I have a server that runs multiple docker containers. I can access my server via SSH, and have set up my ssh_config to allow me to ssh into certain containers that I regularly access: Host some_container HostName my.server.com RemoteCommand docker compose -f /docker-compose.yml exec some_container fish RequestTTY force However, I now need to use a particular piece of software that uses ssh, and access my containers with it. This software sets the ssh command argument. Using the above configuration, this causes ssh to error out with Cannot execute command-line and remote command, due to the presence of RemoteCommand. I do NOT want to have to run an sshd server inside the container. I have attempted to replace RemoteCommand with ProxyCommand, but this reusults in me connecting to my server and the docker command being ignored: ProxyCommand ssh %h -W %h:%p \ -o "RequestTTY=force" \ -o "SessionType=default" \ -o "RemoteCommand=docker compose -f /docker-compose.yml exec some_container fish" (note that this is all one line in my ssh_config - I have split it up here to make it easier to read). Is there any way to ssh into my docker container without running sshd in the container or using RemoteCommand?

I installed Podman Desktop app v. 1.18.1 on macOS Sequoia. If I execute:

podman compose version

… I get this on the console: > \>>>> Executing external compose provider "/usr/local/bin/docker-compose". Please see podman-compose(1) for how to disable this message. >Docker Compose version v2.36.0 I am guessing that Podman is using an implementation of *Docker Compose* rather than its own implementation. (Does Podman even have its own implementation of the *Compose* spec ?) That message seems to suggest I should call a command podman-compose. But there is no such command. Running this:

which podman-compose

… results in : >podman-compose not found Obviously there is a podman compose command+subcommand. I used that above for version. But the message says podman-compose with a hyphen, a command that does not seem to exist. So how do I disable that message?

I'm facing with the following error using docker on my Linux Ubuntu system. ubuntu@ubuntu:~$ docker version Client: Docker Engine - Community Version: 28.0.1 API version: 1.48 Go version: go1.23.6 Git commit: 068a01e Built: Wed Feb 26 10:41:08 2025 OS/Arch: linux/amd64 Context: default ubuntu@ubuntu:~$ docker images REPOSITORY TAG IMAGE ID CREATED SIZE yangsuite-one-container latest 604c66275985 7 weeks ago 2.7GB ubuntu 22.04 a24be041d957 3 months ago 77.9MB carlo-ubuntu latest 1245a5e87e81 5 months ago 608MB ubuntu_iperf latest 072ecc269646 7 months ago 387MB debian latest 617f2e89852e 7 months ago 117MB ubuntu 20.04 6013ae1a63c2 7 months ago 72.8MB networkstatic/iperf3 latest 377c80503c6d 12 months ago 82MB ios-xr/xrd-control-plane 7.11.2 58b64211c345 14 months ago 1.29GB ubuntu 18.04 f9a80a55f492 24 months ago 63.2MB ubuntu@ubuntu:~$ ubuntu@ubuntu:~$ docker inspect ios-xr/xrd-control-plane [] Error: No such object: ios-xr/xrd-control-plane ubuntu@ubuntu:~$ ubuntu@ubuntu:~$ docker inspect "ios-xr/xrd-control-plane" [] Error: No such object: ios-xr/xrd-control-plane ubuntu@ubuntu:~$ As you can see, when I try to access/inspect the tagged image ios-xr/xrd-control-plane I get an error. Which is actually the problem ?

I have a server running this [CMS](https://github.com/mediacms-io/mediacms) as a website by running a Docker file like [this](https://github.com/mediacms-io/mediacms/blob/main/docker-compose-letsencrypt.yaml) by docker-compose which internally uses nginxproxy/nginx-proxy and nginxproxy/acme-companion Docker images. Now, I intend to follow the instructions given [here](https://www.euro-space.net/blog/virtual-server-for-vpn/tutorial/how-to-setup-ikev2-vpn-using-strongswan-and-letsencrypt-on-centos-8.php) to set up a VPN on the same server which uses Let's Encrypt. # Question Can I run the VPN alongside the previous CMS? Would I run into any trouble?

