Re: still several problems with Red Hat Linux 9

Ronald Cole schrieb: > Roland Wintgen <rw@evg.de> writes: > >>after i got a trial version of RDS 7.32UC1 for Linux and installed it >>on my new RH 9 Linux box I always get a core dump when trying to run >>r4gl -V. I installed this version along with the IIUG trial version of >>IDS 9.40UC2 and the newest CSDK 2.81UC2. >>I know that there has been something broken within the compatability >>of the RedHat glibc and the Informix database and tools. The database >>runs fine now but I cannot write any 4GL programs. An older RDS >>7.30UC5 runs fine but isn't capable of the new features built in with >>the new 9.40 version. > > > IBM now has a "patched" csdk-2.81UC2X5 that allows you to create > dynamic 4gl/esql executables. But -static is still broken on both > RHL9 and RHEL3.0beta2; so while I can finally continue development > after waiting *five* months for a "fix", I still cannot deliver > statically linked apps for my clients. IBM now says 2004Q1 for a 4gl > tool chain that works out of the box on RHL9 or RHEL3.0 (which I take > to mean "don't expect it before March 31st"). I guess that's pretty > telling/damning about IBM's Linux strategy for Informix and how > seriously they take their customers who *do* pay for support. > Is the "patched" CSDK officially available because the IBM still offers the "old" version? OK. I did some further research and the PDQ database says that there's a workaround for the RDS problem. Defect 162964 states: Apparently forcing the use of the libraries in /lib instead of /lib/tls solves the problem. In order to do that a possible solution is to define the variable: export LD_ASSUME_KERNEL=2.2.5 that changes the places where ld.so searches for the libs. Adding /lib to LD_LIBRARY_PATH doesn't work. Apparently this works the 4GL BUT with this variable set you won't be able to start up the engine since it fails silently. After bringing up the ADM processes nothing more happens and the engine is offline and no message is written to the online log. If you run onmode -uky it says that you should kill the database process and not to run onmode because the engine is in an undefined state. The workaround for this workaround is to start up the database engine as user root which works fine. I wonder what problems will arise when the RedHat releases the next version of the now called "Fedora" distribution. Has anyone tested the beta? -- Roland Wintgen (Systemadministrator) EVG Elektro-Vertriebs-Gesellschaft Martens GmbH & Co KG Trompeterallee 244-246, D-41189 Moenchengladbach Tel. +49 21 66 / 55 08 23, Fax +49 21 66 / 55 08 90 www.evg.de rw@evg.de

Roland Wintgen <rw@evg.de> writes: > I wonder what problems will arise when the RedHat releases the next > version of the now called "Fedora" distribution. Has anyone tested the > beta? I'd suspect that IBM will not support Fedora distributions, and will decide to only support RHEL distributions from this point on. -- Forte International, P.O. Box 1412, Ridgecrest, CA 93556-1412 Ronald Cole <ronald@forte-intl.com> Phone: (760) 499-9142 President, CEO Fax: (760) 499-9152 My GPG fingerprint: C3AF 4BE9 BEA6 F1C2 B084 4A88 8851 E6C8 69E3 B00B