Please more of these type of videos where you explain theories with real world example. Very useful for me working in a big company primarily on frontend, but want to understand what's going on elsewhere in the backend
Hi Sir, I think there is a correction required as DLQ is used for processing poisonous messages(consumers are not able to process the message/ taking more time etc multiple reasons). Having multiple instances of services and additional queue as shown to get the data from kafka(considering it is shared/externally managed) to our inhouse maintained queue is a good way. But the problem we are trying to solve is a different. In case of crash of our service, till atleast one instance of a service is up, we cant do much. we can make our users to wait and show some message applying circuit breaker, exponential backoff for retries mechanism till service is up. Inbox pattern can be used in the consumer side but not sure on this particular use case. Its just a suggestion. Thanks
Hello Sir, Thanks for the awesome videos. Love from India. If you could write the code and some some example code or provide some example code it would be more appreciated. It's only a suggestion Sir, but having said that your videos are very informative and relevant to real time issues which we might face.✌
Thank you! a. What if the Q fails? Then nobody consumed the event, then Kafka pushes it into its DLQ. is it? 1. what if Kafka fails before it receives the notification from PayPal? 2. What if Kafka fails after the notification is sent but there was network outage so the message was delivered to the Q but not confirmed back to Kafka? 3. How would you describe your solution from the CAP theorem perspective?
Please more of these type of videos where you explain theories with real world example. Very useful for me working in a big company primarily on frontend, but want to understand what's going on elsewhere in the backend
More to come! 😎
Hi Sir,
I think there is a correction required as DLQ is used for processing poisonous messages(consumers are not able to process the message/ taking more time etc multiple reasons). Having multiple instances of services and additional queue as shown to get the data from kafka(considering it is shared/externally managed) to our inhouse maintained queue is a good way. But the problem we are trying to solve is a different.
In case of crash of our service, till atleast one instance of a service is up, we cant do much. we can make our users to wait and show some message applying circuit breaker, exponential backoff for retries mechanism till service is up.
Inbox pattern can be used in the consumer side but not sure on this particular use case.
Its just a suggestion.
Thanks
love the content, please post more 😉
Hello Sir,
Thanks for the awesome videos. Love from India. If you could write the code and some some example code or provide some example code it would be more appreciated. It's only a suggestion Sir, but having said that your videos are very informative and relevant to real time issues which we might face.✌
Your Node.js tutorials, truly divine,
Keep uploading, you're our coding lifeline
🎉
I'm not going to comment on your moustache
You have to go to primary school. To learn basics of life. Or better to kindergarten.
Thank you!
a. What if the Q fails? Then nobody consumed the event, then Kafka pushes it into its DLQ. is it?
1. what if Kafka fails before it receives the notification from PayPal?
2. What if Kafka fails after the notification is sent but there was network outage so the message was delivered to the Q but not confirmed back to Kafka?
3. How would you describe your solution from the CAP theorem perspective?
good questions
Great content! please upload more on following topics
1. Pm2
2. Rabbitmq
Finnaly some real world backend stuff !!
What is the best books for learning modern nodejs ?