Relevant telegraf.conf

[[inputs.prometheus]]
monitor_kubernetes_pods = true
metric_version = 1
monitor_kubernetes_pods_method = "settings+annotations"
monitor_kubernetes_pods_port = 7354
pod_scrape_interval = 60
content_length_limit = "4096KB"

Logs from Telegraf

2026-02-04T14:08:41.094Z | 2026-02-04T14:08:41Z I! [agent] Config: Interval:1m0s, Quiet:false, Hostname:"", Flush Interval:10s
2026-02-04T14:08:41.097Z | 2026-02-04T14:08:41Z W! [inputs.prometheus::myapp] Error while attempting to watch pod: handler {0x452a5a0 0x452a380 0x452a320} was not added to shared informer because it has stopped already

System info

Telegraf 1.37

Docker

No response

Steps to reproduce

1. have prometheus plugin in monitor_kubernetes_pods = true mode
2. (possibly) have many (we have hundreds) pods that plugin would be targeting
3. have telegraf with --watch-config poll (I suspect manual reloading with a signal will trigger it too)

Expected behavior

reload succeeds and metric collection continues

Actual behavior

plugin instances with kubernetes connection stop collecting metrics

Additional info

No response