Migration from 2.2-alpine to 2.3.1-alpine broke healthcheck configuration

[msartor92]

I'm getting tons of healthcheck request after upgrade docker kong from 2.2-alpine to 2.3.1-alpine
I've configured healthcheck to run each 10 seconds. Normally it goes well, but after migration requests seems that is goes on loop and crash application http connector.
Please find log below.

[01/Feb/2021:10:14:04 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:14:14 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:14:24 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:14:34 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:14:44 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:14:54 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:15:04 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:15:14 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:15:24 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:15:34 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:15:44 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:05 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:05 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:05 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:06 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:07 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4
[01/Feb/2021:10:16:08 +0100] "GET /ping HTTP/1.0" 200 4

I've disable passive healthcheck for this upstream via konga, and after reconfigure with previous parameters. That fix the problem for few hours. But after a couple of hour it restart
This happens on multiple upstream

Healthcheck configuration:

healthchecks:
active:
concurrency: 10
healthy:
http_statuses:
- 200
- 302
interval: 10
successes: 3
http_path: /DealerWebapp/backdoor/ping
https_verify_certificate: false
type: http
timeout: 1
unhealthy:
http_failures: 3
http_statuses:
- 429
- 404
- 500
- 501
- 502
- 503
- 504
- 505
tcp_failures: 3
timeouts: 3
interval: 5
passive:
healthy:
http_statuses:
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 226
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
successes: 2
type: http
unhealthy:
http_failures: 3
http_statuses:
- 429
- 500
- 503
tcp_failures: 3
timeouts: 5
threshold: 0

Additional Details & Logs

• Kong version 2.3.1-alpine
• Docker version 20.10.2, build 2291f61
• OS Version Centos 7

[msartor92]

After a rollback to 2.2.1-alpine all seems back to normal.
Bug affect also version 2.3.2-alpine.

[locao]

Hi @msartor92! Thanks for your report. We have this fixed in Kong/lua-resty-healthcheck#67, it should be out when we release Kong 2.3.3.