Re: reliability?
A user on IDS 10.00.UC5 (Linux) found that SELECT MIN(colno) FROM syscolumns WHERE colno>=1 AND colno>1 wrongly returned 1 — an optimizer bug mishandling redundant range predicates. IBM staff reproduced nothing on newer builds and confirmed it was already fixed: present in 9.40.xC8W2–9.40.xC9 (fixed in 9.40.xC9W1+) and 10.00.xC5–xC6 (fixed in 10.00.xC6W1+), absent from 11.10 onward. The poster's client had let maintenance lapse, so the rest of the thread is an unresolved debate about paying for fixpacks/upgrades rather than technical content.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
This problem exists on the following versions: 9.40: 9.40.xC8W2 through 9.40.xC8W7, 9.40.xC9, should be fixed by 9.40.xC9W1 and above. 10.00: 10.00.xC5, 10.00.xC6, should be fixed by 10.00.xC6W1 and above including 10.00.xC7 and 10.00.xC8. It does not exist in 11.10 and above. Bingjie Miao "Kevin Cherkauer" <invalid_address@nowhere.com> wrote in message news:fvt7jp$adn$1@aioe.org... > Bingjie Miao on the optimizer team tells me this bug was indeed fixed > already. He already sent a reply posting via a moderated feed, so it > should arrive here eventually. :-) > > -- > Kevin Cherkauer > Software Engineer > IBM Informix Dynamic Server -- Database Kernel > > > "Kevin Cherkauer" <invalid_address@nowhere.com> wrote in message > news:fvt3g8$c9o$1@aioe.org... >>I just tried this on recent builds of the next-to-be-released 10.00, >>11.10, and 11.50 versions (currently these are numbered 10.00.xC9, >>11.10.xC2W3, 11.50.xC2; of course these numbers might change and my >>posting this is no guarantee from IBM that these will necessarily be >>released...), and it worked correctly on all of them. All of these were on >>64-bit x86 Linux. The same query also worked against a regular table on >>all of them. (sysmaster tables are not normal tables -- they are what we >>call pseudotables. The code to access them produces them out of thin air, >>or rather out of the contents of memory on the fly, so it's possible for >>some things to be broken for particular sysmaster tables that work >>correctly for other sysmaster tables and normal tables.) >> >> This has the flavor of a bug that has already been found and fixed in >> later fixpacks. I will ask the optimizer team if they know of any such >> fix having been done. >> >> -- >> Kevin Cherkauer >> Software Engineer >> IBM Informix Dynamic Server -- Database Kernel >> >> >> "piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message >> news:fvsfda$4q7$1@pippin.nask.net.pl... >>> IBM Informix Dynamic Server Version 10.00.UC5 on redhat >>> >>> Try this: >>> SELECT MIN(colno) FROM syscolumns WHERE colno>=1 AND colno>1 >>> An the answer is ..... 1 >>> A know that stupid questions generates stupid answers but still.... >>> >>> Anyone has come to that? >> >> > >
one more short question: How to obtain fix? Sure - I can find it on fixcentral but as far as I know our client has no more active support... "Bingjie Miao" <nobody@nowhere.com> wrote in message news:fvteot$p16$1@aioe.org... > This problem exists on the following versions: > > 9.40: 9.40.xC8W2 through 9.40.xC8W7, 9.40.xC9, should be fixed by > 9.40.xC9W1 and above. > 10.00: 10.00.xC5, 10.00.xC6, should be fixed by 10.00.xC6W1 and above > including 10.00.xC7 and 10.00.xC8. > > It does not exist in 11.10 and above. > > Bingjie Miao > > > "Kevin Cherkauer" <invalid_address@nowhere.com> wrote in message > news:fvt7jp$adn$1@aioe.org... >> Bingjie Miao on the optimizer team tells me this bug was indeed fixed >> already. He already sent a reply posting via a moderated feed, so it >> should arrive here eventually. :-) >> >> -- >> Kevin Cherkauer >> Software Engineer >> IBM Informix Dynamic Server -- Database Kernel >> >> >> "Kevin Cherkauer" <invalid_address@nowhere.com> wrote in message >> news:fvt3g8$c9o$1@aioe.org... >>>I just tried this on recent builds of the next-to-be-released 10.00, >>>11.10, and 11.50 versions (currently these are numbered 10.00.xC9, >>>11.10.xC2W3, 11.50.xC2; of course these numbers might change and my >>>posting this is no guarantee from IBM that these will necessarily be >>>released...), and it worked correctly on all of them. All of these were >>>on 64-bit x86 Linux. The same query also worked against a regular table >>>on all of them. (sysmaster tables are not normal tables -- they are what >>>we call pseudotables. The code to access