You can detect the cause of system/Sybase IQ failure by looking in the Sybase IQ message file.
Various.
Copy or rename the message log file (dbname.iqmsg) before trying to restart the database. This ensures that any useful information in the file will not be lost.
On UNIX, send a copy of the stack trace to Sybase Technical Support. The stack trace should be in the directory where you started the database server, in a file named stktrc-YYYYMMDD-HHNNSS_#.iq. If the database was open when the failure occurred, the stack trace should also be in the Sybase IQ message log (default name dbname.iqmsg). This information helps Sybase Technical Support determine why the failure occurred.
Restart the server with the start_iq command. When the database restarts, recovery occurs automatically.
Try to start the server without starting a database. If you are able to start the server but not the database, check that database parameters are specified correctly on the startup line and/or in the connection profile.
If you query catalog store tables extensively, restart the server and make sure that the TEMP_SPACE_LIMIT_CHECK option is ON. With this option setting, if a connection exceeds its quota of catalog store temporary file space it receives a non-fatal error.