15min for a restart seems to be on the high side. Check with the network folks if there are any recent changes and if the NIC is still working in full duplex. Make sure oracle and remedy servers are within the same VLAN or as close as possible. 4gb RAM seems be on the low side so check if any swapping happening during the restart. Monitor CPU, disk io, RAM usage during the server restart on both ars, oracle. Check with dbas to see if any kind of logs are turned on the db side that is slowing it down. There are few points in the remedy KBs and performance guide to change some parameters like caching of form display properties, increasing the pre-load threads etc.. I did not really see significant improvements by doing that but you can try if it makes any difference in your environment. Servergroup should not have any impact on restarts because the signalling between servergroup servers happen after the servers are completely started.
Here is what I have seen for restart times: 1.With a full duplex connectivity between remedy and remote oracle db server in a window 2008 64bit 8gb VMs with ITSM 7.6.03 (IM,CM,PM,CMDB,KM,SRM,SLA) we used to see 4min restart time for the complete stack. At somepoint it was like 6 to 8min but when the windows admins fixed some bad nic related issues it came back to 4 min. 2. With a full duplex connectivity between remedy and remote oracle db server in a Linux64bit VMs with base arsystem and custom applications I currently see less than 1 min restart time for ars. In this case all the VMs are carved out of same ESX physical server. _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org "Where the Answers Are, and have been for 20 years"
