You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While investigating some odd healthcheck behaviour during an upgrade from Kong 1.5.0 to Kong 2.4.1 I noticed that every time I patched an upstream (e.g. changing an attribute of the active healthchecks) I got an increasing number of event logs from the healthcheck module. I'm not sure why this is occurring nor whether it had any negative side effects. But I thought I would report it nonetheless as I suspect that this could be seen as a memory leak as it would seem like something is not being correctly disposed of.
Steps To Reproduce
Start Kong - create a single upstream with associated target and active healthchecks enabled
Observe the logs - notice the single target added event log
Patch the upstream - don't actually even need to change any value - just a simple curl -X PATCH http://localhost:8001//upstreams/$guid is sufficient
Observe the logs - notice that there are now 2 (or n depending on iteration) log events like:
...
[healthcheck] (mock) event: local cache cleared
[healthcheck] (mock) event: local cache cleared
...
[healthcheck] (mock) event: target added 'mock(172.26.128.4:3000)'
[healthcheck] (mock) event: target status 'mock(172.26.128.4:3000)' from 'false' to 'true'
[healthcheck] (mock) event: target added 'mock(172.26.128.4:3000)'
[healthcheck] (mock) event: target status 'mock(172.26.128.4:3000)' from 'false' to 'true'
Repeat steps 3 and 4 to your heart's content noticing the ever increasing number of logs
onematchfox
changed the title
Patching an upstream results in an ever increasing number of healthcheck _events_
Patching an upstream results in an ever increasing number of healthcheck events
May 31, 2021
Summary
Kong version:
2.4.1
(reproduced in1.5.0
as well)While investigating some odd healthcheck behaviour during an upgrade from Kong
1.5.0
to Kong2.4.1
I noticed that every time I patched an upstream (e.g. changing an attribute of the active healthchecks) I got an increasing number of event logs from thehealthcheck
module. I'm not sure why this is occurring nor whether it had any negative side effects. But I thought I would report it nonetheless as I suspect that this could be seen as a memory leak as it would seem like something is not being correctly disposed of.Steps To Reproduce
target added
event logcurl -X PATCH http://localhost:8001//upstreams/$guid
is sufficientAdditional Details & Logs
$ kong start --vv
)<KONG_PREFIX>/logs/error.log
)https://docs.konghq.com/latest/admin-api/#retrieve-node-information)
The text was updated successfully, but these errors were encountered: