Re: Informix Bug: Fetch Last

Daniel Wright wrote: > > This particular bug could have caused one of our programmers to waste > several hours, verifiying that we had all the correct versions of > Informix installed in the proper order, and then contacting Tech > Support, only to have them admit (after at least an hour of inevitable > doubt on their part that we had proper versions) that there was a known > bug, which is corrected in a maintenance release, which will be issued > very soon (OK, they said within a month or 2 last Spring, but that was > on 7.21, and this machine was running 7.22!). > Shouldn't Informix's QA dept. include a full complement of FETCH's on > Scroll cursors before they release something? [I'm hoping you're > listening Jonathan Leffler, and any other Informix employees who lurk > this group.] > [ Why is it that it takes Informix Tech Support so much convincing that > there really is a problem?!?!? It seems that they believe they are > infallible and all their customers are idiots. I find it very > insulting, as well as a waste of time. If I can incontrovertibly > demonstrate that their database engine choked using their own > transaction log, that is not good enough. I must provide them with a > small program which they can run from the command line which repeatedly > reproduces the bug....oh yeah, in the best of all possible worlds too.] > Billable hours are very important to us, and unfortunately Informix Bugs > are NOT billable! Perhaps we should write this into our next sales > contract, that any time spent tracking down Informix bugs be chargable > back to Informix (Informix would still make money, but I imagine one of > our customers (who happens to have a corporate wide Oracle license, yet > still bought our software), would have received it for next to > nothing. (This was a different story, different bug, which I spent an > entire week isolating before Informix Tech Support would even look at, > and even then they bitched about the fact that my test case contained > too many lines of code...It was still their bug, not ours!) > -- > - Danny Wright > danwright@bigfoot.com > #include <stddisclaimers.h> > Idiot, n.: > A member of a large and powerful tribe whose influence in human > affairs has always been dominant and controlling. > - Ambrose Bierce, "The Devil's Dictionary" Wouldn't it be great if software companies accepted product liability for defects? Who would trade fewer features for a greater guarantee that they worked? I would. The answer to the waste of time problem seems to be to play dumb and lure the Informix Tech Support person into connecting to your system by modem and seeing the problem themselves. You do need to set up a playpen for them. Peter -- Peter Lancashire Information Systems Specialist, Bayer plc Eastern Way, Bury St Edmunds, Suffolk, IP32 7AH, UK Tel: +44-1635-562258, Fax: +44-1635-562281 Mail: Peter.Lancashire.PL1@bayer.co.uk My Internet plumbing does not allow me to mail and post news together. Sorry. All opinions are my own and not those of Bayer plc.