From 8cd5bbc2dc11e4db035c23541edb81ed2dfb1851 Mon Sep 17 00:00:00 2001
From: Apurva Nisal <anisal@redhat.ocm>
Date: Mon, 8 Jun 2026 00:36:53 +0530
Subject: [PATCH 1/4] =?UTF-8?q?ETCD-704=20=E2=80=94=20cluster-restore.sh:?=
 =?UTF-8?q?=20move=20extra=20/var/lib/etcd=20files=20to=20backup?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When cluster-restore.sh runs the restore-pod path, it moves member/ and
revision.json to /var/lib/etcd-backup, deletes etcd_perf*, then exits if
anything remains in /var/lib/etcd. Extra files (perf artifacts, stray
snapshots, etc.) cause DR restore to fail before the restore pod starts.

Add backup_remaining_etcd_data_dir_contents() to move all remaining
top-level entries to /var/lib/etcd-backup instead of failing.

Fixes: ETCD-704
Related: https://access.redhat.com/solutions/6958920
---
 bindata/etcd/cluster-restore.sh | 27 +++++++++++++++++++--------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/bindata/etcd/cluster-restore.sh b/bindata/etcd/cluster-restore.sh
index 8e36dd7bad..a588224ef8 100644
--- a/bindata/etcd/cluster-restore.sh
+++ b/bindata/etcd/cluster-restore.sh
@@ -61,6 +61,22 @@ function restore_static_pods() {
   done
 }
 
+function backup_remaining_etcd_data_dir_contents() {
+  local entry base
+
+  shopt -s nullglob dotglob
+  for entry in "${ETCD_DATA_DIR}"/*; do
+    base=$(basename "${entry}")
+    if [ -e "${ETCD_DATA_DIR_BACKUP}/${base}" ]; then
+      echo "removing previous backup ${ETCD_DATA_DIR_BACKUP}/${base}"
+      rm -rf "${ETCD_DATA_DIR_BACKUP:?}/${base}"
+    fi
+    echo "Moving ${entry} to ${ETCD_DATA_DIR_BACKUP}/"
+    mv "${entry}" "${ETCD_DATA_DIR_BACKUP}/"
+  done
+  shopt -u nullglob dotglob
+}
+
 BACKUP_DIR="$1"
 # shellcheck disable=SC2012
 BACKUP_FILE=$(ls -vd "${BACKUP_DIR}"/static_kuberesources*.tar.gz | tail -1) || true
@@ -110,14 +126,9 @@ if [ -z "${ETCD_ETCDCTL_RESTORE}" ]; then
   cp -p "${SNAPSHOT_FILE}" "${ETCD_DATA_DIR_BACKUP}"/snapshot.db
   # Move the revision.json when it exists
   [ ! -f "${ETCD_REV_JSON}" ] ||  mv -f "${ETCD_REV_JSON}" "${ETCD_DATA_DIR_BACKUP}"/revision.json
-  # removing any fio perf files left behind that could be deleted without problems
-  rm -f "${ETCD_DATA_DIR}"/etcd_perf*
-
-  # ensure the folder is really empty, otherwise the restore pod will crash loop
-  if [ -n "$(ls -A "${ETCD_DATA_DIR}")" ]; then
-      echo "folder ${ETCD_DATA_DIR} is not empty, please review and remove all files in it"
-      exit 1
-  fi
+  # Move any remaining files (fio perf artifacts, stray snapshots, etc.) out of the data dir.
+  # The restore pod requires /var/lib/etcd to be empty before it runs.
+  backup_remaining_etcd_data_dir_contents
 
   echo "starting restore-etcd static pod"
   cp -p "${RESTORE_ETCD_POD_YAML}" "${MANIFEST_DIR}/etcd-pod.yaml"

From 29d2cdd74b7b3d9a73cefb869354daecf6f967c8 Mon Sep 17 00:00:00 2001
From: Apurva Nisal <anisal@redhat.ocm>
Date: Tue, 9 Jun 2026 23:02:36 +0530
Subject: [PATCH 2/4] ETCD-704: move extra etcd data dir files to backup
 subdirectory

Move leftover /var/lib/etcd entries to extra-data-dir-contents/
so staged snapshot.db, revision.json, and member/ are not overwritten.

Co-authored-by: Cursor <cursoragent@cursor.com>
---
 bindata/etcd/cluster-restore.sh | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/bindata/etcd/cluster-restore.sh b/bindata/etcd/cluster-restore.sh
index a588224ef8..468ab6f7ad 100644
--- a/bindata/etcd/cluster-restore.sh
+++ b/bindata/etcd/cluster-restore.sh
@@ -62,17 +62,19 @@ function restore_static_pods() {
 }
 
 function backup_remaining_etcd_data_dir_contents() {
-  local entry base
+  local entry base extras_dir="${ETCD_DATA_DIR_BACKUP}/extra-data-dir-contents"
+
+  mkdir -p "${extras_dir}"
 
   shopt -s nullglob dotglob
   for entry in "${ETCD_DATA_DIR}"/*; do
     base=$(basename "${entry}")
-    if [ -e "${ETCD_DATA_DIR_BACKUP}/${base}" ]; then
-      echo "removing previous backup ${ETCD_DATA_DIR_BACKUP}/${base}"
-      rm -rf "${ETCD_DATA_DIR_BACKUP:?}/${base}"
+    if [ -e "${extras_dir}/${base}" ]; then
+      echo "removing previous backup ${extras_dir}/${base}"
+      rm -rf "${extras_dir:?}/${base}"
     fi
-    echo "Moving ${entry} to ${ETCD_DATA_DIR_BACKUP}/"
-    mv "${entry}" "${ETCD_DATA_DIR_BACKUP}/"
+    echo "Moving ${entry} to ${extras_dir}/"
+    mv "${entry}" "${extras_dir}/"
   done
   shopt -u nullglob dotglob
 }

