Mirrors/coredns
1
0
Fork 0
mirror of https://github.com/coredns/coredns.git synced 2026-10-26 20:25:43 -04:00
Code Issues Packages Projects Releases Wiki Activity

plugin/forward: add startup health checks with start_failed (#8663)

Browse Source 
* plugin/forward: add startup health checks with start_failed

Signed-off-by: houyuwushang <180804215+houyuwushang@users.noreply.github.com>

* test: separate CoreDNS and third-party imports

Signed-off-by: houyuwushang <180804215+houyuwushang@users.noreply.github.com>

---------

Signed-off-by: houyuwushang <180804215+houyuwushang@users.noreply.github.com>
This commit is contained in:
houyuwushang
2026-10-21 11:34:41 +08:00
committed by GitHub
parent 0fe7b08fbb
commit 3cbe6b5269
8 changed files with 250 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
plugin/forward/README.md
View File

@@ -57,6 +57,7 @@ forward FROM TO... {
tls_servername NAME
policy random|round_robin|sequential
health_check DURATION [no_rec] [domain FQDN]
start_failed
max_concurrent MAX
next RCODE_1 [RCODE_2] [RCODE_3...]
failfast_all_unhealthy_upstreams
@@ -77,6 +78,12 @@ forward FROM TO... {
* `max_fails` is the number of subsequent failed health checks that are needed before considering
an upstream to be down. If 0, the upstream will never be marked as down (nor health checked).
Default is 2.
* `start_failed` starts upstreams in an unhealthy state and immediately runs health checks using
the configured `health_check` settings. Each upstream becomes healthy after its first successful
check. This option is disabled by default and cannot be combined with `max_fails 0`.
Checks run asynchronously; startup does not wait for them. While all upstreams are unhealthy,
requests follow the usual all-unhealthy policy. Combine with `failfast_all_unhealthy_upstreams`
to return SERVFAIL until an upstream passes a check instead of trying a random upstream.
* `max_connect_attempts` caps the total number of upstream connect attempts
performed for a single incoming DNS request. The default cap is twice the number of
configured upstreams, allowing two complete passes when all upstreams are healthy.