Proxy often has error logs（no memory）

[J1a-wei]

We often have error logs recently, suggesting that we have no memory, but we have observed that PROXY memory has not accounted for the host of the PROXY memory through Promethos. Is there any problem?

[error] 1097#0: *720540820 [lua] events.lua:273: post(): worker-events: failed posting event "healthy" by "lua-resty-healthcheck [0dc6f45b-8f8d-40d2-a504-473544ee190b:ethereum-mainnet-api.web3-helm.8599.svc]"; no memory

prometheus.lua:839: log_error(): Error while setting 'memory_workers_lua_vms_bytes{pid="1104",kong_subsystem="http"}' to '160569064': 'no memory'

log_error(): Error while setting 'upstream_target_health{upstream="heco-mainnet-api-graphql.web3-helm.8545.svc",target="172.1.19.101:8545",address="172.1.19.101:8545",state="dns_error",subsystem="http"}' to '0': 'no memory'

Kong Version: 2.7.0 (Helm chart)

[J1a-wei]

grafana snapshot

The third picture seems to be limited to 200MIB. Where can I configure this parameter?

[randmonkey]

I think you could configure the resource limits of containers and pods.