You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We should consolidate these to a single async fn Tenant::shutdown.
Initial plan for this is to change the task_mgr::spawn API such that its callers become responsible for keeping track of the spawned task, e.g., by returning a wrapped tokio::task::JoinHandle.
For tenant shutdown, this means the async fn Tenant::shutdown will then join these join handles, after signalling need for cancellation to these tasks in some way.
We have 2 ways of tenant shutdown, we should have just one.
Changes are mostly mechanical simple refactorings.
Added `warn!` on the "shutdown all remaining tasks" should trigger test
failures in the between time of not having solved the "tenant/timeline
owns all spawned tasks" issue.
Cc: #4327.
We have 2 ways of tenant shutdown, we should have just one.
Changes are mostly mechanical simple refactorings.
Added `warn!` on the "shutdown all remaining tasks" should trigger test
failures in the between time of not having solved the "tenant/timeline
owns all spawned tasks" issue.
Cc: #4327.
Currently we have different ways to shutdown a tenant, at least the following:
pageserver::tenant::mgr::shutdown_all_tenants
pageserver::shutdown_pageserver
pageserver::tenant::mgr::remove_tenant_from_memory
We should consolidate these to a single
async fn Tenant::shutdown
.Initial plan for this is to change the
task_mgr::spawn
API such that its callers become responsible for keeping track of the spawned task, e.g., by returning a wrappedtokio::task::JoinHandle
.For tenant shutdown, this means the
async fn Tenant::shutdown
will then join these join handles, after signalling need for cancellation to these tasks in some way.Follow-up noticed while implementing this:
The text was updated successfully, but these errors were encountered: