John Spray e34059cd18 pageserver: increase DEFAULT_MAX_WALRECEIVER_LSN_WAL_LAG (#6970) ...

## Problem

At high ingest rates, pageservers spuriously disconnect from safekeepers
because stats updates don't come in frequently enough to keep the
broker/safekeeper LSN delta under the wal lag limit.

## Summary of changes

- Increase DEFAULT_MAX_WALRECEIVER_LSN_WAL_LAG from 10MiB to 1GiB. This
should be enough for realistic per-timeline throughputs.

2024-03-01 16:49:37 +00:00

..

benches

bench_walredo: use tokio multi-threaded runtime (#6743)

2024-02-16 16:31:54 +01:00

client

Implement a sharded time travel recovery endpoint (#6821)

2024-02-21 16:35:37 +01:00

compaction

Fix warnings and compile errors on nightly (#6886)

2024-03-01 17:14:19 +01:00

ctl

pageserver: optimise disk io for vectored get (#6780)

2024-02-28 12:06:00 +00:00

pagebench

pageserver: optimise disk io for vectored get (#6780)

2024-02-28 12:06:00 +00:00

src

pageserver: increase DEFAULT_MAX_WALRECEIVER_LSN_WAL_LAG (#6970)

2024-03-01 16:49:37 +00:00

test_data

Add walingest test (#5892)

2023-11-28 12:50:53 -05:00

Cargo.toml

Add new compaction abstraction, simulator, and implementation. (#6830)

2024-02-27 17:15:46 +01:00