Known Problems

This section describes known problems and workarounds for Adaptive Server Enterprise Cluster Edition.

Table 1. IBM-related issues
CR #	Description
551965	Cannot find agents with UDP server discovery Neither sybcluster nor the Adaptive Server plug-in wizards can find the agents running on AIX 6.1.1 machines with UDP server discovery: sybcluster -U uafadmin -P Workaround : Utilities can find all the agents in the local subnet including those running on AIX 6.1.1 machines if you connect specifically to the machine’s agent: sybcluster -U uafadmin -P -F aix1:9999 Prevent this from happening by installing the 6100-02 Service Pack 1 from IBM.
550035	Configuring Job Scheduler fails on IBM AIX srvbuild and srvbuildres fail to configure Job Scheduler. Workaround : Add $SYBASE_JRE6/bin/classic to the LIBPATH environment variable before running srvbuild or srvbuildres.
544399	Registering Adaptive Server plug-in during installation fails The Adaptive Server plug-in does not register automatically during installation. Workaround : Manually register the plug-in: Change to the $SYBASE/ASEP/bin directory. Generate a file called ASEPlugin.jpr under $SYBASE/ASEP/bin by entering `registerASEP`.

Table 2. Known nonplatform-specific Adaptive Server issues
CR #	Description
572710	Taking an engine offline while Java clients are connected may cause an error If there are Adaptive Server users that are executing Java-in-SQL statements while you try to take the engine offline using sp_engine "offline", `engine_id`, Adaptive Server may report stack traces in its error log if the engine is hosting a JVM. Workaround:Do not take the engine offline until Java clients are disconnected. Instead, take the engine offline that is not hosting the JVM.
574278	Timeslice error during `disk init` and disk resize If you run disk init or disk resize involving file system devices, Adaptive Server may hit a timeslice error involving filesystem devices when you: Issue a create database command using the device you just created, or Activate a disk you just created after you restart the server. Workaround: Increase the `cpu grace time` configuration parameter to give enough time for the operating system to flush all dirty buffers of the device that you are creating. The appropriate amount would depend on the disk speed and the amount of data being cached, but typically, a 10Gb device may require that you set the parameter to around 2500. Once you initialize the disk and use it for the first time by issuing create database, you can reset the cpu grace time parameter back to its original value. Warning! Do not use Adaptive Server Cluster Edition on file system devices in a production environment; file system devices are for testing only.
572469	Task may get into an undetected deadlock Under very rare circumstances, a task de-allocating a page may get into an undetected deadlock with another task trying to access the same page from another instance. Workaround: Restart the instance.
549511	Cannot delete entered logical devices for audit tables in the auditinit utility Workaround: Stop and restart the auditinit utility.
487525	Starting Unified Agent may report a permissions error When you initially execute the uafstartup.sh script, you may receive an error stating that you do not have permissions to access some files. This occurs when starting the agents for the first time in the current $SYBASE directory only if starting multiple UAF agents at the same time. This problem is caused by latency issues in the NFS file system layer. Workaround: Wait a few seconds, then retry.
333307	Blank spaces in an ldap server entry If you have a blank space after your ldap server entry, dscp defaults to using the interfaces driver, and does not connect to an ldap server. Workaround : Enable ldap by editing the $SYBASE/OCS-15_0/config/libtcl.cfg file to add the ldap server you plan to use.