them produces them out of thin >>>air, or rather out of the contents of memory on the fly, so it's possible >>>for some things to be broken for particular sysmaster tables that work >>>correctly for other sysmaster tables and normal tables.) >>> >>> This has the flavor of a bug that has already been found and fixed in >>> later fixpacks. I will ask the optimizer team if they know of any such >>> fix having been done. >>> >>> -- >>> Kevin Cherkauer >>> Software Engineer >>> IBM Informix Dynamic Server -- Database Kernel >>> >>> >>> "piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message >>> news:fvsfda$4q7$1@pippin.nask.net.pl... >>>> IBM Informix Dynamic Server Version 10.00.UC5 on redhat >>>> >>>> Try this: >>>> SELECT MIN(colno) FROM syscolumns WHERE colno>=1 AND colno>1 >>>> An the answer is ..... 1 >>>> A know that stupid questions generates stupid answers but still.... >>>> >>>> Anyone has come to that? >>> >>> >> >>
"piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message news:fvufun$d14$1@pippin.nask.net.pl... > one more short question: > How to obtain fix? > Sure - I can find it on fixcentral but as far as I know our client has no > more active support... > In other words: is it true that we have to pay extra for any (even most critical) fix to IDS?? If so - this makes Informix an unique rdbms (or any other) software vendor... Especially taking into account IDS buggynes....
I'm not sure that makes IDS unique....i've seen other vendors do this. Typically they all copy each other! I thought IBM changed the way the s/w was licensed so that you didn't "buy" the product at all. Rather you pay maintenance...aka lease the s/w. I also thought that under these terms if you stopped paying maintenance you should stop using the s/w. It's been a long time since i've had to bother about such details. > -----Original Message----- > From: informix-list-bounces@iiug.org > [mailto:informix-list-bounces@iiug.org]On Behalf Of piotrek > Sent: Thursday, May 08, 2008 6:13 AM > To: informix-list@iiug.org > Subject: Re: reliability? > > > > "piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message > news:fvufun$d14$1@pippin.nask.net.pl... > > one more short question: > > How to obtain fix? > > Sure - I can find it on fixcentral but as far as I know our > client has no > > more active support... > > > > In other words: > is it true that we have to pay extra for any (even most > critical) fix to > IDS?? > If so - this makes Informix an unique rdbms (or any other) software > vendor... > Especially taking into account IDS buggynes.... > > > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > Notice of Confidentiality: **This E-mail and any of its attachments may contain Lincoln National Corporation proprietary information, which is privileged, confidential, or subject to copyright belonging to the Lincoln National Corporation family of companies. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. Thank You.**
> "Denham, Mark" <Mark.Denham@lfg.com> wrote in message > news:mailman.1072.1210253144.20610.informix-list@iiug.org... > I'm not sure that makes IDS unique....i've seen other vendors do this. > Typically they all copy each other! > I thought IBM changed the way the s/w was licensed so that you didn't > "buy" the product at all. Rather you pay maintenance...aka lease the s/w. > > I also thought that under these terms if you stopped paying maintenance > you should stop using the s/w. > It's been a long time since i've had to bother about such details. Happy you! We are working also with Oracle and MSSQL. Both have many week points but both give: - easy access to developement versions of engine - FREE and immediate and easy to find access to patches I always liked informix and promoted it in client talks. But now I have no more sentiment... > -----Original Message----- > From: informix-list-bounces@iiug.org > [mailto:informix-list-bounces@iiug.org]On Behalf Of piotrek > Sent: Thursday, May 08, 2008 6:13 AM > To: informix-list@iiug.org > Subject: Re: reliability? > > > > "piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message > news:fvufun$d14$1@pippin.nask.net.pl... > > one more short question: > > How to obtain fix? > > Sure - I can find it on fixcentral but as far as I know our > client has no > > more active support... > > > > In other words: > is it true that we have to pay extra for any (even most > critical) fix to > IDS?? > If so - this makes Informix an unique rdbms (or any other) software > vendor... > Especially taking into account IDS buggynes.... > > > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > Notice of Confidentiality: **This E-mail and any of its attachments may contain Lincoln National Corporation proprietary information, which is privileged, confidential, or subject to copyright belonging to the Lincoln National Corporation family of companies. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. Thank You.**
