[BUG] Could not bring Docker Engine back online

[marithpl]

Describe the bug
I try to install Synology-docker to use my new docker images which requires new docker version.
I stuck on error during syno_docker_update.sh logger
I suppose that docker is not stopped on first step of script. Because all of my services ale working.

I have EXT4 file system

To reproduce

1. $ sudo ./syno_docker_update.sh logger

Expected behavior
IDK what should be displayed but no error for sure ;)

Log file

Step 4 from 4: Starting Docker service - May take a while. 
   - timeout set to 24m based on 16 active containers
{"action":"prepare","error":{"code":150,"description":"Failed to load package info"},"last_stage":"prepare","stage":"prepare_start","success":false}
ERROR: Could not bring Docker Engine back online

Docker daemon configuration

$ sudo cat /var/packages/Docker/etc/dockerd.json
{
  "data-root": "/var/packages/Docker/var/docker",
  "group": "administrators",
  "log-driver": "local",
  "log-opts": {
    "max-file": "5",
    "max-size": "20m"
  },
  "registry-mirrors": [],
  "storage-driver": "aufs"
}

Additional context
./syno_docker_list_containers.sh show all containers are db

[telnetdoogie]

If all containers still show as db then you haven't yet followed step 2 and rebuilt containers to use the local logger.

Containers still showing db will not start if you do a full update using this script, and if there are any containers still set up for the db logger, an updated docker will not start.

So... you need to rebuild containers to use the local logger as per the instructions.

If you're on ext4 instead of btrfs you may also need overlay drivers for your NAS.

[marithpl]

Step 2 from 4: Backing up current Docker binaries (/volume1/homes/MaRith/synology-docker/docker_backup_20260310_155004.tgz)
bin/
bin/dockerd
bin/docker-compose
bin/docker-proxy
bin/containerd
bin/docker-init
bin/auplink
bin/containerd-stress
bin/ctr
bin/containerd-shim
bin/containerd-shim-runc-v1
bin/runc
bin/docker
bin/containerd-shim-runc-v2
dockerd.json
start-stop-status

im not sure if I understood. Should I follow step 2 and then come back to step 1?

[telnetdoogie]

Step 2 in the instructions:

---

---

Step 2: Recreate containers still using db

For each compose‑managed container using db:

cd /volume1/docker/jellyfin
docker-compose up -d --force-recreate

Re‑run syno_docker_list_containers.sh until everything says local.

Containers created via docker run will show a best‑guess recreate command. Verify it before running.

---

Step 3: Upgrade Docker & Compose

sudo ./syno_docker_update.sh update

If you did the logger step correctly, containers should come back automatically.

---

[marithpl]

Even though there was an error in the first step, should I complete step 2 and step 3?

[telnetdoogie]

You said all of your services are working, so I assume things are fine? Sometimes the startup times out.
If your log driver is changed (the file looks correct, but you can validate with docker info | grep logger ) then you can continue to move all the containers over to the logger driver and confirm with syno_docker_list_containers.sh

[github-actions]

This issue has been marked as stale and will be closed

[ehoberg]

I actually have the same issue. However after i verified that my containers were using the local logger, i decided to give it a shot and continue anyways.

No beuno unfortunately.

When attempting to upgrade i get the following error:

Step 2 from 12: Checking for / Installing iptables modules.
Based on this version of docker, we'll need to check for / install iptables modules...
Kernel Version: 4.4.180+
Platform: apollolake

• .ko files in place ? 🔴

• kernel modules loaded ? 🔴

• available for download ? 🔴

• CM script loads modules ?grep: /var/packages/ContainerManager/scripts/start-stop-status: No such file or directory
  🔴

  The kernel modules for your platform and kernel are not available for download.
  You can compile them yourself and place them in /lib/modules or request a compile for your platform
  We cannot continue without these modules.
  ERROR: Could not install iptables modules. Stopping.

Maybe my nas is too old to support it?

using a DS718+

[telnetdoogie]

@ehoberg your NAS isn't too old, but your kernel version is an odd one I've not seen before. You should update to the latest DSM, which for your NAS should be 7.3.x and that will bring your kernel up to 4.4.302+ - the modules for 4.4.302+ are available.

[ehoberg]

Hmm that weird, I always try to keep it up-to-date, so I just checked and I'm currently running the latest version which is 7.1.1.

Looks like it is necessary to manually update to 7.2 and probably 7.3. I'll look into it later. Thanks for the info, was unaware of even newer dsm versions

[ehoberg]

I've upgraded to the latest dsm version (7.3.2). I'm getting a lot further now, however it still crashes:

Step 2 from 12: Checking for / Installing iptables modules.
Based on this version of docker, we'll need to check for / install iptables modules...
Kernel Version: 4.4.302+
Platform: apollolake

• .ko files in place ? 🟢
• kernel modules loaded ? 🟢
• available for download ? 🟢
• CM script loads modules ? 🔴

Modifying the ContainerManager startup to install modules...

• .ko files in place ? 🟢
• kernel modules loaded ? 🟢
• available for download ? 🟢
• CM script loads modules ? 🟢

Step 10 from 12: Starting Docker service - May take a while.

• timeout set to 10m based on 5 active containers
  {"action":"start","beta":false,"error":{"code":272,"description":"Failed to run script, script=[start]"},"finished":true,"language":"enu","last_stage":"prepare_start","package":"ContainerManager","pid":30746,"scripts":[{"code":1,"message":"","type":"start"}],"stage":"start_failed","status":"stop","status_code":311,"status_description":"translate from systemd status","success":false,"username":"","version":"24.0.2-1606"}
  ERROR: Could not bring Docker Engine back online

It crashes so bad that the restore option doesn't work. Fortunately its pretty easy to reinstall / repair via synology package manager so its pretty easy to recover.

The restore gives the following errors:

sudo ./syno_docker_update.sh restore
Update Docker Engine and Docker Compose on Synology to target version

./syno_docker_update.sh: line 195: containerd: command not found
./syno_docker_update.sh: line 196: runc: command not found
Current DSM version: 7.3.2
Current Docker version: Unknown
Current Docker Compose version: Unknown
Current containerd version: Unknown
Current runc version: Unknown
ERROR: Could not detect current Docker version, use --force to override

[telnetdoogie]

are you using btrfs?

It's hard to know what's happening without some logging from the failed startup. can you see what errors were thrown in sudo journalctl or sudo dmesg when it's failing to start?