I want to debug a failing docker compose build --no-cache (it succeeds with the cache), however due to the log limit, I cannot see the reason for the failure, only the message output clipped, log limit 1MiB reached. Following the Docker documentation, I created the following /etc/docker/daemon.json:

{
  "features": { "buildkit": true },
  "log-driver": "local",
  "log-opts": 
  {
    "max-size": "100m",
    "max-file": "3"
  }
}

After systemctl docker restart, docker info shows:

Client:
 Context:    default
 Debug Mode: false
 Plugins:
  buildx: Docker Buildx (Docker Inc., v0.8.2-docker)
  compose: Docker Compose (Docker Inc., 2.6.0)

Server:
 Containers: 0
  Running: 0
  Paused: 0
  Stopped: 0
 Images: 0
 Server Version: 20.10.17
 Storage Driver: overlay2
  Backing Filesystem: extfs
  Supports d_type: true
  Native Overlay Diff: false
  userxattr: false
 Logging Driver: local
 Cgroup Driver: systemd
 Cgroup Version: 2
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
 Swarm: inactive
 Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: 10c12954828e7c7c9b6e0ea9b0c02b01407d3ae1.m
 runc version: 
 init version: de40ad0
 Security Options:
  seccomp
   Profile: default
  cgroupns
 Kernel Version: 5.18.5-arch1-1
 Operating System: Arch Linux
 OSType: linux
 Architecture: x86_64
 CPUs: 8
 Total Memory: 15.41GiB
 Name: konradspc
 ID: 6NFV:RQP7:V6XK:7D6Z:W2LC:LPGR:HQBQ:V55P:BECL:WXPP:YPC5:2QQ2
 Docker Root Dir: /var/lib/docker
 Debug Mode: false
 Registry: https://index.docker.io/v1/ 
 Labels:
 Experimental: false
 Insecure Registries:
  127.0.0.0/8
 Live Restore Enabled: false

Notably, the logging driver is now shown as "local", where before it was "json-file", so it seems to me as if Docker is successfully loading /etc/docker/daemon.json. However when I execute docker compose build --no-cache again, I still get the message output clipped, log limit 1MiB reached. How do I get docker compose to not ignore daemon.json or is there any other way to access the full build log? I am using Docker version 20.10.17, build 100c70180f, Docker Compose version 2.6.0 with BuildKit on Arch Linux with Kernel 5.18.5. BuildKit cannot be deactivated because the build uses --mount=type=cache,target=/root/.m2 to cache Maven dependencies.

I run my docker stuff with a dedicated user, and installed the docker rootlesskit. I start docker with systemctl --user start docker.service. Everything related to docker, executed with that user, works. I am now installing a nostr relay. I followed the instructions and the thing actually works. But. I have to run it via tmux and then scripts/start_local. This script basically runs docker compose -f $DOCKER_FILE up -d. I tried setting up a systemd script for that. First, I tried the /etc/systemd/system/ location, with the correct USER variable and working directory. However, it would fail to start with the famous Cannot connect to the Docker daemon at unix:///home/me/.docker/run/docker.sock. Is the docker daemon running? error message. So I thought maybe it's because I should run the systemd script as the user. So I installed the script into $HOME/.config/systemd/user with a symlink to $HOME/.config/systemd/user/default.target.wants. Basically, the same way my docker service runs. To my surprise, I get the same error. Going back to running it in tmux, it just works. What is different here? For clarity, here is the service file I am using:

[Unit]
Description=Nostr TS Relay

[Service]
Type=simple
Restart=always
RestartSec=5
WorkingDirectory=/home/me/nostream
ExecStart=/home/me/nostream/scripts/start_local
ExecStop=/home/me/nostream/scripts/stop

[Install]
WantedBy=default.target

Note: it's not a big deal I can have it running with a tmux, but I'd prefer the systemd variant.