From: "piotrek" <piotrUSUN_T_O110@o2.pl> > - easy access to developement versions of engine > - FREE and immediate and easy to find access to patches link to IDS 11.1 Developer Edition (All features enabled, no time limit): http://www.ibm.com/developerworks/downloads/im/ids/dev.html?S_TACT=105AGX28&S_CMP=DLMAIN I followed the link labeled "Download the IDS Developer Edition" on the IDS 11 main page here: http://www-306.ibm.com/software/info/ids/index.jsp I don't know what to say about the FREE and immediate and easy to find access to patches. I'm not sure how the other guys do it but IDS doesn't do "patches", just new releases with the bug fixes included. If you have support it is not a problem to get these fixes from IBM's download site, but if you don't have support, are you really entitled to the bug fixes? Andrew
> > http://www.ibm.com/developerworks/downloads/im/ids/dev.html?S_TACT=105AGX28&S_CMP=DLMAIN > > I followed the link labeled "Download the IDS Developer Edition" on the > IDS 11 main page here: Ok, I give up - but it's a quite new thing > I don't know what to say about the FREE and immediate and easy to find > access to patches. I'm not sure how the other guys do it but IDS doesn't > do "patches", just new releases with the bug fixes included. If you have > support it is not a problem to get these fixes from IBM's download site, > but if you don't have support, are you really entitled to the bug fixes? > - If IDS does not have pathes (fixes) then what is this: http://www-1.ibm.com/support/docview.wss?uid=swg24009130 - Such a bug in optimiser is like faulty stearing mechanism in a car - you cannot trust an application build on something like this. Maybe 'other guys' are also not faultless but what is a problem with fixing (whithout extra charge) things which when not fixed makes a product unusable (what a kind of agrement is that)? In automotive industry they are even installing a path for you (product recall)
> Happy you! We are working also with Oracle and MSSQL. Both > have many week > points but both give: > - easy access to developement versions of engine > - FREE and immediate and easy to find access to patches > IDS has easy access to developer versions. I presume to get "free" patches you had to pay for the s/w up front. When you get patches do you also get new features/enhancements too? IDS has never supported a patch mechanism as such and you sometimes get new features/enhancements. I'm not sure you get free patches to Oracle on Unix platforms. Notice of Confidentiality: **This E-mail and any of its attachments may contain Lincoln National Corporation proprietary information, which is privileged, confidential, or subject to copyright belonging to the Lincoln National Corporation family of companies. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. Thank You.**
----- Original Message ----- From: "piotrek" <piotrUSUN_T_O110@o2.pl> Newsgroups: comp.databases.informix To: <informix-list@iiug.org> Sent: Thursday, May 08, 2008 9:43 AM Subject: Re: reliability? > > >> http://www.ibm.com/developerworks/downloads/im/ids/dev.html?S_TACT=105AGX28&S_CMP=DLMAIN >> >> I followed the link labeled "Download the IDS Developer Edition" on the >> IDS 11 main page here: > > Ok, I give up - but it's a quite new thing > >> I don't know what to say about the FREE and immediate and easy to find >> access to patches. I'm not sure how the other guys do it but IDS doesn't >> do "patches", just new releases with the bug fixes included. If you have >> support it is not a problem to get these fixes from IBM's download site, >> but if you don't have support, are you really entitled to the bug fixes? >> > > - If IDS does not have pathes (fixes) then what is this: > http://www-1.ibm.com/support/docview.wss?uid=swg24009130 > - Such a bug in optimiser is like faulty stearing mechanism in a car - you > cannot trust an application build on something