We're running CiscoWorks 2000 and HP Openview on a Solaris 8 E4500. If I don't shut down the ANIServer service right within 15 minutes after a reboot it will eventually lock up CW and snmpcollect and netmon fail. The only way to recover is to perform another reboot and shut down the service. Everything else runs fine after that. We opened a case with the TAC who asked up to un-install and re-install CW. This didn't change anything. Does this sound familiar to anyone?
We don't run on Solaris, but we do have problems with discovery locking up. In our case it is due to "Out of Memory" exceptions and subsequent "Null Pointer" exceptions. Review you aniserver.log file for those two strings. If you have them, then your database is corrupt and will have to be rebuilt.
Campus Manager has problems with memory when you get over about 1200-1400 devices.
We finally had to call Sun on this issue and they informed us that it may have been caused by a problem with a patch (I don't have the number in front of me) that was removing the Loopback interface on the box when a high number of connections were established. We uninstalled the patch and everything seems to be working now.