storcon: do not update observed state on node activation (#11155)

## Problem

When a node becomes active, we query its locations and update the
observed state in-place.
This can race with the observed state updates done when processing
reconcile results.

## Summary of changes

The argument for this reconciliation step is that is reduces the need
for background reconciliations.
I don't think is actually true anymore. There's two cases.

1. Restart of node after drain. Usually the node does not go through the
offline state here, so observed locations
were not marked as none. In any case, there should be a handful of
shards max on the node since we've just drained it.
2. Node comes back online after failure or network partition. When the
node is marked offline, we reschedule everything away from it. When it
later becomes active, the previous observed location is extraneous and
requires a reconciliation anyway.

Closes https://github.com/neondatabase/neon/issues/11148
This commit is contained in:
Vlad Lazar
2025-03-12 15:31:28 +00:00
committed by GitHub
parent c7717c85c7
commit 02a83913ec
5 changed files with 48 additions and 12 deletions

View File

@@ -36,7 +36,9 @@ use pageserver_api::config::{
use pageserver_api::controller_api::{
NodeAvailabilityWrapper, PlacementPolicy, TenantCreateRequest,
};
use pageserver_api::models::{ShardParameters, TimelineCreateRequest, TimelineInfo};
use pageserver_api::models::{
ShardParameters, TenantConfigRequest, TimelineCreateRequest, TimelineInfo,
};
use pageserver_api::shard::{ShardCount, ShardStripeSize, TenantShardId};
use postgres_backend::AuthType;
use postgres_connection::parse_host_port;
@@ -1129,12 +1131,16 @@ async fn handle_tenant(subcmd: &TenantCmd, env: &mut local_env::LocalEnv) -> any
let tenant_id = get_tenant_id(args.tenant_id, env)?;
let tenant_conf: HashMap<_, _> =
args.config.iter().flat_map(|c| c.split_once(':')).collect();
let config = PageServerNode::parse_config(tenant_conf)?;
pageserver
.tenant_config(tenant_id, tenant_conf)
let req = TenantConfigRequest { tenant_id, config };
let storage_controller = StorageController::from_env(env);
storage_controller
.set_tenant_config(&req)
.await
.with_context(|| format!("Tenant config failed for tenant with id {tenant_id}"))?;
println!("tenant {tenant_id} successfully configured on the pageserver");
println!("tenant {tenant_id} successfully configured via storcon");
}
}
Ok(())

View File

@@ -14,7 +14,7 @@ use pageserver_api::controller_api::{
NodeConfigureRequest, NodeDescribeResponse, NodeRegisterRequest, TenantCreateRequest,
TenantCreateResponse, TenantLocateResponse,
};
use pageserver_api::models::{TimelineCreateRequest, TimelineInfo};
use pageserver_api::models::{TenantConfigRequest, TimelineCreateRequest, TimelineInfo};
use pageserver_api::shard::TenantShardId;
use pageserver_client::mgmt_api::ResponseErrorMessageExt;
use postgres_backend::AuthType;
@@ -878,4 +878,9 @@ impl StorageController {
)
.await
}
pub async fn set_tenant_config(&self, req: &TenantConfigRequest) -> anyhow::Result<()> {
self.dispatch(Method::PUT, "v1/tenant/config".to_string(), Some(req))
.await
}
}