2012년 3월 16일 금요일 오전 7:25
Server 2008 R2 Enterprise SP2 with all updates but the last two months worth of .NET patches
Exchange 2010 SP1
Running as a VM on Citrix Xen server
In site DC/GC which is also the FSMO master
IIS 6 and 7
All 3rd party software has been removed except for Citrix Tools for Virtual Machines.
.NET 3.5SP1 and .NET 4.0
After a failed .NET patching the entire server was broken, almost no snap-ins would load in MMCs, no .NET based services would start (net.tcp services etc), almost all Exchange services, everything related to IIS, and some others would start, the network wouldn't come up for 5 minutes after logging in, and many other broken things. After removing IIS and .NET and reinstalling 3.5SP1 and 4.0, everything is mostly up and running aside from four Exchange services.
The Transport service is stuck in starting, and does not give any errors, ever. The Mailbox Replication, Protected Service Host and normal Exchange Service Host services fail with error 1053; did not respond in a timely fashion.
All IIS pages are giving out 500 errors, but I can deal with that later, right now I just need mail to work.
When I start the transport service, or reboot the server, I get event ID 12025 from MSExchangeTransport, but based on Googling around, this doesn't seem to be indicative of the problem, just something that happens:
Transport service is disconnecting performance counters with process lifetime from their old process.
I am receiving these errors every so often, but haven't paid much attention to them as I assume they're generated because the transport service isn't running:
Service MSExchangeMailSubmission. Unable to process anything for over 30 minutes. Diagnostic info: Governor: ServerGovernor for DatabaseManagerEvent, Last run time: 3/16/2012 2:24:58 AM, Next retry interval: 00:00:05. Current exception: Microsoft.Exchange.Assistants.TransientServerException
at Microsoft.Exchange.Assistants.Util.TraceAndThrow(CatchMe function, AIException aiException)
at Microsoft.Exchange.Assistants.Util.CatchMeIfYouCan(CatchMe function)
at Microsoft.Exchange.Assistants.Base.CatchMeIfYouCan(CatchMe function)
at Microsoft.Exchange.Assistants.EventDispatcherPrivate.DangerousProcessItem(EmergencyKit kit, InterestingEvent interestingEvent)
I also receive event 1009 from MSExchangeMailSubmission, which just says that no hub transport servers can be contacted. This doesn't surprise me since it's the only Exchange server in this environment.
Currently, I'm wondering if building a new Exchange VM and migrating the databases is going to be a better fix than trying to get this all sorted out because I've been working on this for the last 7 hours, and I'm only marginally better off than I was at the start. I'm not sure, however, about what will be required to do so. If I can just dismount the virtual drive, and then attach it to the new server, update firewall rules, and recreate the send/receive connectors, that's great. I can worry about all the rules and policies later. I would rather get everything working in this VM, but I need to get everything back up and running in the least amount of time possible.
We are not using online archiving, circular logging is not enabled, there is an old Exchange 2003 server in the organization that hasn't been decommissioned, but is not in active use.
- 편집됨 RobbieCrash 2012년 3월 16일 금요일 오전 7:26
2012년 3월 17일 토요일 오후 2:31Try putting on SP2 to see if this address the issue before deplying another server.
2012년 4월 1일 일요일 오후 12:47
I would usually start my troubleshooting based on KB http://support.microsoft.com/kb/944752 when there is a 1053 error when starting managed code services.
What is the current status of the issue? Did you manage to build a new server or any progress of the troubleshooting on the existing server?