Exchange 2016

We have implemented this template and find that the default EWS Advanced Health monitor fails. Where can we seek assistance troubleshooting this monitor transaction? With simple monitoring EWS functions operate as expected.

Burrell, what happens if you run the curl command used by the monitor from the command line: curl-apd -g -s --ntlm -k -X POST -H 'Content-Type: text/xml; charset=utf-8' -d "<?xml version="1.0" encoding="UTF-8"?><soap:Envelope xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/" xmlns:t="${DIR}/types" xmlns:m="${DIR}/messages"><soap:Header></soap:Header><soap:Body><GetFolderxmlns="http://schemas.microsoft.com/exchange/services/2006/messages"><FolderShape><t:BaseShape>IdOnly</t:BaseShape></FolderShape><FolderIds><t:DistinguishedFolderId Id="inbox"><t:Mailbox><t:EmailAddress>youremail@domain.com</t:EmailAddress></t:Mailbox></t:DistinguishedFolderId></FolderIds></GetFolder></soap:Body></soap:Envelope>" -u domain\user:password http://<your_server_address>/EWS/Exchange.asmx | grep -i "<m:GetFolderResponseMessage ResponseClass="Success">"

We found that our conflict with the EWS Advanced Health monitor was due to NTLM version of our new AD Domain. We are using the simple monitors going forward.

When implementing this template I have found that using the advanced health monitor and 2 monitor mailboxes causes the node to stay active even when you put it into maintenance mode. The pool monitors are set to require 2 of the 3 monitors to be up but it appears that putting the server into maintenance mode does not affect the mailbox tests so both mailboxes report up and cause the node to remain in the pool.

JTrimble, Could you provide some more details on the advanced monitor setup? You are saying if you put the Exchange server into maint mode the two advanced monitor tests with two unique user accounts are staying up? Could you run the command listed above in Mikeshimkus's comment from the BIG-IP command line and see the result? Also to check the result when the server is in maint mode would be helpful to see if that curl command produces a different result as expected.

Hi - a few of us have been trying to find the Exchange 2016 iApp template - where is is exactly...? We cannot find it in downloads.f5.com or devcentral or google etc

Hi B.Earp. As stated in the introduction above, the 2016 Template is in the RELEASE_CANDIDATES directory of the zip file on downloads.f5.com. Once you download the zip file, you'll see the RELEASE_CANDIDATES directory inside. Extract the 2016 template from there.

When I run the code by mikeshimkus I get nothing returned. No error message, but nothing.

Hi JTrimble, Maybe you should add a -I (i uppercase) in the curl command to see response headers

Yann is correct, I forgot to add a flag to display the response. You could also use "-vv" in the curl-apd command.

Still get nothing, even adding both -vv and -I (upper case i). I remove the grep statement to see if anything at all is returned and I get nothing. This is on 11.4.1 so I'm not sure if things have changed in more recent updates... *Update* - Updating to the final release of the template has fixed this for me, the nodes now show down when put into maintenance mode.

with thank

Exchange 2016 Deployment Guide references an iAPP available in versions 11.4.1 - 12.1. However, it is not present in 11.5.4.

Hello Mike, the iApp is downloadable, and does not ship with the BIG-IP by default (if fact, you should only be using downloadable iApps for Exchange of any version, and not the Exchange 2010 that ships on the box). The deployment guide and the solution linked above both have instructions on downloading the iApp for Exchange 2016. Note that this template is fully supported by F5 networks support. Thanks, Joe

Hi, I tried to use the iApp at customer site but have trouble with it. We are working with software 11.6.0 and apm/ltm setup. The exchange mrsproxy test (Powershell Command: Test-MigrationServerAvailability -ExchangeRemoteMove -Autodiscover -EmailAddress ... -Credentials ...) is failing because the application profile for exchange does provide only basic and ntlm www-authentication header, but mrsproxy is expecting a negotiate www-authentication header. If I disable the application profile mrsproxy is working (header with negotiate now offered to client), but in turn autodiscover is not working any more.

Is there anything I missed? I found the extra irule for mapi-over-https wihtin deployment guide and included it. But is there anything other to do?

Regards, Lars

Lars, did you try the iRule on page 69 of the deployment guide: https://www.f5.com/pdf/deployment-guides/microsoft-exchange-2016-dg.pdf

That should allow the necessary requests for EWS and Autodiscover to pass through APM directly to the server.

Thanks for the answer. I tried something similar - but in a not very sophisticated way. I will try the iRule.

As a feedback: The irule from page 69 seems to do the job so far. We also discovered the exchange powershell testcommand needs a user with administrative rights.