-
Notifications
You must be signed in to change notification settings - Fork 461
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[elastic-agent] Update package with additional data streams #1298
Conversation
💚 Build Succeeded
Expand to view the summary
Build stats
Test stats 🧪
Trends 🧪 |
list of processes agent is able to run
i think endpoint, heartbeat and packetbeat are missing, |
I assume the dataset names for apm-server and fleet-server both use |
Why is the dataset for logs called |
What about the |
Yes, I just hadn't pushed the most recent change yet as I'm still trying to find some issues around why not all are recognised properly. I pushed the most recent changes but it is not fully working yet. @andrewkroh can you have a quick look if this is what you meant? |
Yes, that is what I was expected assuming there is coverage for all the processes Michal listed above. 👍 Not sure why that one just has One other concern I have is around indexing the structured logs from Beats. There could be mapping conflicts since there is no guarantee as to what a structured log field may contain (e.g. calls to My personal approach to this when collecting Beat logs is to decode the message into |
It seems we ingest logs into An alternative instead of using the json flattened trick could be to just not index all the other fields and use runtime fields if query on it is needed? |
I prefer convention used in naming |
Relates: #953 |
Reminder for myself: Currently still missing are endpoint and auditbeat logs and metrics. |
I'm making progress here. By now all processes/data_streams with metrics and logs should be part of the package.
|
Currently getting the following error on CI, investigating:
I'll mark the PR as ready for review to get more eyes / opinions on it. |
That sounds like a safe approach to begin with for these log data streams. 👍 |
@jen-huang @afgomez Do you have any idea why we get the following error on CI? |
List of data streams was missing created by Elastic Agent for monitoring. push logs one too add endpoint and auditbeat cleanup container metrics update changelog update changelog and formatting rename endpoint to endpoint_security
0a7467a
to
0948aad
Compare
I incremented the version to 1.1.0 on this as 1.0 has been released and changed the compatibility to 7.15 so we can get this out quickly for testing. |
@nchaulet During development I hit the following error:
I found now the reason and it was that I had specified the same data stream twice. Of course this is a bug in the package and I'm glad Kibana / CI complained but I'm a bit surprised by the error. Is this expected? There is also an other error I saw (unfortunately did not copy) that told me foo@custom template already exists which I think was for the duplicated data stream. As @Custom for an upgrade always exists I was wondering if this could be a potential issue? |
This one should be ready for a round of reviews. It puts the minimal mappings in place and nothing more. Over time we should add more mappings for the ones we need. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, but I didn't analyze all fields.
elasticsearch: | ||
index_template: | ||
mappings: | ||
dynamic: false |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm curious: is it enabled by default?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, it is enabled by default in all data streams. There were some discussions in the past around turning this off especially now with runtime fields around.
@@ -0,0 +1,12 @@ | |||
- name: data_stream.type |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@mtojek Sorry, only stumbled over this after merging. Where did we land on who is responsible to set the type? Will the tooling do this or Fleet?
What about module and dataset? There was never an Elastic Agent module does it mean it can be skipped? @andrewkroh ?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What about module and dataset?
There's no requirement to set them. I think the cost is really low since they are constant_keyword so I personally would set them to have consistency.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If I add module here, what would you set the value here?
…1298) Adding all processes as data streams.
List of data streams was missing created by Elastic Agent for monitoring.