Issue on one of our CAS Servers, clients disconnect and some bad events in the Cas server events log.
Exchange systems: We got 2 Mailbox Servers they carry both 6 Databases Also 2 Cas Servers behind a load balancer.
Hardware: CAS1: ESX VM – 4 CPU 16GB MEM
CAS2: ESX VM - 4 CPU 16GB MEM
CAS1 ON active HOST09: HP Proliant BL490c G6, 8 CPU, 98GB
CAS2 ON active HOST08: HP Proliant BL490c G6, 8 CPU, 98GB
During the day we get very often clients calling about their outlook say’s trying to connect at that moment we look at our Cas servers and one of them spikes till a 100% Cpu Load for about 10mins.
Cas1 is running normal without any issue but Cas2 uses high CPU load for the above timeframe.
When looking at the RPC connections
CAS1 holds about ~ 512 conss
CAS2 holds about ~ 200 conns
(so you should expect Cas1 to have a heavier load but that is not the case )
Event logs from Cas2 show the following error’s
- Application Log:
EVENT ID 1, MS Exchange Auto discover
EVENT 4002, MS Exchange Availability
- System Log:
EVENT: 10009, DistributedCom
Warning: EVENT 5010, WAS
We did a lot of research to these events, but nothing helps.
Any suggestions for us, if something is not clear please let me know I will try to specify a bit more.
