Arseny Sher 729ac38ea8 Centralize suspending/resuming timeline activity on safekeepers. ...

Timeline is active whenever there is at least 1 connection from compute or
pageserver is not caught up. Currently 'active' means callmemaybes are being
sent.

Fixes race: now suspend condition checking and callmemaybe unsubscribe happen
under the same lock.

2022-02-03 02:34:10 +03:00

..

bin

Refactoring: use anyhow::Context's methods where possible

2022-01-19 16:33:48 +03:00

http

Prefix per-cluster directory with ztenant_id in safekeeper.

2022-01-27 17:22:07 +03:00

callmemaybe.rs

walkeeper: use named type as a key in callmemaybe subscriptions hashmap

2022-01-24 17:20:15 +03:00

handler.rs

Prefix per-cluster directory with ztenant_id in safekeeper.

2022-01-27 17:22:07 +03:00

json_ctrl.rs

Refactor Postgres command parsing in safekeeper.

2021-12-24 15:48:13 +03:00

lib.rs

Prefix per-cluster directory with ztenant_id in safekeeper.

2022-01-27 17:22:07 +03:00

receive_wal.rs

Centralize suspending/resuming timeline activity on safekeepers.

2022-02-03 02:34:10 +03:00

s3_offload.rs

Switch safekeeper from log to tracing logging.

2021-12-09 06:57:46 +03:00

safekeeper.rs

Extend replication protocol with ZentihFeedback message

2022-01-27 11:20:45 +03:00

send_wal.rs

Centralize suspending/resuming timeline activity on safekeepers.

2022-02-03 02:34:10 +03:00

timeline.rs

Centralize suspending/resuming timeline activity on safekeepers.

2022-02-03 02:34:10 +03:00

upgrade.rs

Add safekeeper --dump-control-file option.

2022-01-12 19:47:24 +03:00

wal_service.rs

Make safekeeper namings more consistent with reality.

2021-12-21 13:24:23 +03:00