Re: Error -406 all the time (SE7.1)

batonnet@zeta.org.au (Bryan Tonnet) wrote: :In <3286AB00.2906@msoft.alphanet.com.au> Pete McDonald <petem@msoft.alphanet.com.au> writes: :>Anyone upgraded from RDS4.1/SE5.0 to RDS6.0/SE7.1?? :>If the answer is yes, please help!!! Running under Sco Unix, our :>existing database systems run without a problem. Problems start to :>happen, however, when we try to re-compile and run them with RDS6.0 and :>SE7.1. :>We get no compile-time errors, but when we run them we either get a -406 :>error (Memory allocation failed) or simply a core dump (at various :>seemingly un-related points in the code). At the suggestion of Informix :>AUS, we have bumped up our ULIMIT, but to no avail. We've even :>re-created the databases with the new engine. :>Are we being really stupid and missing something obvious, or is there :>some magic trick we have to perform? :>Thanks in advance. :>Our company is having a problem transferring existing Informix 4gl-RDS :>systems from RDS4.1/SE5.0 to RDS6.0/SE7.1. :Gidday Pete :I have recently completed the same excercise on the same operatinf system :with the same release numbers of Informix RDS and SE. I found similar :problems to your self, and heres a list; :Random core dumps within the code : : I found that RUN RETURNING xxxx caused the problem. Disabling these : lines returned the code to normal operation (apart from the RUN :(). : Informix AUS told me it was a bug and I am awaiting a patch release. There have been nown bugs with the run returning construct (although I can't remember core dumps as one of them). For this reason there is a c-function available at www.iiug.org to replace the run statement. :Odd arithmetical errors : : Turned out to be the 'float' variable type. Changing to decimal(X,Y) : fixed the problems, Informix has no answer at this stage. There where some versions with aritmetic bugs on float datatypes. This should be fixed by some update. Don't remember the versions, but a search of c.d.i. on dejanews might give you some more info. :Odd arethmetical errors 2 : : RDS 6.x reports dont seem to default initialise declared variables in : the report the way 4.x did. This means you've got to do iit yourself : in FIRST PAGE HEADER, which makes for messy code if you already have : a PAGE HEADER block. Informix investigating, no answer yet. If this is a problem you don't need to define a first page header. We do this inside the page header to avoide the messy code (duplication of page header code): if pageno = 1 then ... do initialization on first page only ... end if ... rest of page header ... This is also needed for compatibility with 4GL compiled which do not initialize any variables. I don't know anything about the following problems. :INPUT ARRAY REQUIRED attribute : : INPUT ARRAY on fields with REQUIRED set dont work WITHOUT DEFAULTS. : The default value displays OK, but a simple <enter> to get out of the : field returns "A value is required in this field". Overtyping just : one character in the field satisfies it, and the whole default value : is returned to the input variable in the correct fashion. : Informix have called it a bug, and I await their patch with baited : breath. :RDS falls over with "myfuncfail - failed at 860 spipe.c". : Dont know, cant fix it, Informix has been told of the problem. No answer : at this stage. :Also, read the release notes carefully ( as I'm sure you already have ), :there are quite a lot of differences and fixes in the Toolset (much more :in the engine of course, but I dont think this is causing your heartache). :Good luck, please e-mail me if you have other questions, or you get any :good news that might ease my own situation. :SCO 5.0.2 :SE 7.12.UC1 :RDS 6.03.UC1 :Bryan Tonnet :batonnet@zeta.org.au : Nils.Myklebust@idg.no NM Data AS, P.O.Box 9090 Gronland, N-0133 Oslo, Norway My opinions are those of my company The Informix FAQ is at http://www.iiug.org