SIGBUS or SIGSEGV signals don't necessarily mean that you've run out of memory, only that you've attempted to access a memory address that isn't owned by your process (in the case of SIGSEGV). You've violated your (memory) segment boundaries. The bad news is that there's almost nothing you can check, given that it's from SQL and/or the underlying query parser or engine, and you didn't write it. You need to log a support call. IBM *may* be able to discern something from the core file, so don't discard it just yet.
----- Original Message ----- From: "Noah Hart" <[EMAIL PROTECTED]> Date: Fri, 24 Sep 2004 17:05:24 -0700 To: <[EMAIL PROTECTED]> Subject: [U2] Solaris SEGMENTATION FAULT > When I run SQL from the UNIDATA prompt, executing the following command > against a SQL mapped Unidata file, > I get either BUS ERROR or SEGMENTATION FAULT follwed by a core dump. > > > :SQL > sql> select COUNT(*) from lialogMEMBER_MV20S WHERE ID IN (SELECT RID > FROM LIA_LOG_SNAP); > > Bus Error - core dumped > > > Now, I'm not really a unix guy, but we are running Solaris 2.7 with > plenty of memory and disk. > > Any idea what to check? > > Thanks, > > Noah Hart > ------- > u2-users mailing list > [EMAIL PROTECTED] > To unsubscribe please visit http://listserver.u2ug.org/ ------- u2-users mailing list [EMAIL PROTECTED] To unsubscribe please visit http://listserver.u2ug.org/
