Exchange2010 witness server/folder stops automatic

I have 2 exchange servers with all the roles installed.
and DAG configured. But evry 2-3 weeks witness I have to delete the witness folder and recreate otherwise if I restart any of these servers database will get dismounted.

But if delete the folder and recreate the folder and update from DAG properties it works fine.


Did anyone face the same issue before?
LVL 29
MAS (MVE)EE Solution GuideAsked:
Who is Participating?
 
suriyaehnopCommented:
0
 
Will SzymkowskiSenior Solution ArchitectCommented:
Have you tried putting the witness server on another machine? Do you get the same results?
What service pack are you running for exchange 2010? It would be recommended that you are running the last one so avoid any bugs.

You can verify the SP and Roll up at the below link.
Service Pack Roll Up

Will.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
I tried another witness but it is the same result
I am running exchange SP2
0
Simplify Active Directory Administration

Administration of Active Directory does not have to be hard.  Too often what should be a simple task is made more difficult than it needs to be.The solution?  Hyena from SystemTools Software.  With ease-of-use as well as powerful importing and bulk updating capabilities.

 
MAS (MVE)EE Solution GuideAuthor Commented:
It is working for few days and stops.
FM
Now I have a doubt that my FSMO holder is having performance issue as now internal emails taking more than 30seconds to deliver even test email. before it was less than 5 seconds.

It could be a reason?
0
 
pcmghouseCommented:
Can you explain where your witness share is? I hope it's not on the dag.
Exchange Trusted Subsystem is the security group you need to be concerned about.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
I configured in my DHCP server not in exchange server.
And Exchange Trusted Subsystem is a member of Local admin group
0
 
pcmghouseCommented:
Can you please check your Failover Cluster manager on the DAG. What is the state of the Witness resource during that time. Can you bring it online manually if its offline.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
I tried restarting one server and saw node is online is cluster manager but outlook is showing "Trying to connect..." or "Disconnected"


Now I can see the attached error but I can ping each other
screenshot.docx
0
 
pcmghouseCommented:
check your witness share using this command.
cluster dag.domain.com res

And second thing:
your client connects to cas server. how are you doing load balancing for the cas servers as they are on the same servers as mailbox role. you cannot use nlb. hardware load balancer will solve your situation efficiently.
One cas server restarts and your outlook disconnects and tries to connect to the second cas server. If you have a hw load balancer that will be more than quick.
And third:
Can you please run in exchange mgmt shell: get-clientaccessarray|fl
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
Resource             Group                Node            Status
-------------------- -------------------- --------------- ------
Cluster Name         Cluster Group        EXCH2010-1      Online
File Share Witness (\\win-2008.domain.com\DAG.nasser.com) Cluster Group
EXCH2010-1      Online
IPv4 DHCP Address 1 (Cluster Group) Cluster Group        EXCH-1      Online
IPv4 DHCP Address 2 (Cluster Group) Cluster Group        EXCH-1      Failed

Thsi is teh result of the first command
0
 
pcmghouseCommented:
Can you fix the IPs of the DAG instead of taking them from DHCP. make sure those IPs are excluded from DHCP distribution.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
It is already excluded from DHCP if I am not mistaken
0
 
pcmghouseCommented:
then fix the static ips for the dag.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
It is already assigned before
0
 
pcmghouseCommented:
IPv4 DHCP Address 2 (Cluster Group) Cluster Group        EXCH-1      Failed

Can you explain what this network is. The naming is not clear.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
EXCH-1  is the name of the server
This is our internal network not replication network
0
 
pcmghouseCommented:
I understand that.
Your IP(s) should be online on EXCH2010-1 as that is holding the other two resources.
Do you have assigned two V-IPs to the DAG.
0
 
pcmghouseCommented:
can you give the output for this:
get-databaseavailabilitygroup|fl *ip*

Also check if dag is getting IP from DHCP. I suggest you set it to static. Go to DAG/Properties/Ip Addresses.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
Result of command get-databaseavailabilitygroup|fl *ip*

DatabaseAvailabilityGroupIpv4Addresses : {10.0.0.27}
DatabaseAvailabilityGroupIpAddresses   : {10.0.0.27}

I still didn't understand how NODE IP is related to DAG IP?
0
 
pcmghouseCommented:
I still didn't understand how NODE IP is related to DAG IP?  Can you explain in detail.


node1 - 2 ips (lan, heartbeat).
node2 - 2 ips (lan, heartbeat).

dag - 1 virtual ip (note that this ip is not used for client connectivity purpose).

I hope all of the above are different.

Now can you check what is the IP assigned to the failed resource. From my point of view, you should see only three resources (cluster name, cluster ip [dag ip] and witness). Looks like the failed resource is not used.
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
Now emails receiving after 5-6 minutes. Before it was less than 15 seconds.
But if switch databases to EXCH-1 emails flowing fast but when I switch DBs to EXCH-2 all the emails are very slow.
Any idea ?
0
 
pcmghouseCommented:
Can you show the cluster dagname res output now.

And second thing:
your client connects to cas server. how are you doing load balancing for the cas servers as they are on the same servers as mailbox role. you cannot use nlb. hardware load balancer will solve your situation efficiently.
One cas server restarts and your outlook disconnects and tries to connect to the second cas server. If you have a hw load balancer that will be more than quick.
And third:
Can you please run in exchange mgmt shell: get-clientaccessarray|fl
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
Attached results
results.txt
0
 
MAS (MVE)EE Solution GuideAuthor Commented:
I ended up installing one more mailbox server and added to and now its stable.

Thanks to all
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.