fix cron after first run + add hc

.env.example

@@ -7,3 +7,7 @@ MINIO_ACCOUNT_ID=account
 MINIO_APPLICATION_KEY=key
 MINIO_ENDPOINT=https://s3.domain.ru
 MINIO_LOCATION=ru-nsk
+
+# healthchecks.io (self-hosted instance is fine), leave empty to disable
+HEALTHCHECK_URL=https://hc.domain.ru/ping
+HEALTHCHECK_UUID=00000000-0000-0000-0000-000000000000

.gitignore

@@ -1,2 +1,3 @@
 .idea
 .env
+var/*

Dockerfile

@@ -1,17 +1,19 @@
-FROM alpine:latest
+FROM alpine:3.19
 
 ENV TZ=UTC
 LABEL maintainer="Vassiliy Yegorov <vasyakrg@gmail.com>"
 
 RUN apk add --no-cache \
     sqlite \
+    wget \
     curl \
     bash \
     tzdata \
-    openssl
+    openssl \
+    && rm -rf /var/cache/apk/*
 
-RUN curl -O https://downloads.rclone.org/rclone-current-linux-amd64.zip && \
+RUN wget https://github.com/rclone/rclone/releases/download/v1.71.0/rclone-v1.71.0-linux-amd64.zip && \
-    unzip rclone-current-linux-amd64.zip && \
+    unzip rclone-v1.71.0-linux-amd64.zip && \
     cd rclone-*-linux-amd64 && \
     cp rclone /usr/bin/ && \
     chown root:root /usr/bin/rclone && \

README.md

@@ -18,6 +18,8 @@ services:
       - MINIO_APPLICATION_KEY
       - MINIO_ENDPOINT
       - MINIO_LOCATION
+      - HEALTHCHECK_URL
+      - HEALTHCHECK_UUID
 ```
 
 пример .env
@@ -32,8 +34,34 @@ MINIO_ACCOUNT_ID=account
 MINIO_APPLICATION_KEY=key
 MINIO_ENDPOINT=https://s3.domain.ru
 MINIO_LOCATION=ru-nsk
+
+# необязательно: пинг в healthchecks.io (в т.ч. self-hosted)
+HEALTHCHECK_URL=https://hc.domain.ru/ping
+HEALTHCHECK_UUID=00000000-0000-0000-0000-000000000000
 ```
 
 1. myminio - системно! не менять
 2. проверьте настройки локации, не должно быть пусто
 3. DELETE_AFTER - в днях
+
+## Healthchecks
+
+Если `HEALTHCHECK_URL` пуст — пинги не отправляются, бекап работает как обычно.
+
+Итоговый адрес: `${HEALTHCHECK_URL}/${HEALTHCHECK_UUID}`
+
+| Момент | Запрос |
+|---|---|
+| старт бекапа | `.../${HEALTHCHECK_UUID}/start` |
+| дамп снят, проверен `PRAGMA integrity_check`, запакован, залит в s3, размер в s3 сверен с локальным, старое удалено | `.../${HEALTHCHECK_UUID}` |
+| любая ошибка на любом шаге | `.../${HEALTHCHECK_UUID}/fail` |
+
+Для self-hosted инстанса `HEALTHCHECK_URL` — это адрес его ping-эндпоинта, например `https://hc.domain.ru/ping`.
+
+## Как передаётся окружение в cron
+
+`crond` из busybox запускает джобы с пустым окружением, поэтому `entrypoint.sh`
+сохраняет env контейнера в `/tmp/container.env` (права 600), а cron-запись
+вызывает `/scripts/backup.sh --from-cron`, который его подгружает.
+Ручной запуск (`docker exec ... /scripts/backup.sh`) использует то окружение,
+с которым его вызвали.

docker-build.sh

@@ -1,6 +1,9 @@
 #!/bin/bash
 
-DOCKER_BASEIMAGE=alpine:latest
+TAG=$1
 
-docker buildx build --platform linux/amd64,linux/arm64 --push -t hub.realmanual.ru/pub/sqlite-backup \
+if [ -z "$TAG" ]; then
-        --build-arg DOCKER_BASEIMAGE=${DOCKER_BASEIMAGE} .
+  TAG=latest
+fi
+
+docker buildx build --platform linux/amd64,linux/arm64 --push -t git.realmanual.ru/pub/sqlite-backup:${TAG} .

