we have an eight node cluster with 2 passive and for some reason it fails with keep-alive and then the cluster trys to flip. We upgraded all the drivers for cards intel 100/1000 and broadcom 1gig eithernet thinking this was the issue. but still getting the cluster failures. Did run best practice but did not give any useful information. 00000fb8.000016d4::2009/08/26-20:59:57.293 ERR Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] DwMonitorPrimaryThread: Service process terminated unexpectedly, aborting. 00000fb8.0000194c::2009/08/26-20:59:57.293 ERR Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] DwMonitorSecondaryThread: process terminated unexpectedly, aborting 00000fb8.000016d4::2009/08/26-20:59:57.293 ERR Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] DwMonitorPrimaryThread: Service process terminated unexpectedly, aborting. 00000fb8.0000194c::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] Thread exited: DwMonitorSecondaryThread. Returning error 1067. 00000fb8.0000194c::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] DwResourceThread::decrement:: g_lThreadCount=3. 00000fb8.0000194c::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] EXCHANGE_RESOURCE::Release: Count=2. 00000fb8.000016d4::2009/08/26-20:59:57.293 ERR Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] EventLogging: Clustered Mailbox Server: VMBX108 Physical Server: MBX102 The IsAlive check for the resource Exchange Information Store Instance (VMBX108) failed. Error Code: 1067. 00000fb8.000016d4::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] State change: from 2 (Online) to 4 (Failed). 00000fb8.000016d4::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] Thread exited: DwMonitorPrimaryThread. Returning error 1067. 00000fb8.000016d4::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] DwResourceThread::decrement:: g_lThreadCount=2. 00000fb8.000016d4::2009/08/26-20:59:57.293 WARN Microsoft Exchange Information Store <Exchange Information Store Instance (VMBX108)>: [EXRES] EXCHANGE_RESOURCE::Release: Count=1.

First check if the Heartbeat is set properly in Network properties. Recommended private "Heartbeat" configuration on a cluster server http://support.microsoft.com/kb/258750/en-us also check if exchange folders are excluded from AV Vinod |CCNA|MCSE 2003 +Messaging|MCTS|ITIL V3|

Error 1067 equals to RES_DISK_CORRUPT_FILE, which indicates it cannot access disk resource properly Please upgrade the exchange to the latest patch, similar symptom has been fixed to the update Please disable all 3rd party products on the nodes, including the anti-virus software, and monitor the issue Please run the ExBPA against the CMS for a health check