like this. Maybe 'other > guys' > are also not faultless but what is a problem with fixing (whithout extra > charge) things which when not fixed makes a product unusable (what a kind > of > agrement is that)? > In automotive industry they are even installing a path for you (product > recall) > > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list Like Mark said IDS does not support a patch mechanism, the bugs and new features are rolled into the next release. To take advantage of a bug fix you must install a new version of IDS. It has always been this way, even before the IBM aquisition. As for as needing to be current with support to get these new releases, I kind of agree that this is a weird area. If you bought something and it has a defect, then shouldn't you be entitled to the fix for free? Maybe. But on the other hand, if you're running a system important enough to care about an optimizer bug, then shouldn't you have current vendor support for that system? Andrew
piotrek wrote: > >> "Denham, Mark" <Mark.Denham@lfg.com> wrote in message >> news:mailman.1072.1210253144.20610.informix-list@iiug.org... >> I'm not sure that makes IDS unique....i've seen other vendors do this. >> Typically they all copy each other! > >> I thought IBM changed the way the s/w was licensed so that you didn't >> "buy" the product at all. Rather you pay maintenance...aka lease the >> s/w. > I also thought that under these terms if you stopped paying >> maintenance you should stop using the s/w. >> It's been a long time since i've had to bother about such details. > > Happy you! We are working also with Oracle and MSSQL. Both have many > week points but both give: > - easy access to developement versions of engine > - FREE and immediate and easy to find access to patches > > I always liked informix and promoted it in client talks. But now I have > no more sentiment... > > >> -----Original Message----- >> From: informix-list-bounces@iiug.org >> [mailto:informix-list-bounces@iiug.org]On Behalf Of piotrek >> Sent: Thursday, May 08, 2008 6:13 AM >> To: informix-list@iiug.org >> Subject: Re: reliability? >> >> >> >> "piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message >> news:fvufun$d14$1@pippin.nask.net.pl... >> > one more short question: >> > How to obtain fix? >> > Sure - I can find it on fixcentral but as far as I know our >> client has no >> > more active support... >> > >> >> In other words: >> is it true that we have to pay extra for any (even most >> critical) fix to >> IDS?? >> If so - this makes Informix an unique rdbms (or any other) software >> vendor... >> Especially taking into account IDS buggynes.... >> >> >> _______________________________________________ >> Informix-list mailing list >> Informix-list@iiug.org >> http://www.iiug.org/mailman/listinfo/informix-list >> > > > > > Notice of Confidentiality: **This E-mail and any of its attachments may > contain > Lincoln National Corporation proprietary information, which is > privileged, confidential, > or subject to copyright belonging to the Lincoln National Corporation > family of > companies. This E-mail is intended solely for the use of the individual > or entity to > which it is addressed. If you are not the intended recipient of this > E-mail, you are > hereby notified that any dissemination, distribution, copying, or action > taken in > relation to the contents of and attachments to this E-mail is strictly > prohibited > and may be unlawful. If you have received this E-mail in error, please > notify the > sender immediately and permanently delete the original and any copy of > this E-mail > and any printout. Thank You.** > AFAIK: You can download Oracle XE and use it for production, but you don't get access to Metalink. You can download development version of the Enterprise, but you cannot use it for production or even to access your customer's production database. You can get Micro$oft SQL Server with one of the developer's programs, but you have to renew subscription every year or stop using software (and now pass some sort of exam on the web to continue using it - so I moved everything to Linux :-) ) All of the above are licensing restrictions, but not functional limitations. Informix support gives you access to the new versions and emergency support for production environment. HTH Michael
