging tears of disappointment and rebuild that index!
    No because it doesn't disappoint me and it is more than likely not needed and a waste of effort.

    http://richardfoote.files.wordpress.com/2007/12/index-internals-rebuilding-the-truth.pdf

    Can you not run the examples?

    If you can run any of the examples in the presentation do you get the same results or not?

    If you don't get the same results do you know why?

    Do you just want everyone to rebuild indexes so you don't feel alone in wasting your time?

    Or are you still saying that the majority of Oracle behavior is random and cannot be reproduced and you need to pay one to say "trust me. I am en expert" to you?
  • 153. Re: Index rebuild
    706417 Explorer
    Currently Being Moderated
    Hi Pointless,

    +"Whereas I know the majority of Oracle behavior is consistent and can be reproduced."+
    How much, expressed as a percentage? Proof, if you please.

    +"I really don't care how you want to twist and label it, the majority of Oracle behavior is consistent and can be reproduced."+
    How much, expressed as a percentage? Proof, if you please.

    The "rebuild that index" thing is what we humans call humour. You have been caught out Tharg Zqilshm'Ping (of the Casseopieian gas planet Trumpus, which orbits star γ Cas) and your chief globule alien lord will, I fear, smite you for exposing your mission so early on.


    Live Long and Prosper,

    Don Lewis
  • 154. Re: Index rebuild
    6363 Guru
    Currently Being Moderated
    Don Lewis wrote:

    +"Whereas I know the majority of Oracle behavior is consistent and can be reproduced."+
    How much, expressed as a percentage? Proof, if you please.
    So you want me to prove that more than half the time Oracle functionality behaves the same on multiple databases regardless of environment rather than producing inconsistent results? I don't think a spreadsheet would sell if it had that reliability never mind an enterprise database.

    I could go through all the posts marked as correct that provided an example that was reproduced on another system. I don't think this is that surprising but you appear to think that more than half of them should fail or behave differently for reasons other than version dependency and fixed bugs. For tuning questions resultant plans are often different, but the reasons why are consistently identified and I can find no examples of the totally inexplicably different behavior you are asking me to believe in.

    But I won't because I can't be bothered with your definition of what should be proof when you can't even supply a single case of what you are talking about.

    >
    The "rebuild that index" thing is what we humans call humour. You have been caught out Tharg Zqilshm'Ping (of the Casseopieian gas planet Trumpus, which orbits star γ Cas) and your chief globule alien lord will, I fear, smite you for exposing your mission so early on.
    Oh.

    [Very funny|http://www.youtube.com/watch?v=IrNonqsWphU].
  • 155. Re: Index rebuild
    706417 Explorer
    Currently Being Moderated
    +"So you want me to prove that more than half the time Oracle functionality behaves the same on multiple databases regardless of environment rather than producing inconsistent results? I don't think a spreadsheet would sell if it had that reliability never mind an enterprise database."+

    Yes. I want you to prove that. You claimed it as a truth, so where is your evidence? OTN posts marked as correct are, obviously, not good enough. Or was your claim plucked from the air like anecdotal "evidence" you so despise?

    +"you appear to think that more than half of them should fail or behave differently for reasons other than version dependency and fixed bugs"+

    Research, Tharg, research! All along I have said, categorically, that inconsistent behaviour can happen as a result of bugs and other factors. I have not said, nor would I, that non-bug issues account for > 50% of such inconsistencies.

    +"But I won't because I can't be bothered with your definition of what should be proof"+

    I win! 10 points to me, I think. Can't be bothered? Can't prove it, more like. Go on, prove your claim. Prove that "the majority of Oracle behavior is consistent and can be reproduced", you naughty empiricist, you!

    Love - Don Lewis
  • 156. Re: Index rebuild
    user20335 Explorer
    Currently Being Moderated
    This is most entertaining for us neutrals - reminds me of the time the two philosophy professors came to blows over some abstruse theory. However, much as I enjoy the Pointless/DL show, it's not as good as the main act - DKB/JL.
  • 157. Re: Index rebuild
    661463 Newbie
    Currently Being Moderated
    Well, Pointless/DL kind of stole the show. This was a funny read.

    --
    Best,
    Johann

    "I may not agree with what you say but I will defend to the death your right to say it"
  • 158. Re: Index rebuild
    6363 Guru
    Currently Being Moderated
    Don Lewis wrote:

    OTN posts marked as correct are, obviously, not good enough.
    Why not? They demonstrate instances where Oracle behavior is consistent and reproducible among a wide range of Oracle installations for thousands of users in multiple countries.

    a) Prove that Santa Clause does not exist within a given percentage of certainty!
    b) No one has ever seen or ever produced any evidence of the existence of Santa Clause
    a) That is not proof, I win, I win, Santa Clause exists!

    (shuffles off stage pinning school debating society badge to anorak)
  • 159. Re: Index rebuild
    706417 Explorer
    Currently Being Moderated
    +"They demonstrate instances where Oracle behavior is consistent and reproducible among a wide range of Oracle installations for thousands of users in multiple countries."+

    Tharg, it's a fact, isn't it, that some poor souls will mark a post as correct, even if it were to later prove to be incorrect. So, posts marked as correct are not 100% reliable. That's why I said "not good enough".

    With 2 lifelines gone, you may phone a friend, if you wish. Do you have one?


    Regards - Don Lewis
  • 160. Re: Index rebuild
    user20335 Explorer
    Currently Being Moderated
    Johann.

    Not on topic, but you mentioned elsewhere that Mr Burleson is that the opinion that Sweden is communist. Where did he express that, and more importantly, at least for my understanding, why does he think that is true ? It clearly isn't.
  • 161. Re: Index rebuild
    Hoek Guru
    Currently Being Moderated
    Where did he express that
    See:
    Re: Path to become a DBA
  • 162. Re: Index rebuild
    tkyte Employee ACE
    Currently Being Moderated
    I fixed that - I had an issue in my procedure that text was indexing that cropped up during the move.

    I rebuilt the index and all is well...

    maybe I shouldn't have said that - someday someone might quote me as having said "he said he rebuilds indexes and all is well, just rebuild" :)
  • 163. Re: Index rebuild
    Hoek Guru
    Currently Being Moderated
    What a complete surprise: I never knew you posted here on OTN, Tom (on the other hand, I never even bothered to search on kyte over here, it just never crossed my mind).
    maybe I shouldn't have said that - someday someone might quote me as having said "he said he rebuilds indexes and all is well, just
    rebuild"
    :D

    Well, this is for sure the right thread to 'spread the word' ;)

    But, yes: all is 100% fixed. Thanks a lot for noticing and fixing so fast, despite OOW'09 etc. going on.
    All the favorite/dear threads I almost use daily (is that normal, I now wonder?) get selected as expected again.
1 7 8 9 10 11 Previous Next

Legend

  • Correct Answers - 10 points
  • Helpful Answers - 5 points