Refreshing dacha2 cache after split brain

[chkp-liordan]

we have deployment of 3 pods for dacha. we see that two pod with updated data and one is behind.
there is API to refresh that data?

[rvowles]

heya - dacha2 is an empty cache setup, so killing the pod is the best option and it will gradually refill as it is asked for api keys.

[chkp-liordan]

just FYI, the problem was in NATS. it wasn't deployed as a cluster (by mistake). therefor dacha was not updated. after redeploy NATS as a cluster it works.
how to check Dacha using the api (

featurehub/backend/dacha-api/dacha-api.yaml

Line 7 in a125472

/v2/environment/{eId}/{serviceAccountKey}:

):

1. port-forward to each pod
2. curl http://localhost:8601/v2/environment/"<client_key>" | md5
3. check for each pod if md5 is different.