forked from bottlerocket-os/bottlerocket-update-operator
-
Notifications
You must be signed in to change notification settings - Fork 0
/
.env
27 lines (23 loc) · 1.32 KB
/
.env
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
# Example .env file
BRUPOP_CONTAINER_IMAGE=public.ecr.aws/bottlerocket/bottlerocket-update-operator:v0.2.2
# If testing against a private image registry, you can set the pull secret to fetch images.
# This can likely remain as `brupop` so long as you run something like the following:
# kubectl create secret docker-registry brupop \
# --docker-server 109276217309.dkr.ecr.us-west-2.amazonaws.com \
# --docker-username=AWS \
# --docker-password=$(aws --region us-west-2 ecr get-login-password) \
# --namespace=brupop-bottlerocket-aws
# BRUPOP_CONTAINER_IMAGE_PULL_SECRET=brupop
# External load balancer setting.
# When EXCLUDE_FROM_LB_WAIT_TIME_IN_SEC is set to positive value
# brupop will exclude the node from load balancer and
# wait for EXCLUDE_FROM_LB_WAIT_TIME_IN_SEC seconds before draining node.
EXCLUDE_FROM_LB_WAIT_TIME_IN_SEC=0
# Concurrent update nodes setting.
# When MAX_CONCURRENT_UPDATE is set to a positive integer value,
# brupop will concurrently update max MAX_CONCURRENT_UPDATE nodes.
# When MAX_CONCURRENT_UPDATE is set to "unlimited",
# brupop will concurrently update all nodes with respecting `PodDisruptionBudgets`
# Note: the "unlimited" option does not work well with `EXCLUDE_FROM_LB_WAIT_TIME_IN_SEC`
# option, which could potential exclude all nodes from load balancer at the same time.
MAX_CONCURRENT_UPDATE=1