Added information about Lambda Interruption #199
Conversation
There was a problem hiding this comment.
Thanks for the submission, @halilduygulu!
Do you have a linkable source source for this (even AWS support comment forums-level)?
| @@ -1055,6 +1055,7 @@ Lambda | |||
| - 🔸Lambda is a new technology. As of mid 2016, only a few companies are using it for large-scale production applications. | |||
| - 🔸Managing lots of Lambda functions is a workflow challenge, and tooling to manage Lambda deployments is still immature. | |||
| - 🔸AWS’ official workflow around managing function [versioning and aliases](https://docs.aws.amazon.com/lambda/latest/dg/versioning-aliases.html) is painful. | |||
| - ❗Currently (October,2016) Lambda functions can sometimes stop working for 2-3 minutes for failure recovery purposes according to a support ticket answer from Lambda dev team. They are working to prevent this in the future. | |||
There was a problem hiding this comment.
Suggestion: "(as of October, 2016)"
| @@ -1055,6 +1055,7 @@ Lambda | |||
| - 🔸Lambda is a new technology. As of mid 2016, only a few companies are using it for large-scale production applications. | |||
| - 🔸Managing lots of Lambda functions is a workflow challenge, and tooling to manage Lambda deployments is still immature. | |||
| - 🔸AWS’ official workflow around managing function [versioning and aliases](https://docs.aws.amazon.com/lambda/latest/dg/versioning-aliases.html) is painful. | |||
| - ❗Currently (October,2016) Lambda functions can sometimes stop working for 2-3 minutes for failure recovery purposes according to a support ticket answer from Lambda dev team. They are working to prevent this in the future. | |||
There was a problem hiding this comment.
I'm not 100% clear on what "for failure recovery purposes" means - is it 2-3 minutes while recovering from failure?
There was a problem hiding this comment.
As i sad it was a support ticket response. I don't have link to it because it is in our account, but i can copy paste the exact response here.
I've received an update from the engineering team.
They replied that customer may see their stream stop processing for three minute or less very infrequently due to certain failure recovery modes in our system. Their stream processing should then resume and recover from the delay.
They are working improvements to our system that should eliminate these sorts of pauses, we don't have a solid ETA on when that will be available.
There was a problem hiding this comment.
Also it happened to my 2 functions 4 times in total last week.
There was a problem hiding this comment.
There's a huge difference between "Lambda functions can stop working" and "streams can stop processing" though. This should belong in the Kinesis section then, imo.
There was a problem hiding this comment.
This is an excellent addition, thank you @halilduygulu - a few requests related to the undocumented nature of the behavior.
| @@ -1055,6 +1055,7 @@ Lambda | |||
| - 🔸Lambda is a new technology. As of mid 2016, only a few companies are using it for large-scale production applications. | |||
| - 🔸Managing lots of Lambda functions is a workflow challenge, and tooling to manage Lambda deployments is still immature. | |||
| - 🔸AWS’ official workflow around managing function [versioning and aliases](https://docs.aws.amazon.com/lambda/latest/dg/versioning-aliases.html) is painful. | |||
| - ❗Currently (as of October, 2016) Lambda functions can sometimes stop working for 2-3 minutes for failure recovery purposes according to a support ticket answer from Lambda dev team. They are working to prevent this in the future. | |||
There was a problem hiding this comment.
Since this isn't documented officially, would you mind doing the following?
- Adding a "undocumented" scroll icon after the serious gotcha red exclamation point? (📜)
- Convert "October, 2016" to be a link pointing to your support ticket text? ( https://github.com/open-guides/og-aws/pull/199/files/c99bddb4ee2437587f1e188d47be2bb1da01f81d#r83529126 )
|
@ThanosBaskous I made changes, could you check again. Thanks for reviews. |
There was a problem hiding this comment.
Thanks for the fixes, @halilduygulu! I have one more request to change something that was in the original review - sorry that I didn't catch it the first time around.
| @@ -1055,6 +1055,7 @@ Lambda | |||
| - 🔸Lambda is a new technology. As of mid 2016, only a few companies are using it for large-scale production applications. | |||
| - 🔸Managing lots of Lambda functions is a workflow challenge, and tooling to manage Lambda deployments is still immature. | |||
| - 🔸AWS’ official workflow around managing function [versioning and aliases](https://docs.aws.amazon.com/lambda/latest/dg/versioning-aliases.html) is painful. | |||
| - ❗📜 Currently [as of October, 2016](https://github.com/open-guides/og-aws/pull/199/files/c99bddb4ee2437587f1e188d47be2bb1da01f81d#r83529126) Lambda functions can sometimes stop working for 2-3 minutes for failure recovery purposes according to a support ticket answer from Lambda dev team. They are working to prevent this in the future. | |||
There was a problem hiding this comment.
Please replace "dev" with "development.
There was a problem hiding this comment.
done and thanks for accepting.
|
Thanks @halilduygulu! Since this doesn't have a topic, assigning to @jlevy for secondary review. |
|
👍 Thanks! |
Added information about Lambda Interruption which happened on my project this week. The information comes from aws support ticket and lambda engineering team according to support staff.