Within Redis Enterprise Cloud, the alerts section allows for 5 standard alerts to be set - what are some of the settings that people are using and why? We just stick with the default values and that has worked for us, but curious if people have reasons for using other values.
The defaults are good and provide the proper level of alert when you start to use Redis/Redis Cloud. These values will help you to see when you are close to the limits of your plan, and you may need to change it (take a bigger one for example because your application needs more data).
Also, the alerts that you see here, in addition to the ones that are available on a bigger plan (and Redis Enterprise Software) allow you to know when something not expected is happening.
What I mean by “not expected” is that when you selected the plan, you have worked on the sizing of your database based on some ideas of what the application will do with Redis. For example:
- my application will use 8Gb of data
- and will do between 12.000 and 15.000 ops/s
- and the latency should always be < 2ms
The alerts are here to check that the applications and databases are working as initially thought.
But this could have changed, more data than expected, more activity, or simply because your service is having more and more users.
The alerts will inform you of this; allowing you to work on the new sizing or new code depending of the type of events.
Dear @Joe in my experience I suggest my customers to configure alerts based:
- Use case for which they are using Redis, example for transactions use case you always need latency avg 1 ms but for caching use case you may be ok with latency avg 2 ms.
- Watch on actual traffic beyond Expected: Example in your case average traffic (throughput) may be 1000 ops/sec so you need to configure if alert if traffic increase and you may want to troubleshot why it happened.
Normally the default alerts works fine in most of the cases but for best your should configure it based on your use case and the expected traffic. you can also enable emails which you can further configure with slacks or pageduty kind of tools for further escalation.
Hope this helps.