docker-compose.yml

@@ -1,6 +1,9 @@
 services:
   sqlite_backup:
-    image: git.realmanual.ru/pub/sqlite-backup/backup:latest
+    # image: git.realmanual.ru/pub/sqlite-backup/backup:latest
+    build:
+      context: .
+      dockerfile: Dockerfile
     container_name: sqlite_backup
     restart: always
     volumes:
@@ -15,3 +18,5 @@ services:
       - MINIO_APPLICATION_KEY
       - MINIO_ENDPOINT
       - MINIO_LOCATION
+      - HEALTHCHECK_URL
+      - HEALTHCHECK_UUID

scripts/backup.sh

@@ -1,15 +1,79 @@
 #!/bin/sh
 
-BACKUP_FILE="sqlite_$(date "+%F-%H%M%S")"
+# crond gives the job a bare environment (HOME/PATH/SHELL only), so the cron
-
+# entry passes --from-cron and the env dumped by entrypoint.sh is restored here.
-sqlite3 ${DB_FILE} ".backup '/tmp/db.sqlite'"
+# A manual run keeps the environment it was invoked with.
-tar -zcvf /tmp/${BACKUP_FILE}.tar.gz /tmp/db.sqlite
+if [ "${1:-}" = "--from-cron" ] && [ -f /tmp/container.env ]; then
-
+  . /tmp/container.env
-/scripts/minio_uploader.sh copy /tmp/${BACKUP_FILE}.tar.gz ${MINIO_PATH}
-
-rm /tmp/${BACKUP_FILE}.tar.gz
-
-if [ ! -z $DELETE_AFTER ] && [ $DELETE_AFTER -gt 0 ]
-then
-  /scripts/deleteold.sh ${MINIO_PATH}
 fi
+
+set -u
+
+BACKUP_FILE="sqlite_$(date "+%F-%H%M%S").tar.gz"
+DUMP=/tmp/db.sqlite
+ARCHIVE="/tmp/${BACKUP_FILE}"
+DEST="${MINIO_PATH%/}/"
+
+log() { echo "[$(date '+%F %T')] $*"; }
+
+# healthchecks.io ping. Self-hosted instance is supported via HEALTHCHECK_URL.
+# Final url: ${HEALTHCHECK_URL}/${HEALTHCHECK_UUID}[/start|/fail]
+hc_ping() {
+  [ -n "${HEALTHCHECK_URL:-}" ] || return 0
+  url="${HEALTHCHECK_URL%/}"
+  [ -n "${HEALTHCHECK_UUID:-}" ] && url="${url}/${HEALTHCHECK_UUID}"
+  [ -n "${1:-}" ] && url="${url}/$1"
+  curl -fsS -m 10 --retry 3 -o /dev/null "$url" \
+    || log "WARN: healthcheck ping failed: ${url}"
+}
+
+cleanup() { rm -f "$DUMP" "$ARCHIVE"; }
+
+fail() {
+  log "ERROR: $*"
+  cleanup
+  hc_ping fail
+  exit 1
+}
+
+[ -n "${DB_FILE:-}" ] || fail "DB_FILE is not set"
+[ -f "$DB_FILE" ] || fail "database not found: $DB_FILE"
+[ -n "${MINIO_PATH:-}" ] || fail "MINIO_PATH is not set"
+
+hc_ping start
+log "backup started: $DB_FILE -> ${DEST}${BACKUP_FILE}"
+
+# 1. Consistent snapshot
+rm -f "$DUMP"
+sqlite3 "$DB_FILE" ".backup '$DUMP'" || fail "sqlite3 .backup failed"
+[ -s "$DUMP" ] || fail "dump is empty"
+
+# 2. Verify the snapshot before shipping it anywhere
+INTEGRITY=$(sqlite3 "$DUMP" "PRAGMA integrity_check;" 2>&1) \
+  || fail "integrity_check failed to run: $INTEGRITY"
+[ "$INTEGRITY" = "ok" ] || fail "integrity_check: $INTEGRITY"
+
+# 3. Pack
+tar -C /tmp -zcf "$ARCHIVE" "$(basename "$DUMP")" || fail "tar failed"
+LOCAL_SIZE=$(wc -c < "$ARCHIVE" | tr -d ' ')
+[ "$LOCAL_SIZE" -gt 0 ] || fail "archive is empty"
+
+# 4. Upload
+/scripts/minio_uploader.sh copy "$ARCHIVE" "$DEST" || fail "upload failed"
+
+# 5. Verify the remote object exists and matches the local size
+REMOTE_SIZE=$(/scripts/minio_uploader.sh size "${DEST}${BACKUP_FILE}")
+[ -n "$REMOTE_SIZE" ] || fail "uploaded file not found at ${DEST}${BACKUP_FILE}"
+[ "$REMOTE_SIZE" = "$LOCAL_SIZE" ] \
+  || fail "size mismatch: local=${LOCAL_SIZE} remote=${REMOTE_SIZE}"
+
+log "uploaded and verified: ${DEST}${BACKUP_FILE} (${LOCAL_SIZE} bytes)"
+cleanup
+
+# 6. Rotate old backups
+if [ -n "${DELETE_AFTER:-}" ] && [ "${DELETE_AFTER}" -gt 0 ] 2>/dev/null; then
+  /scripts/deleteold.sh || fail "rotation of old backups failed"
+fi
+
+log "backup finished successfully"
+hc_ping

scripts/deleteold.sh

@@ -1,24 +1,28 @@
 #!/bin/bash
 
+set -u
+
 # Seconds since epoch for current time
 DATE_NOW=$(date +%s)
+DEST="${MINIO_PATH%/}/"
 
