Merge pull request #4437 from pmtk/healthcheck-docs

USHIFT-4288: Update greenboot.mds with healthcheck

Author: openshift-merge-bot[bot]

docs/config/busybox_running_check.sh

@@ -2,29 +2,10 @@
 set -e
 
 SCRIPT_NAME=$(basename $0)
-PODS_NS_LIST=(busybox)
-PODS_CT_LIST=(1      )
 
 # Source the MicroShift health check functions library
 source /usr/share/microshift/functions/greenboot.sh
 
-# Set the exit handler to log the exit status
-trap 'script_exit' EXIT
-
-# The script exit handler logging the FAILURE or FINISHED message depending
-# on the exit status of the last command
-#
-# args: None
-# return: None
-function script_exit() {
-    [ "$?" -ne 0 ] && status=FAILURE || status=FINISHED
-    echo $status
-}
-
-#
-# Main
-#
-
 # Exit if the current user is not 'root'
 if [ $(id -u) -ne 0 ] ; then
     echo "The '${SCRIPT_NAME}' script must be run with the 'root' user privileges"
@@ -33,36 +14,7 @@ fi
 
 echo "STARTED"
 
-# Exit if the MicroShift service is not enabled
-if [ $(systemctl is-enabled microshift.service 2>/dev/null) != "enabled" ] ; then
-    echo "MicroShift service is not enabled. Exiting..."
-    exit 0
-fi
-
 # Set the wait timeout for the current check based on the boot counter
 WAIT_TIMEOUT_SECS=$(get_wait_timeout)
 
-# Wait for pod images to be downloaded
-for i in ${!PODS_NS_LIST[@]}; do
-    CHECK_PODS_NS=${PODS_NS_LIST[$i]}
-
-    echo "Waiting ${WAIT_TIMEOUT_SECS}s for pod image(s) from the '${CHECK_PODS_NS}' namespace to be downloaded"
-    wait_for ${WAIT_TIMEOUT_SECS} namespace_images_downloaded
-done
-
-# Wait for pods to enter ready state
-for i in ${!PODS_NS_LIST[@]}; do
-    CHECK_PODS_NS=${PODS_NS_LIST[$i]}
-    CHECK_PODS_CT=${PODS_CT_LIST[$i]}
-
-    echo "Waiting ${WAIT_TIMEOUT_SECS}s for ${CHECK_PODS_CT} pod(s) from the '${CHECK_PODS_NS}' namespace to be in 'Ready' state"
-    wait_for ${WAIT_TIMEOUT_SECS} namespace_pods_ready
-done
-
-# Verify that pods are not restarting
-for i in ${!PODS_NS_LIST[@]}; do
-    CHECK_PODS_NS=${PODS_NS_LIST[$i]}
-
-    echo "Checking pod restart count in the '${CHECK_PODS_NS}' namespace"
-    namespace_pods_not_restarting ${CHECK_PODS_NS}
-done
+/usr/bin/microshift healthcheck -v=2 --timeout="${WAIT_TIMEOUT_SECS}s" --namespace busybox --deployments busybox-deployment

docs/contributor/greenboot.md

@@ -46,25 +46,30 @@ sudo journalctl -o cat -u greenboot-healthcheck.service
 
 ### Health Check Implementation
 
-The script utilizes the MicroShift health check functions that are available
-in the `/usr/share/microshift/functions/greenboot.sh` file to reuse procedures
-already implemented for the MicroShift core services. These functions need a
-definition of the user workload namespaces and the expected count of pods.
-
-```bash
-PODS_NS_LIST=(busybox)
-PODS_CT_LIST=(1      )
-```
+The script utilizes the MicroShift `healthcheck` command that is part of the
+`microshift` binary to reuse procedures already implemented for the MicroShift
+core services. 
 
 The script starts by running sanity checks to verify that it is executed from
-the `root` account and that the MicroShift service is enabled.
-
-Finally, the MicroShift health check functions are called to perform the
-following actions:
-- Get a wait timeout of the current boot cycle for the `wait_for` function
-- Call the `namespace_images_downloaded` function to wait until pod images are available
-- Call the `namespace_pods_ready` function to wait until pods are ready
-- Call the `namespace_pods_not_restarting` function to verify pods are not restarting
+the `root` account.
+
+Then, it executes `microshift healthcheck` command with following options:
+- `-v=2` to increase verbosity of the output
+- `--timeout="${WAIT_TIMEOUT_SECS}s"` to override default 300s timeout value
+- `--namespace busybox` to specify the Namespace of the workloads
+- `--deployments busybox-deployment` to specify Deployment to check the readiness of
+
+Internally, `microshift healthcheck` checks if workload of the provided type exists and verifies its
+status for the specified timeout duration, so the amount of ready replicas (Pods) matches the expected amount.
+
+`microshift healthcheck` also accepts other parameters to specify other kinds
+of workload: `--daemonsets` and `--statefulsets`. These options take
+comma-delimited list of resources, e.g.: `--daemonsets ovnkube-master,ovnkube-node`.
+
+Alternatively, a `--custom` option can be used with a JSON string, for example:
+```
+microshift healthcheck --custom '{"openshift-storage":{"deployments": ["lvms-operator"], "daemonsets": ["vg-manager"]}, "openshift-ovn-kubernetes":{"daemonsets": ["ovnkube-master", "ovnkube-node"]}}'
+```
 
 ## MicroShift Service Failure
 

docs/user/greenboot.md

@@ -49,16 +49,12 @@ following commands:
 Exiting the health check script with a non-zero status will have the boot declared
 as failed. The following validations are performed by the script.
 
-|Validation                                           |Pass  |Fail  |
-|-----------------------------------------------------|------|------|
-|Check the script runs with 'root' permissions        |Next  |exit 0|
-|Check microshift.service is enabled                  |Next  |exit 0|
-|Wait for microshift.service to be active (!failed)   |Next  |exit 1|
-|Wait for Kubernetes API health endpoints to be OK    |Next  |exit 1|
-|Wait for any Pod to start                            |Next  |exit 1|
-|For each core namespace, wait for images to be pulled|Next  |exit 1|
-|For each core namespace, wait for Pods to be ready   |Next  |exit 1|
-|For each core namespace, check Pods not restarting   |exit 0|exit 1|
+| Validation                                                  | Pass | Fail   |
+|-------------------------------------------------------------|------|--------|
+| Check the script runs with 'root' permissions               | Next | exit 0 |
+| Check microshift.service is enabled                         | Next | exit 0 |
+| Wait for microshift.service to be active (!failed)          | Next | exit 1 |
+| For each core namespace, wait for readiness of the workload | Next | exit 1 |
 
 The pre-rollback script runs the `sudo microshift-cleanup-data --ovn` command
 to prepare the system for a potential software downgrade.