2

Fresh build of Exchange 2016 CU10 on Server 2016. Forgetting outbound for the minute, mailboxes on this server cannot send messages to each other on the same database. They can't even send messages to themselves. They just go to the Drafts folder.

Here's what I've tried:

  • In the send connector, used MX records and routing via smart host. Neither makes a difference.
  • Tried using external DNS to resolve, no differnce
  • Deleted the send connector altogether (in theory shouldn't need it for this), no difference
  • Ensured all name resolution is correct (DC configured in Exchange and on NIC, PTR records resolved, logs show resolution is ok)

Not really sure where else to go from here. I've scoured every log, I keep seeing this recurring error:

421 4.3.2 Service not available

This shows up in multiple logs, but I can't see any errors or reasons why this might be the case. I also can't figure out what service is not available, in fact I can see SMTP handshake with the smarthost in the protocol logs. Running Test-Mailflow is equally unhelpful:

RunspaceId         : 3ae5c152-51a0-46f8-b886-c6e6d80a033b
TestMailflowResult : *FAILURE*
MessageLatencyTime : 00:00:00
IsRemoteTest       : False
Identity           :
IsValid            : True
ObjectState        : New

Ok, FAILURE, but why? Adding -Debug -Verbose doesn't make any difference either, it just takes longer and displays the same information.

Get-ServerComponentState shows everything as 'Active' except:

ForwardSyncDaemon
ProvisioningRps

Not even sure I need 2016, at this point I'm wondering if I admit defeat and install 2013 instead. Anything else I've missed?

EDIT:

I found this blog post which (among countless others) seems to describe a similar situation. I have followed through the steps in here and subsequently messages are no longer stuck in the drafts folder. Now they are actually going to sent item,s but they are not yet appearing in the recipients inbox.

Improve this question
5
  • You could of missed a number of things, or not completed a number of steps. Take a look at this guide practical365.com/exchange-server/… start right at the beginning. I suspect either the namespaces and/or send connectors but like I say follow the guide as a check list
    – Drifter104
    Aug 29, 2018 at 10:00
  • Ok, let's assume I know what I'm doing and have done all the usual steps to set up my server. What I meant was is there anything I've missed in terms of troubleshooting.
    – Matt
    Aug 29, 2018 at 11:40
  • 1
    Going back the setup with a guide/check list is a troubleshooting step. There is an old saying about assuming anything :-) However.... Take outlook out of the question can you send mail using owa? Have you done any network captures to see what is actually going network wise? Are your DNS settings on the servers nic correct (only internal nothing external for example).
    – Drifter104
    Aug 29, 2018 at 11:50
  • Actually in fairness I did go through those articles and double check everything, so thanks for the link! I've actually only tested with OWA so far, but I'll try with Outlook too. Unlikely but worth ruling out. DNS setting seem good, I can resolve internal and external addresses with no trouble.
    – Matt
    Aug 29, 2018 at 12:11
  • 1
    I was trying to chase down this issue in a 2010 2016 coexistence setup where mail flow on 2016 was not working either direction. Thank you Matt for the one clue I found out of many blogs. Make sure your ex2016 DNS has ONLY INTERNAL address listed. For my example I had my PDC as primary DNS and my ISP's dns as secondary. Mail flow did not work. I removed the secondary external DNS entry and BINGO! After days of working this issue mail instantly started flowing in both directions.
    – istadmin
    Oct 26, 2018 at 17:23

2 Answers 2

Reset to default
0

Check if this blog provides any helps: Exchange 2013: Stuck messages in OWA’s Drafts folder and DNS

Moreover, please also run the following command to check if all Exchange services are running:

Test-ServiceHealth

If services are ok, try to move a mailbox to another new-created database and check again.

Improve this answer
1
  • As per my fourth bullet, DNS already checked and is ok. I ran the Test-ServiceHealth command, it just tells me the services are installed.
    – Matt
    Sep 18, 2018 at 11:44
0

I found my receive connector had been removed during a MS Patch. Remember, a server needs to be able to receive mail properly to be able to use the send connector. Here is a link to the article I used to rebuild all 5 receive connectors. After a rebuild and reboot, issue was solved. https://www.petenetlive.com/KB/Article/0001314

Improve this answer
2
  • If that link dies, or it is blocked in a user's country or corporate network, is there enough information here to solve the problem? If not, please add the essential bits in case something does happen.
    – Michael Hampton
    Sep 2, 2020 at 0:06
  • Doesn't help - the receive connected was not deleted by a patch. More importantly, the issue affects users in the same mailbox database, which requires no send or receive connector.
    – Matt
    Sep 4, 2020 at 1:55

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .