Cloudwatch dasbhoards and alarms
1. Cloudwatch dasbhoards and alarms
Let's dive in and explore Cloudwatch logs for our current architecture. Then we will set up an alarm to send email when something goes wrong! First, head over to AWS Cloudwatch dashboard The first thing we see is alarms - but we have non yet. Head over to the cross service dashboard. The Cross service dashboard shows the most important metrics for any services being used in the account. If we scroll down a bit, we can see some high level stats for Kinesis Firehose - Incoming Bytes sum and Incoming Records sum. Right below that, we see Lambda invocations and errors. Click on "View Kinesis Firehose dashboard to drill down" Here we can see metrics on Incoming Bytes, incoming records, and delivery to Elastic Search. Click on Incoming Records, then View in Metrics. Uncheck everything except the Relevant delivery Stream. Then go to actions, add to Dashboard and Select a new dashboard. Create the dashboard and add the metric to the dashboard. Now let's do the same for for Delivery to Elasticsearch Success Now, let's create an alarm. From the dashboard, click on the menu above the incoming records sum metric, and click "view in metrics". Next to the twitterStream name, click the bell icon. Configure the alarm to trigger when there are less than 5 tweets in 5 minutes. Click next. Next, select an existing SNS topic. I have already pre-created a topic to send an alert to. Write name and description, and create the alarm!2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.