Jaeger v2 ILM #9153
Replies: 1 comment
|
Jaeger v2 still ships the same index lifecycle tools from v1 — they just moved into the unified binary. Here's the current state:
It supports
That said, the project is actively working toward Elasticsearch Data Streams support (RFC 0004), which would eliminate the need for external cron jobs, rollover init steps, and manual alias management entirely. The config refactoring (Phase 1) is done, but the actual data stream write/read path (Phase 2) is still in progress [5]. Once that lands, you'd configure For now, your best options are:
To reply, just mention @dosu. Share context across your team and agents. Try Dosu. |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Jaeger v1 → v2 migration: Elasticsearch ILM management
Hi everyone,
We are migrating from Jaeger v1 to Jaeger v2 and would like to get some guidance on the recommended approach for trace data retention and Elasticsearch ILM management.
In Jaeger v1, we had a Jaeger-provided cronjob that handled Elasticsearch ILM configuration for trace indices. With Jaeger v2, we would like to understand what the recommended replacement approach is.
Should we manage ILM directly in Elasticsearch by creating our own:
Or does Jaeger v2 provide any built-in/default mechanism for managing trace index lifecycle?
For reference, we are currently deploying Jaeger v2.20.0 with the Elasticsearch storage backend and custom index prefixes.
Our current deployment configuration:
Any recommendations, migration guidance, or examples of the preferred production setup would be greatly appreciated.
Thanks!
All reactions