mirror of
https://github.com/neondatabase/neon.git
synced 2026-01-08 05:52:55 +00:00
storcon: automatically clear Pause/Stop scheduling policies to enable detaches (#10011)
## Problem We saw a tenant get stuck when it had been put into Pause scheduling mode to pin it to a pageserver, then it was left idle for a while and the control plane tried to detach it. Close: https://github.com/neondatabase/neon/issues/9957 ## Summary of changes - When changing policy to Detached or Secondary, set the scheduling policy to Active. - Add a test that exercises this - When persisting tenant shards, set their `generation_pageserver` to null if the placement policy is not Attached (this enables consistency checks to work, and avoids leaving state in the DB that could be confusing/misleading in future)
This commit is contained in:
@@ -3230,3 +3230,55 @@ def test_multi_attached_timeline_creation(neon_env_builder: NeonEnvBuilder, migr
|
||||
# Always disable 'pause' failpoints, even on failure, to avoid hanging in shutdown
|
||||
env.storage_controller.configure_failpoints((migration_failpoint.value, "off"))
|
||||
raise
|
||||
|
||||
|
||||
@run_only_on_default_postgres("Postgres version makes no difference here")
|
||||
def test_storage_controller_detached_stopped(
|
||||
neon_env_builder: NeonEnvBuilder,
|
||||
):
|
||||
"""
|
||||
Test that detaching a tenant while it has scheduling policy set to Paused or Stop works
|
||||
"""
|
||||
|
||||
remote_storage_kind = s3_storage()
|
||||
neon_env_builder.enable_pageserver_remote_storage(remote_storage_kind)
|
||||
|
||||
neon_env_builder.num_pageservers = 1
|
||||
|
||||
env = neon_env_builder.init_configs()
|
||||
env.start()
|
||||
virtual_ps_http = PageserverHttpClient(env.storage_controller_port, lambda: True)
|
||||
|
||||
tenant_id = TenantId.generate()
|
||||
env.storage_controller.tenant_create(
|
||||
tenant_id,
|
||||
shard_count=1,
|
||||
)
|
||||
|
||||
assert len(env.pageserver.http_client().tenant_list_locations()["tenant_shards"]) == 1
|
||||
|
||||
# Disable scheduling: ordinarily this would prevent the tenant's configuration being
|
||||
# reconciled to pageservers, but this should be overridden when detaching.
|
||||
env.storage_controller.allowed_errors.append(".*Scheduling is disabled by policy.*")
|
||||
env.storage_controller.tenant_policy_update(
|
||||
tenant_id,
|
||||
{"scheduling": "Stop"},
|
||||
)
|
||||
|
||||
env.storage_controller.consistency_check()
|
||||
|
||||
# Detach the tenant
|
||||
virtual_ps_http.tenant_location_conf(
|
||||
tenant_id,
|
||||
{
|
||||
"mode": "Detached",
|
||||
"secondary_conf": None,
|
||||
"tenant_conf": {},
|
||||
"generation": None,
|
||||
},
|
||||
)
|
||||
|
||||
env.storage_controller.consistency_check()
|
||||
|
||||
# Confirm the detach happened
|
||||
assert env.pageserver.http_client().tenant_list_locations()["tenant_shards"] == []
|
||||
|
||||
Reference in New Issue
Block a user