Experimental replication parameters #725
irevoire
started this conversation in
Experimental features
Replies: 1 comment
-
Update: v1.7 has been released 🐇 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Meilisearch v1.7.0 (released on March 11, 2024) introduces a new CLI parameter that changes the behavior of Meilisearch in a few ways that should help you run meilisearch in a cluster by externalizing the task queue.
Enable the CLI flags
What this CLI flag does
This CLI flag changes three big things in the engine around the task queue:
Disable the auto-deletion of tasks
By default, Meilisearch will auto-delete your latest tasks finished (succeeded, failed, or canceled) when you reach 1M tasks in your task queue.
When using this flag, you will be in charge of auto-deleting the tasks. Meilisearch won’t do it anymore.
But you must do it because once the task queue reaches 10GiB, Meilisearch will stop everyone from registering tasks.
Chose your own task ID
When the
--experimental-replication-parameters
flag is enabled, you can specify your own task ID with theTaskId
header instead of letting Meilisearch choose a new one, but there are a few rules around that:TaskId
header (i.e.,TaskId: 0
)TaskId: 1000
, then you won’t be able to register the task id1
or even1000
; your task ID must be strictly> 1000
.Dry-register a task
When the
--experimental-replication-parameters
flag is enabled, you also gain the ability to « dry register » a task.That means, when you try to register a task, if you specify the header
DryRun: true
in your request, Meilisearch won’t register your task but will answer with a 200 and a validated task.Beta Was this translation helpful? Give feedback.
All reactions