From bd47518facfc4fd64a9b439af0011157ade1b46c Mon Sep 17 00:00:00 2001
From: Apurva Nisal <anisal@redhat.ocm>
Date: Thu, 11 Jun 2026 20:02:52 +0530
Subject: [PATCH 3/4] ETCD-704: move backup helper to etcd-common-tools Share
 backup_remaining_etcd_data_dir_contents via etcd-common-tools per review
 feedback. Behavior unchanged.

---
 bindata/etcd/cluster-restore.sh | 18 ------------------
 bindata/etcd/etcd-common-tools  | 18 ++++++++++++++++++
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/bindata/etcd/cluster-restore.sh b/bindata/etcd/cluster-restore.sh
index 468ab6f7ad..962a5610b3 100644
--- a/bindata/etcd/cluster-restore.sh
+++ b/bindata/etcd/cluster-restore.sh
@@ -61,24 +61,6 @@ function restore_static_pods() {
   done
 }
 
-function backup_remaining_etcd_data_dir_contents() {
-  local entry base extras_dir="${ETCD_DATA_DIR_BACKUP}/extra-data-dir-contents"
-
-  mkdir -p "${extras_dir}"
-
-  shopt -s nullglob dotglob
-  for entry in "${ETCD_DATA_DIR}"/*; do
-    base=$(basename "${entry}")
-    if [ -e "${extras_dir}/${base}" ]; then
-      echo "removing previous backup ${extras_dir}/${base}"
-      rm -rf "${extras_dir:?}/${base}"
-    fi
-    echo "Moving ${entry} to ${extras_dir}/"
-    mv "${entry}" "${extras_dir}/"
-  done
-  shopt -u nullglob dotglob
-}
-
 BACKUP_DIR="$1"
 # shellcheck disable=SC2012
 BACKUP_FILE=$(ls -vd "${BACKUP_DIR}"/static_kuberesources*.tar.gz | tail -1) || true
diff --git a/bindata/etcd/etcd-common-tools b/bindata/etcd/etcd-common-tools
index 99bf470f1e..f3c36da647 100644
--- a/bindata/etcd/etcd-common-tools
+++ b/bindata/etcd/etcd-common-tools
@@ -78,6 +78,24 @@ function wait_for_containers_to_stop() {
   done
 }
 
+function backup_remaining_etcd_data_dir_contents() {
+  local entry base extras_dir="${ETCD_DATA_DIR_BACKUP}/extra-data-dir-contents"
+
+  mkdir -p "${extras_dir}"
+
+  shopt -s nullglob dotglob
+  for entry in "${ETCD_DATA_DIR}"/*; do
+    base=$(basename "${entry}")
+    if [ -e "${extras_dir}/${base}" ]; then
+      echo "removing previous backup ${extras_dir}/${base}"
+      rm -rf "${extras_dir:?}/${base}"
+    fi
+    echo "Moving ${entry} to ${extras_dir}/"
+    mv "${entry}" "${extras_dir}/"
+  done
+  shopt -u nullglob dotglob
+}
+
 function mv_static_pods() {
   local containers=("$@")
 

From 04434a0b6d9b326a91848fbd3234214ce539ba88 Mon Sep 17 00:00:00 2001
From: Apurva Nisal <anisal@redhat.ocm>
Date: Fri, 12 Jun 2026 18:06:41 +0530
Subject: [PATCH 4/4] ETCD-704: backup extra data dir contents in
 cluster-restore-tnf.sh Call backup_remaining_etcd_data_dir_contents before rm
 -rf so leftover files are moved to extra-data-dir-contents/ instead of
 deleted.

---
 bindata/etcd/cluster-restore-tnf.sh | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/bindata/etcd/cluster-restore-tnf.sh b/bindata/etcd/cluster-restore-tnf.sh
index 63074cd342..a0e5737d5d 100644
--- a/bindata/etcd/cluster-restore-tnf.sh
+++ b/bindata/etcd/cluster-restore-tnf.sh
@@ -196,6 +196,9 @@ if [ -d "${ETCD_DATA_DIR}/member" ]; then
   mv "${ETCD_DATA_DIR}"/member "${ETCD_DATA_DIR_BACKUP}"/
 fi
 
+# Move any remaining files out of the data dir before wiping it.
+backup_remaining_etcd_data_dir_contents
+
 echo "removing etcd data dir..."
 rm -rf "${ETCD_DATA_DIR}"
 mkdir -p "${ETCD_DATA_DIR}"