"Denham, Mark" <Mark.Denham@lfg.com> wrote in message news:mailman.1072.1210253144.20610.informix-list@iiug.org... I'm not sure that makes IDS unique....i've seen other vendors do this. Typically they all copy each other! >> I thought IBM changed the way the s/w was licensed so that you didn't >> "buy" the product at all. Rather you pay maintenance...aka lease the s/w. >> I also thought that under these terms if you stopped paying maintenance >> you should stop using the s/w. It's been a long time since i've had to >> bother about such details. Absolutely not. You buy from IBM (or hopefully from an Informix-specialist reseller!) a perpetual licence to use. This has bundled within it a non-negotiable 12-months' maintenance, which entitles you to support and any subsequent releases, bug fixes or patches. Once that 12-month period expires you may optionally renew the maintenance to retain the entitlement to support and to product updates. If you let that lapse then need a new release you have to pay to have the maintenance reinstated - this is charged at 3 times the maintenance cost. If the reinstatement charge were not punitive, everyone would let manitenance lapse and pay only when they had a problem.
>If you let that lapse then need a new release you have to pay to have the >maintenance reinstated - this is charged at 3 times the maintenance cost. >If the reinstatement charge were not punitive, everyone would let >manitenance lapse and pay only when they had a problem. > Exactly this is the case. And the client wants help from us since we made the appliction...
"piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message news:fvv9al$pft$1@pippin.nask.net.pl... > >If you let that lapse then need a new release you have to pay to have the > >maintenance reinstated - this is charged at 3 times the maintenance cost. > >If the reinstatement charge were not punitive, everyone would let > >manitenance lapse and pay only when they had a problem. >> > > Exactly this is the case. And the client wants help from us since we made > the appliction... OK. So your an OEM reseller of Informix? You bundled Informix in with your application and sold both to the client?
>In automotive industry they are even installing a path for you (product > recall) A great example....and IDS support works almost the same. To determine a problem with a component a track record has to be established. This means that many customers go to the dealer, or their favorite repair shop and pay for diagnosis/repair. Even if the part is covered by warranty you coud still have to pay for the labor. You will have to pay for both if your warranty has expired (an extended warranty is just like paying for support). Only when a recall is issued can the rest of the customers benefit - and that's where IDS falls short. Then again you don't get the equivalent of a model upgrade for each recall! Anyone who runs critical production systems without support is quite mad in my view. If they are a publicly traded US company then maybe they are falling fowl of federal regs too - the ones that came into play after 911. I would say that not being able to run a production system is just as bad as mind numbingly dumb as not having good backups to recover one! There's always the other option too....fix the app that is generating a daft query....using the free support it gets. Notice of Confidentiality: **This E-mail and any of its attachments may contain Lincoln National Corporation proprietary information, which is privileged, confidential, or subject to copyright belonging to the Lincoln National Corporation family of companies. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. Thank You.**
>Anyone who runs critical production systems without support is quite mad in >my view. If they are a publicly traded US company then maybe >they are >falling fowl of federal regs too - the ones that came into play after 911. >I would say that not being able to run a production system is just > as bad >as mind numbingly dumb as not having good backups to recover one! Take it easy. We are just an 'pretedning to be civilised' wild and remote small country :) No danger for any US citizens as far as I know. >There's always the other option too....fix the app that is generating a >daft query....using the free support it gets. The problem with the query is that it is partialy generated and not that trivial. The sysmasters tables was just an easy to reproduce example - in real the problem was with dates columns.