I want to configure cgroups V2 resource limitation on a Docker-Compose container. How do I do this?

I hava a senrio as follow: on basic.target stage. My host start running a serviceB, which create a unix socket file "/tmp/.test_sock" when service started. on the multi-user.target stage. The docker container "test_dc" starting created. what can i do, to create a volume in the docker test_dc, which is the mount of "/tmp/.test_sock"?

i have a pihole server running in docker compose on my Debian Linux server. i also host a wireguard vpn (also in docker compose) running on the same server. by using the tcpdump command i have confirmed that all traffic that happens on my laptop is routed to the Debian server. my only issue at this stage is pihole ad blocking doesn't seem to work while using it which is really important to me in some circumstances. is there anyway to have my wireguard vpn traffic use my pihole dns server?

I am trying to configure Owncloud in docker. The docker-compose yaml I am using look like this:

services:
  owncloud:
    image: owncloud/server:10.15
    container_name: owncloud_server
    restart: always
    ports:
      - "8084:8080"
    depends_on:
      - mariadb
      - redis
    environment:
      - OWNCLOUD_DOMAIN=https://owncloud.example.com 
      - OWNCLOUD_TRUSTED_DOMAINS=owncloud.example.com
      - OWNCLOUD_DB_TYPE=mysql
      - OWNCLOUD_DB_NAME=owncloud
      - OWNCLOUD_DB_USERNAME=owncloud
      - OWNCLOUD_DB_PASSWORD=owncloud
      - OWNCLOUD_DB_HOST=mariadb
      - OWNCLOUD_ADMIN_USERNAME=admin
      - OWNCLOUD_ADMIN_PASSWORD=admin
      - OWNCLOUD_MYSQL_UTF8MB4=true
      - OWNCLOUD_REDIS_ENABLED=true
      - OWNCLOUD_REDIS_HOST=redis
    healthcheck:
      test: ["CMD", "/usr/bin/healthcheck"]
      interval: 30s
      timeout: 10s
      retries: 5
    volumes:
      - /home/ubuntu/oci-owncloud-config:/var/www/config
      - /home/ubuntu/oci-owncloud-data:/mnt/data

  mariadb:
    image: mariadb:10.11
    container_name: owncloud_mariadb
    restart: always
    environment:
      - MYSQL_ROOT_PASSWORD=owncloud
      - MYSQL_USER=owncloud
      - MYSQL_PASSWORD=owncloud
      - MYSQL_DATABASE=owncloud
      - MARIADB_AUTO_UPGRADE=1
    command: ["--max-allowed-packet=128M", "--innodb-log-file-size=64M"]
    healthcheck:
      test: ["CMD", "mysqladmin", "ping", "-u", "root", "--password=owncloud"]
      interval: 10s
      timeout: 5s
      retries: 5
    volumes:
      - /home/ubuntu/oci-owncloud-db:/var/lib/mysql

  redis:
    image: redis:6
    container_name: owncloud_redis
    restart: always
    command: ["--databases", "1"]
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    volumes:
      - /home/ubuntu/oci-owncloud-redis:/data

When I run it, I get three containers (redis, mariadb, and owncloud). The folders in the host vm are connected to my OCI Object Storage with s3fs. The issue is on the folder user and permissions. What users and permissions should I use with Owncloud in order to connect them to my host vm's folders? I tried:

sudo chown -R root:root /home/ubuntu/oci-owncloud-(name)
sudo chmod -R 755 /home/ubuntu/oci-owncloud-(name)

but it doesn't work.