-/scripts/minio_uploader.sh list $MINIO_PATH | grep "sqlite_" | while read LINE
+/scripts/minio_uploader.sh list "$DEST" | grep "sqlite_" | while read -r LINE
 do
-  BACKUP_FILENAME=$(echo $LINE | awk '{ print $3 }')
+  # `rclone ls` prints "<size> <path>"
+  BACKUP_FILENAME=$(echo "$LINE" | awk '{ print $2 }')
 
-  BACKUP_DATE=$(echo $BACKUP_FILENAME | awk 'BEGIN { FS = "[_-]" } ; { printf "%s-%s-%s",$2,$3,$4 }')
+  BACKUP_DATE=$(echo "$BACKUP_FILENAME" | awk 'BEGIN { FS = "[_-]" } ; { printf "%s-%s-%s",$2,$3,$4 }')
 
   if [[ $BACKUP_DATE =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}$ ]]
   then
-    BACKUP_DATE_SECS=$(date -d $BACKUP_DATE +%s)
+    BACKUP_DATE_SECS=$(date -d "$BACKUP_DATE" +%s)
 
-    DAYS_DIFF=$(( ($DATE_NOW - $BACKUP_DATE_SECS) / (60*60*24) ))
+    DAYS_DIFF=$(( (DATE_NOW - BACKUP_DATE_SECS) / (60*60*24) ))
 
     if [ "$DAYS_DIFF" -gt "$DELETE_AFTER" ]
     then
       echo "File $BACKUP_FILENAME is $DAYS_DIFF days old (greater than $DELETE_AFTER days). Deleting it."
-      /scripts/minio_uploader.sh delete /$BACKUP_FILENAME
+      /scripts/minio_uploader.sh delete "${DEST}${BACKUP_FILENAME}"
     fi
   fi
 done

scripts/entrypoint.sh

@@ -1,8 +1,22 @@
 #!/bin/sh
 
-# Create crontab in a writable location
+set -e
-mkdir -p /tmp/crontabs
-echo "${CRONTAB} /scripts/backup.sh" > /tmp/crontabs/root
 
-/scripts/backup.sh
+# crond runs jobs with a bare environment (HOME/PATH/SHELL only), so the
-exec crond -f -c /tmp/crontabs
+# container env is dumped here and sourced back by backup.sh on every run.
+ENV_FILE=/tmp/container.env
+export -p | grep -v -E "^export (PWD|SHLVL|OLDPWD|_)=" > "$ENV_FILE"
+chmod 600 "$ENV_FILE"
+
+# Create crontab in a writable location and set proper permissions
+mkdir -p /tmp/crontabs
+echo "${CRONTAB:-"0 * * * *"} /scripts/backup.sh --from-cron >> /proc/1/fd/1 2>&1" > /tmp/crontabs/root
+chmod 644 /tmp/crontabs/root
+
+echo "[entrypoint] schedule: ${CRONTAB:-"0 * * * *"} (TZ=${TZ:-UTC})"
+
+# Run initial backup (do not abort the container if it fails)
+/scripts/backup.sh || echo "[entrypoint] initial backup failed, continuing to crond"
+
+# Start crond in foreground with debug output
+exec crond -f -c /tmp/crontabs -d 0

scripts/minio_uploader.sh

@@ -1,5 +1,7 @@
 #!/bin/bash
 
+set -u
+
 # Create rclone config on-the-fly in a writable location
 RCLONE_CONFIG_DIR="/tmp/rclone"
 mkdir -p "$RCLONE_CONFIG_DIR"
@@ -14,19 +16,30 @@ secret_access_key = $MINIO_APPLICATION_KEY
 endpoint = $MINIO_ENDPOINT
 region = $MINIO_LOCATION
 acl = private
+no_check_bucket = true
 EOF
+chmod 600 "$RCLONE_CONFIG"
 
-case $1 in
+# All commands take full remote paths; callers own the path building.
+case "${1:-}" in
 	copy)
 		echo "copy from $2 to $3"
-		rclone copy --progress $2 myminio:$3
+		rclone copy --stats-one-line --retries 3 "$2" "$3"
 	;;
 	list)
-		echo "list $2"
+		echo "list $2" >&2
-		rclone ls myminio:$2
+		rclone ls "$2"
+	;;
+	size)
+		# bytes of a single remote object, empty output if it does not exist
+		rclone lsf --format s "$2" 2>/dev/null
 	;;
 	delete)
 		echo "delete $2"
-		rclone delete myminio:$2
+		rclone deletefile "$2"
+	;;
+	*)
+		echo "usage: $0 {copy SRC DST|list PATH|size FILE|delete FILE}" >&2
+		exit 2
 	;;
 esac