addon¶
Inspect and update the managed EKS add-ons (vpc-cni, coredns, kube-proxy,
and others) on a cluster. List shows installed versions and status, describe
drills into one add-on, and update rolls a single add-on or every add-on
(--all) to a compatible version with optional health gating and waiting.
The cluster is a positional on each subcommand, or --cluster/-c, falling back
to the active context.
list¶
List the managed EKS add-ons installed on a cluster along with their current
version, status, and (with --show-health) a health badge.
Flags¶
| Flag | Description |
|---|---|
--cluster, -c |
EKS cluster name or pattern (or pass as positional) |
--show-health, -H |
Include a health mapping/badge in table output |
--format, -o |
table (default), json, yaml, plain |
--watch, -w |
Re-run and redraw every --watch-interval until interrupted |
--watch-interval |
Refresh interval for --watch (default 10s) |
--timeout, -t |
Operation timeout (env REFRESH_TIMEOUT) |
Watch an update land
refresh addon list my-cluster --watch keeps the listing live, so you can
watch an add-on update progress without re-running the command.
Examples¶
refresh addon list my-cluster
refresh addon list my-cluster --show-health
refresh addon list my-cluster -o plain
refresh addon list my-cluster --watch --watch-interval 5s
describe¶
Detailed information for one add-on: its version, status, and configuration.
describe has the alias get. The add-on name may be the second positional or
--addon/-a, and a case-insensitive substring is resolved against the
installed add-ons.
Flags¶
| Flag | Description |
|---|---|
--cluster, -c |
EKS cluster name or pattern (or pass as positional) |
--addon, -a |
Add-on name (e.g. vpc-cni); or pass as second positional |
--format, -o |
table (default), json, yaml, plain |
--timeout, -t |
Operation timeout (env REFRESH_TIMEOUT) |
Examples¶
update¶
Update a single managed add-on to a target version, or with --all update every
add-on in the cluster to its latest compatible version.
For a single add-on, pass the add-on name and an optional version (the third
positional or --version, defaulting to latest). The command exits non-zero
if any add-on update fails.
Flags¶
| Flag | Description |
|---|---|
--cluster, -c |
EKS cluster name or pattern (or pass as positional) |
--addon, -a |
Add-on name (or pass as second positional) |
--version |
Target version or latest (default; or pass as third positional) |
--all |
Update every add-on in the cluster to its latest version |
--health-check |
Verify the add-on is ACTIVE and version-compatible before updating |
--dry-run, -d |
Preview without applying changes |
--wait |
Wait for each update to complete |
--wait-timeout |
Per-add-on wait timeout, with --wait (default 5m) |
--parallel, -p |
(--all only) Update add-ons in parallel |
--dependency-order |
(--all only) Update in dependency-safe order: vpc-cni → coredns/kube-proxy → others |
--skip, -s |
(--all only) Skip specific add-ons (repeatable) |
--format, -o |
table (default), json, yaml, plain |
--timeout, -t |
Operation timeout (default 10m; env REFRESH_TIMEOUT) |
--all-only flags
--parallel, --dependency-order, and --skip apply only with --all. On
a single-add-on update they're ignored with a warning. --parallel and
--dependency-order are mutually exclusive (parallel defeats ordering).
A single-add-on update honors -o json|yaml for a machine-readable result.
Examples¶
# vpc-cni -> latest
refresh addon update my-cluster vpc-cni
# Pin a version
refresh addon update my-cluster coredns v1.11.1
# Preview only
refresh addon update my-cluster vpc-cni --dry-run
# All add-ons, dependency-safe order, waiting for each to settle
refresh addon update my-cluster --all --dependency-order --wait
# All add-ons in parallel, skipping vpc-cni
refresh addon update my-cluster --all --skip vpc-cni --parallel