> Exactly this is the case. And the client wants help from us > since we made > the appliction... > Better jump to it then. Should have been able to fix it by now....failing that we could prepare an invasion force so long as you have oil. If not you are on your own. Notice of Confidentiality: **This E-mail and any of its attachments may contain Lincoln National Corporation proprietary information, which is privileged, confidential, or subject to copyright belonging to the Lincoln National Corporation family of companies. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout. Thank You.**
On 8 May, 11:12, "piotrek" <piotrUSUN_T_O...@o2.pl> wrote: > "piotrek" <piotrUSUN_T_O...@o2.pl> wrote in message > > news:fvufun$d14$1@pippin.nask.net.pl... > > > one more short question: > > How to obtain fix? > > Sure - I can find it on fixcentral but as far as I know our client has no > > more active support... > > In other words: > is it true that we have to pay extra for any (even most critical) fix to > IDS?? > If so - this makes Informix an unique rdbms (or any other) software > vendor... > Especially taking into account IDS buggynes.... How do you Larry Ellision has billions and can afford one of the biggest boats in the world? By paying people to create patches and then not charging for them? Believe me he is not making a loss on oracle...and neither are microsoft or db2 or sybase otherwise they would soon go out of business... Why it is we have these people who think all software development companies should be run at a loss? Fucking idiots...
> Why it is we have these people who think all software development > companies should be run at a loss? Fucking idiots... > sure, that's my point - especially as as I am software developer (and after work I am used to whip myself )... So according to your wisdom when buying a software you cannot relay on it (event on basic functionality like selecting right data from databse). Instead you have to pay for constant upgrades to new versions. And install them with new features and new bugs to find one day that 2+2=5. Ok - I will try to pomote this model in our company :) (but I think that's a privliege of huge enterprises) regards :)
"piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message news:g01gi5$7ob$1@pippin.nask.net.pl... > >> Why it is we have these people who think all software development >> companies should be run at a loss? Fucking idiots... >> > > sure, that's my point - especially as as I am software developer (and > after work I am used to whip myself )... > So according to your wisdom when buying a software you cannot relay on it > (event on basic functionality like selecting right data from databse). > Instead you have to pay for constant upgrades to new versions. > And install them with new features and new bugs to find one day that > 2+2=5. > Ok - I will try to pomote this model in our company :) > (but I think that's a privliege of huge enterprises) > regards :) To be fair, you will have ahad a year's maintenance on the software product you bought. Has it taken over a year to discover the fault?
> To be fair, you will have ahad a year's maintenance on the software product > you bought. Has it taken over a year to discover the fault? to be fair - are you all work in IBM??? In this model (no patches - only paid upgrade to new version) we could catch an 'broken' version on the last day of first year's maintenance. So it is not making a deal more fair at all. And once again - We just want basic funcionality to work as it should and as it worked before.
"piotrek" <piotrUSUN_T_O110@o2.pl> wrote in message news:g03an0$47i$1@atlantis.news.neostrada.pl... >> To be fair, you will have ahad a year's maintenance on the software > product >> you bought. Has it taken over a year to discover the fault? > to be fair - are you all work in IBM??? > In this model (no patches - only paid upgrade to new version) we could > catch > an 'broken' version on the last day of first year's maintenance. So it is > not making a deal more fair at all. > And once again - We just want basic funcionality to work as it should and > as it worked before. Well, I don't agree with you. It doesn't matter whether fixes are delivered by upgrades or patches. As someone else pointed out, you can't expect IBM or any other software vendor that wants to remain solvent to provide limitless R&D on the product for nothing. Product maintenance goes to support fixes and new and improved features. Your company has chosen, as is its perfect right, not to pay for product support. The downside is unfortunately ... that you don't get product support! Just my opinion though ...