Okay. We're done. Now, remember the ping configuration I did and do the configuration for that? So, I better do that now. I'll go ahead and run heroku config colon set and the first one was the app URL and that's going to be HTTP colon because we didn't set up a code to deal with HTTPS, but the URL will be our Probot app. The app URL is the Probot app URL. So, I'll paste in the Probot URL part portion there or the host portion there and then the other configuration was the Prometheus server and that's specified up by PROM underscore URL and that's going to be equal to our Prometheus server and that's right over here. So, I'll grab this portion of the URL and we're going to make that an HTTP URL as well, HTTP. That just basically means it's an unsecure URL which is not encrypted. There's not really much there for us to hide, but if you are doing this for production, you would probably want to improve that code and make sure that it works on our HTTPS. I'm going to just restart the app real quick so that I know that it's taking effect on our server. So, we can run heroku logs minus tail and take a look at log output. Looks like our Probot app is starting up just fine and we'll use the logs to come in and look at it real quick if we're testing it. So, let's go back to Probot and we'll refresh this window. That's great. We have one dot zero dot one. Now, let's try the metrics endpoint. So, that's going to be the Probot URL with the slash Probot and the additional URL called metrics. It looks like we have something in here. Okay, that's looking okay. Now, let's take a look at and run. Actually, let's take a look at the logs and see what we have. Yes, that's looking good. Now, let's take a look at the reset. Make sure the reset is working. Yes, that's working. So, we can see the resetting point message here and take a look one more time at the metrics. Okay. So at this point, I think we're ready to put some additional data together or some additional testing together by actually sending a message to our Probot app. So, the easiest way to do that is with GitHub. If we go back to github.com and we choose our repositories and we choose the repository that we were working on which is the WEB1066 Probot hello. That'll be your username. Then we look at our pull request. There's quite a few pull requests out there. There's actually some with failures. So, we'll have to look at that in a little bit to see why that's failing, but what's important here is that we're making sure that we're getting checks. These are failing but that's okay. So, let's go to our settings. We're going to look at the developer settings and choose GitHub apps and choose our Probot app. Take a look at the advanced settings. We can see some webhook deliveries here and we can see a check suite for that one. That's not the one we want. We do want this check run. All right so, if our Probot app is configured correctly, then what should happen is that when we deliver this message, this message should be received by our Probot app. So, we can see the last message here was Probot metrics, which probably means that we are getting some things there. All right, so let's redeliver that. That was redelivered. We can see the message come across here that a webhook was received by Probot. It was the check run completed and then that triggered our code and we can see the actual data being transmitted there. So, we can see that it was a pull request message. We got a status. It was failure message that as well we saw in the build. The repository that we're dealing with and the repository name. The duration that that executed in which is about a 122,000 milliseconds and that it finished and it was posted. So, that's great news. That means the webhooks are being delivered to our Probot app. Now, if we go look at the metrics page again and we refresh that, here's the metrics that are being published for us. So, that's going to show up in Prometheus for us. So, let's go check out Prometheus and see what we see there. So, here's Prometheus. It's up and running and I'm just going to click here again to make sure that it's actually started. There we go. We should refresh this. Let's give it a second to refresh. It'll take about 10 seconds here. If for some reason we're not getting the metrics, one of the troubleshooting things that we can do is we can go to the prom server and we can double check that the settings has config values for our Probot host. So, we're expecting to communicate with this Probot host and that should match Probot wenlockca in this case or your app name and then Heroku app. So, if there's a configuration problem with the Prometheus server reaching your Probot app, it might be right here. Okay, now that looks okay. All right. Great. So, our Prometheus server did reach our Probot app and we do see some data coming in here. There should be at least one metric for the number of counts for this particular. We can execute that and we can see, here's the data that Prometheus received. Now visually, this isn't very useful for us. It's like one data point. Not very interesting. So, this is where Grafana comes in. If we go back to Grafana and just visit the Grafana website. If you don't remember the website, it's really easy to find it again. Just go to grafana.com. Remember to log in with your credentials, choose your user ID, go to your org profile. Okay and then click on your hosted Grafana server and run your Grafana server. You should have your dashboard. You can load that up. We see that we do have some problem here with collecting our metrics. So, let's take a quick look at what's going on there. Notice that we have this week configured here. That should actually be set up for let's just say the last one hour. That way, we can actually see data because the amount of data here is really small. So, we see that Probot has an up time of 100 percent right now. We have 100 percent failure. That's not so good for us and we have one build that has been detected here. We also have zero percent success so far as well. All right, so there's the whole thing in the end.
