Skip to content

Conversation

@manototh
Copy link
Collaborator

No description provided.

Users were confused about which join types are currently supported, leading to
query failures when attempting to use leftouter, rightouter, or fullouter joins.
The existing note was easy to miss.

Changes:
- Changed Note to Warning to make the limitation more prominent
- Explicitly state that unsupported join types will cause query failures
- Added Tip suggesting lookup operator as a workaround for some use cases
- Improved discoverability of current limitations

Related support ticket:
- https://app.plain.com/workspace/w_01HA1Z15QQY1X6QR29KFMBMYTH/thread/th_01K897KEWHRYYY8412F9CAM37H/
Users were unclear about the maximum number of rows they could retrieve using
limit and take operators, leading to confusion when trying to export large datasets.
The 50k limit was mentioned in some places but not prominently in the operator docs.

Changes:
- Added explicit 50,000 row limit to parameter descriptions
- Added Note callout explaining the limit and suggesting alternatives
- Provided guidance on pagination for larger datasets
- Made the limitation discoverable where users configure these operators

Related support ticket:
- https://app.plain.com/workspace/w_01HA1Z15QQY1X6QR29KFMBMYTH/thread/th_01K7WNSFCZHXFZHK50PZ0CKQPS/
Users with ingestion delays were missing alerts because their match monitors
evaluated events before they arrived in Axiom. The secondDelay parameter exists
to handle this but was not documented in the monitor guides.

Changes:
- Added new "Handle ingestion delays" section with step-by-step guidance
- Explained how to measure ingestion delay using ingestion_time() function
- Documented the secondDelay API parameter with practical examples
- Added Tip explaining how match monitors evaluate events based on _time
- Clarified the relationship between ingestion delay and monitor timing

Related support ticket:
- https://app.plain.com/workspace/w_01HA1Z15QQY1X6QR29KFMBMYTH/thread/th_01K86V64NV96VW7W9KK215FF2D/
Users received unclear error messages when creating anomaly monitors without
proper time binning. The requirement for bin(_time) was mentioned but not
emphasized enough, leading to confusion about why monitors failed to create.

Changes:
- Added Warning callout explaining time series query requirement
- Provided clear example of correct query structure with bin(_time)
- Explained that summarize must bin by _time for anomaly detection
- Made the requirement visible during the query definition step

Related support ticket:
- https://app.plain.com/workspace/w_01HA1Z15QQY1X6QR29KFMBMYTH/thread/th_01K80W57WSN7G541TF0D8MDEY8/
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant