Christmas Presents Come Early
I'm happy to announce the SenseDeep summer "goodies" release which is a collection of many little improvements.
Every so often we like to accumulate requested features and suggestions into a "goodies" release. While there may not be a single compelling feature in this update, when added together, all these small improvements have an outsized impact.
Thank you to all of you who suggested the features and fixes you would like addressed.
Many Small Changes
The following improvements have been made in this update:
- More granular function and application error metrics
- Improved function invocations filtering and searching
- Improved correlation of graphs and metrics for functions and applications
- Functions force update and count down
- More links
- Resolve alerts from function log panel
- CloudWatch log create and delete from log list
- Update CloudWatch log retention
More granular function and application error metrics
AWS Lambda detects system errors, but your Lambda functions can fail in many ways that are not detected by AWS and are not counted in the standard AWS/Lambda Errors metric.
SenseDeep tracks application log errors, memory errors, timeout errors and exceptions. These are tracked separately with metrics in the SenseDeep/Lambda namespace.
The functions page now displays App Errors and Memory Errors separately. SenseDeep now relies on the SenseDeep/Lambda Errors metric rather than the AWS/Lambda metric as it includes these normally unseen error sources.
Improved function invocations filtering and searching
The functions invocations list has improved filtering and searching with quick actions to filter for App errors (log patterns), memory errors and timeouts. These quick actions filter the displayed logs for errors of a specific class.
You can also enter filter text in the filter box to search for specific request IDs. This will search all invocations within the selected date range.
Improved correlation of graphs and metrics for functions and applications
Previously we used the CloudWatch AWS/Lambda Errors metric to supply data for the graphs on the function page. However, because AWS does not count internal Lambda errors, this sometimes caused confusion as the Metric cards and invocations displayed a different error count. We've now standardized on using the SenseDeep/Lambda error metrics which renders a more consistent picture.
The Application pages have similarly been updated to use the SenseDeep metrics.
Immediate functions update
The SenseDeep function page will automatically update to display new data. Invocations are updated every ten seconds while the more costly graphs and metrics are updated once per minute.
The top of the function page now features a countdown timer to the next update. Click on this time to force an immediate update. Click on the App name to view that application. Click on the cloud to display the cloud list.
The description under the function title now has quick links to the relevant application and cloud.
Resolve alerts from function log panel
If you click on an Alert email message, you will display the invocation that triggered the alarm. From there you can view the invocation details and exact log messages. From there, you may decide the alert has already been addressed by your dev team and so you want to mark it as
Previously you had to navigate back to the alert list to mark the alert. But now, we've added a
resolve button to the alert details so you can immediately mark the alert as resolved.
Update CloudWatch log retention
CloudWatch does not provide a facility to set the log retention for multiple logs in one command. No worries, with SenseDeep you can now easily set the CloudWatch log retention for any number of logs in one operation.
You can also simply delete logs and create new log groups from the log list.
I hope you get value from all these smaller improvements and enjoy using SenseDeep.
Please speak up if there is a missing feature that would make your life easier. We love to make SenseDeep a joy to use.
SenseDeep is an observability platform for AWS developers to accelerate the delivery and maintenance of serverless applications.
SenseDeep helps developers through the entire lifecycle to create observable, reliable and maintainable apps via an integrated serverless developer studio that includes deep insights into how your apps are performing.
To try SenseDeep, navigate your browser to: https://app.sensedeep.com.
To learn more about SenseDeep please see: https://www.sensedeep.com/product.
Please let us know what you think, we thrive on feedback. firstname.lastname@example.org.