Hi Maciej, I'll quote from an email our systems person sent to me:
*** I kinda solved this issue with a sledgehammer - I put 'MALLOC_CHECK_=0' on the kernel command line in the grub.conf file, as well as adding the line 'export MALLOC_CHECK_=0' to /etc/profile. Most likely, if there's a single script that starts the remedy services, you can just add 'export MALLOC_CHECK_=0' to the beginning of the script, and it'll take care of it. RedHat has info on this in the RHEL4 release notes: http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/release-notes/as-x86/ ... and also in this knowledgebase article: http://kbase.redhat.com/faq/FAQ_85_5217.shtm Note that that articles says to use a value of 1, but according to the release notes that causes a performance hit; so we use 0. *** If you need more info I would be glad to put you in touch with our systems person. I'm sure he would be glad to help you. Good luck sorting this one out. It was a major ordeal, setting the variable seems to have solved it. Dwayne Martin ---- Original message ---- >Date: Tue, 26 Jun 2007 06:28:07 -0700 >From: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> >Subject: Re: "glibc detected" error SOLVED >To: arslist@ARSLIST.ORG > >Dear Dwayne, >can you ask your system experts what are thouse kernel parameters? I >am experiencing similar problems with ARS7.0.1 on RHEL 4 + Oracle. >Regards, >Maciej Stepniewski, >CERN Remedy Team. > >On Jun 11, 6:36 pm, Dwayne Martin <[EMAIL PROTECTED]> wrote: >> I reported this problem as '"Asynchronous even occured" and "glibcdetected" >> when opening form' back in May 29. I thot I had solved it, and when I >> discovered that it wasn't fixed afterall, the title got so long that it was >> changed (I don't know who changed it) to "ARSList Question NOT SOLVED >> AFTERALL". >> >> The problem turned out to be a Red-Hat Linux issue. One of our Systems >> experts figured it out, and I'll simply pass on what he wrote to me: >> >> "There's a problem where double mallocs (double clearing of a memory >> allocation) causes a hard error in newer versions of RH Linux - the program >> crashes. The custom form in use here apparently does this somewhere in a >> function call. This is an issue we've dealt with (sort of) before and the >> systems guys just need to change a kernel paramater to cause a warning, >> versus error, to be generated when this occurs. This is the MALLOC >> parameter which, when set, causes a warning, versus hard error (crash) to be >> generated when a double clear occurs." >> >> Thanks to everyone who wrote in on this problem. >> >> Dwayne Martin >> Computing Support >> James Madison University >> >> _______________________________________________________________________________ >> UNSUBSCRIBE or access ARSlist Archives atwww.arslist.orgARSlist:"Where the >> Answers Are" > >_______________________________________________________________________________ >UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the >Answers Are" Dwayne Martin Computing Support James Madison University _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org ARSlist:"Where the Answers Are"
