mirror of
https://github.com/neondatabase/neon.git
synced 2026-10-27 05:56:39 +00:00
pageserver: remove trace_read_requests (#8338)
`trace_read_requests` is a per `Tenant`-object option. But the `handle_pagerequests` loop doesn't know which `Tenant` object (i.e., which shard) the request is for. The remaining use of the `Tenant` object is to check `tenant.cancel`. That check is incorrect [if the pageserver hosts multiple shards](https://github.com/neondatabase/neon/issues/7427#issuecomment-2220577518). I'll fix that in a future PR where I completely eliminate the holding of `Tenant/Timeline` objects across requests. See [my code RFC](https://github.com/neondatabase/neon/pull/8286) for the high level idea. Note that we can always bring the tracing functionality if we need it. But since it's actually about logging the `page_service` wire bytes, it should be a `page_service`-level config option, not per-Tenant. And for enabling tracing on a single connection, we can implement a `set pageserver_trace_connection;` option.
This commit is contained in:
@@ -302,17 +302,6 @@ pub struct TenantId(Id);
|
||||
|
||||
id_newtype!(TenantId);
|
||||
|
||||
/// Neon Connection Id identifies long-lived connections (for example a pagestream
|
||||
/// connection with the page_service). Is used for better logging and tracing
|
||||
///
|
||||
/// NOTE: It (de)serializes as an array of hex bytes, so the string representation would look
|
||||
/// like `[173,80,132,115,129,226,72,254,170,201,135,108,199,26,228,24]`.
|
||||
/// See [`Id`] for alternative ways to serialize it.
|
||||
#[derive(Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize, PartialOrd, Ord)]
|
||||
pub struct ConnectionId(Id);
|
||||
|
||||
id_newtype!(ConnectionId);
|
||||
|
||||
// A pair uniquely identifying Neon instance.
|
||||
#[derive(Debug, Clone, Copy, PartialOrd, Ord, PartialEq, Eq, Hash, Serialize, Deserialize)]
|
||||
pub struct TenantTimelineId {
|
||||
|
||||
Reference in New Issue
Block a user