Exchange Server 2016 standalone deployment/coexistence Deployment (vii)--DAG functional Testing

Set up the DAG, and the DAG Replication network, and finally create a copy of the database, only the three have been completed, so congratulations, your exchange in the DAG function is complete, so today we will test the effect of this DAG function.

First, let's look at the status of the configured DAG feature

Since the primary database is in mail01, we log on to MAIL02 and use the following command to see the replication status

Get-mailboxdatabaseCopystatus

650) this.width=650; "height=" 147 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M02 /7d/88/wkiol1bqxjhdhvn-aaagelv9av0686.png "border=" 0 "/>

From here we can see that the two databases MDB01 and MDB02 in the mail02 are in a healthy replication state, and the replication queue is 0, which shows that real-time replication works well.

But it's no good to look at this, or I'm finished with this article ... we're still going to test it! To test! To test!

Come on, go on, first we find a database of user mailtest that is stored on the new database, and its database is MDB02

650) this.width=650; "height=" 456 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M00 /7d/8b/wkiom1bqxasdgxmraahtqdzokmk018.png "border=" 0 "/>

Next we jump to the list of databases in the ECP console and we can see the current copy status of our MDB01 and MBD02.

BTW: Because I do not use the physical NLB load balancer device in my own environment, we are here to manually specify the MAIL01 (192.168.1.8) of the ECP, we can also see that our two new database is activated on MAIL01,

This means that MAIL01 is now the primary database, and mail02 belongs to the standby state.

650) this.width=650; "height=" 461 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M01 /7d/8b/wkiom1bqxawrcoa7aaevbgr3av8791.png "border=" 0 "/>

Next, we shut down this mail01.

650) this.width=650; "height=" 356 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M01 /7d/88/wkiol1bqxjxjb_17aabwy5ie_fu402.png "border=" 0 "/>

We log on to mail02 on this server, open failover Cluster Manager, we can see immediately, inside the node, a total of two devices, one of which has been down.

650) this.width=650; "height=" 401 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M02 /7d/8b/wkiom1bqxaibjcmgaacemsuen3o250.png "border=" 0 "/>

Click to see the details, we can see the specific outage of the server is which one.

650) this.width=650; "height=" "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M00 /7d/8b/wkiom1bqxanxd-osaabrzgah1ie873.png "border=" 0 "/>

Take this time, does our mail system dag function really play a role? We wait and see!

At this time, the Mail01 ECP we opened was unresponsive, and we re-signed in to the ECP of Mail02 (192.168.1.9) (such as configuring a physical NLB load balancer device, ignoring this step)

650) this.width=650; "height=" 462 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M01 /7d/88/wkiol1bqxjnda3ahaaeljonwwpu600.png "border=" 0 "/>

As we can see clearly in the, the server that hosts the database automatically jumps to mail02 from the previous mail01.

Looks like a jump, but how about sending and receiving mail? We open MAIL02 's OWA to test it.

First we use an internal mailbox and an external mailbox to send a message to this mailtest, respectively. Check to see if the email feature has been affected.

After opening OWA, I am glad that we have received it!

650) this.width=650; "height=" 420 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M01 /7d/8b/wkiom1bqxavbmo-baabqe3bqods073.png "border=" 0 "/>

650) this.width=650; "height=" "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M02 /7d/88/wkiol1bqxjrbt416aaauceiap0e510.png "border=" 0 "/>

650) this.width=650; "height=" 239 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M02 /7d/88/wkiol1bqxjuqsngcaaacpksxzwk531.png "border=" 0 "/>

We also reply to these two emails to see if the effects of sending emails are affected.

Haha, internal users and external users have received the mail I sent.

650) this.width=650; "height=" 414 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M00 /7d/8b/wkiom1bqxa7xpdtpaabsa15v71s386.png "border=" 0 "/>

650) this.width=650; "height=" 539 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M01 /7d/8b/wkiom1bqxa-srfewaaftu0h35ic592.png "border=" 0 "/>

This is basically the end of our Dag's functional testing. The effect of this dag is very good. In my this test, about the server's jump time of less than 1 minutes, the effect is very satisfactory.

By the way, finally, the main server hosting the database we can manually switch, this can be based on server performance and some security considerations, if it is due to server downtime, the main database will automatically jump, as I did this test.

However, if you do not manually adjust the back, the database will still continue to stay on the server after the jump. So after we restore the server state, we can modify it as appropriate.

650) this.width=650; "height=" 657 "title=" image "style=" border:0px; "alt=" image "src=" http://s3.51cto.com/wyfs02/M00 /7d/88/wkiol1bqxj-bgjnbaacrypbjjqe554.png "border=" 0 "/>

This article is from the "June Ma Run Space" blog, be sure to keep this source http://horse87.blog.51cto.com/2633686/1752162

Exchange Server 2016 standalone deployment/coexistence Deployment (vii)--DAG functional Testing