mirror of
https://github.com/neondatabase/neon.git
synced 2026-01-15 09:22:55 +00:00
17672c88ff
When endpoint is stopped in immediate mode and started again there is a chance of old connection delivering some WAL to safekeepers after second start checked need for sync-safekeepers and thus grabbed basebackup LSN. It makes basebackup unusable, so compute panics. Avoid flakiness by waiting for walreceivers on safekeepers to be gone in such cases. A better way would be to bump term on safekeepers if sync-safekeepers is skipped, but it needs more infrastructure. ref https://github.com/neondatabase/neon/issues/9079
21 lines
698 B
Python
21 lines
698 B
Python
from fixtures.common_types import TenantId, TimelineId
|
|
from fixtures.log_helper import log
|
|
from fixtures.safekeeper.http import SafekeeperHttpClient
|
|
from fixtures.utils import wait_until
|
|
|
|
|
|
def wait_walreceivers_absent(
|
|
sk_http_cli: SafekeeperHttpClient, tenant_id: TenantId, timeline_id: TimelineId
|
|
):
|
|
"""
|
|
Wait until there is no walreceiver connections from the compute(s) on the
|
|
safekeeper.
|
|
"""
|
|
|
|
def walreceivers_absent():
|
|
status = sk_http_cli.timeline_status(tenant_id, timeline_id)
|
|
log.info(f"waiting for walreceivers to be gone, currently {status.walreceivers}")
|
|
assert len(status.walreceivers) == 0
|
|
|
|
wait_until(30, 0.5, walreceivers_absent)
|