From 04889e66db524bd7e4b29d8b80c09e40e12b05fe Mon Sep 17 00:00:00 2001
From: Niels de Vos <ndevos@redhat.com>
Date: Tue, 26 Jul 2022 10:45:06 +0200
Subject: [PATCH] ci: verify that Ceph Mgr is running

The Ceph v17.2.2 container-image fails to start Ceph Mgr. This causes
issues while the e2e test suite is running. It is better to check if
Ceph Mgr is available, before continuing with the rest of the CI job.

Updates: #3259
Signed-off-by: Niels de Vos <ndevos@redhat.com>
---
 scripts/rook.sh | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/scripts/rook.sh b/scripts/rook.sh
index 4d732c270..1b650eaa5 100755
--- a/scripts/rook.sh
+++ b/scripts/rook.sh
@@ -86,6 +86,9 @@ function deploy_rook() {
 		check_ceph_cluster_health
 	fi
 
+	# Make sure Ceph Mgr is running
+	check_ceph_mgr
+
 	# Check if CephFileSystem is empty
 	if ! kubectl_retry -n rook-ceph get cephfilesystems -oyaml | grep 'items: \[\]' &>/dev/null; then
 		check_mds_stat
@@ -166,6 +169,22 @@ function check_ceph_cluster_health() {
 	echo ""
 }
 
+function check_ceph_mgr() {
+	for ((retry = 0; retry <= ROOK_DEPLOY_TIMEOUT; retry = retry + 5)); do
+		echo "Waiting for Ceph Mgr... ${retry}s" && sleep 5
+
+		MGR_POD=$(kubectl_retry -n rook-ceph get pods -l app=rook-ceph-mgr -o jsonpath='{.items[0].metadata.name}')
+		MGR_POD_STATUS=$(kubectl_retry -n rook-ceph get pod "$MGR_POD" -ojsonpath='{.status.phase}')
+		[[ "$MGR_POD_STATUS" = "Running" ]] && break
+	done
+
+	if [ "$retry" -gt "$ROOK_DEPLOY_TIMEOUT" ]; then
+		echo "[Timeout] Ceph Mgr is not running (timeout)"
+		return 1
+	fi
+	echo ""
+}
+
 function check_mds_stat() {
 	for ((retry = 0; retry <= ROOK_DEPLOY_TIMEOUT; retry = retry + 5)); do
 		FS_NAME=$(kubectl_retry -n rook-ceph get cephfilesystems.ceph.rook.io -ojsonpath='{.items[0].metadata.name}')