I am using Ubuntu on my host machine, and I have a docker container also running Ubuntu that contains an ASP .NET website. Now the issue is I can't seem to figure out how to get the container to mount my SSL keys from my host machine. My docker-compose.yml file has the following volumes specified.

volumes:
      - /etc/letsencrypt/archive/example.com/fullchain.pem:/etc/ssl/certs/fullchain.pem:ro
      - /etc/letsencrypt/archive/example.com/privkey.pem:/etc/ssl/private/privkey.pem:ro
      - app-data:/app/data
      - app-data:/root/.aspnet/DataProtection-Keys

volumes:
  app-data:

I also verified these files exist by using cat /etc/letsencrypt/archive/example.com/privkey1.pem and cat /etc/letsencrypt/archive/example.com/fullchain1.pem which all worked perfectly. But when I compose my container, I always get the following errors because it can't seem to find the file.

Unhandled exception. Interop+Crypto+OpenSslCryptographicException: error:2006D080:BIO routines:BIO_new_file:no such file
   at Interop.Crypto.CheckValidOpenSslHandle(SafeHandle handle)
   at Internal.Cryptography.Pal.OpenSslX509CertificateReader.FromFile(String fileName, SafePasswordHandle password, X509KeyStorageFlags keyStorageFlags)
   at System.Security.Cryptography.X509Certificates.X509Certificate..ctor(String fileName, String password, X509KeyStorageFlags keyStorageFlags)
   at System.Security.Cryptography.X509Certificates.X509Certificate2..ctor(String fileName, String password)
   at Microsoft.AspNetCore.Hosting.ListenOptionsHttpsExtensions.UseHttps(ListenOptions listenOptions, String fileName, String password)

I also ensured the permissions are right by running the following to no avail.

sudo chmod 644 /etc/letsencrypt/archive/example.com/fullchain1.pem
sudo chmod 600 /etc/letsencrypt/archive/example.com/privkey1.pem
sudo chmod 755 /etc/letsencrypt/archive
sudo chmod 755 /etc/letsencrypt/archive/example.com

Next, I tried manually starting the container, but I get the same error where the container instantly closes due to the exception.

docker run -it --rm \
  -v /etc/letsencrypt/archive/example.com/fullchain1.pem:/etc/ssl/certs/fullchain.pem:ro \
  -v /etc/letsencrypt/archive/example.com/privkey1.pem:/etc/ssl/private/privkey.pem:ro \
  server /bin/bash

Lastly, here is my Program class which is trying to read the HTTPS cert which is what is generating the actual no file found exception.

public class Program
    {
        public static void Main(string[] args)
        {
            CreateHostBuilder(args).Build().Run();
        }

        public static IHostBuilder CreateHostBuilder(string[] args) =>
        Host.CreateDefaultBuilder(args)
            .ConfigureWebHostDefaults(webBuilder =>
            {
                webBuilder.UseStartup()
                    .UseKestrel(options =>
                    {
                        options.ListenAnyIP(80); // HTTP
                        options.ListenAnyIP(443, listenOptions =>
                        {
                            listenOptions.UseHttps("/etc/ssl/certs/fullchain.pem", "/etc/ssl/private/privkey.pem");
                        });
                    });
            });
    }

So I have a docker network named home that all of my root-based (or docker containers that were simply too hard to port to podman) containers live. sudo docker network ls NETWORK ID NAME DRIVER SCOPE 9d5788b45048 bridge bridge local b1f4756feab4 home bridge local 5d7ee6579f19 host host local 8678a773e2f2 none null local And, for podman, I have a very similar configuration. podman network ls NETWORK ID NAME DRIVER 8b17ae3d5d67 home bridge 2f259bab93aa podman bridge The problem? Well, it turns out that my name resolution for containers doesn't work within one network to another. So, for example, I have nginx-proxy-manager running on podman and I want to redirect http://domain/freshrss to the freshrss service by specifying freshrss and the associated port number. This doesn't work, and that makes sense to me as the docker network and the podman network are fundamentally *detached* from one another. So my question is simple: Is there some way to **treat these two networks as *one* network by bridging them together without hurting the sanctity of individualized network configurations**? Alternatively, is there *some way* for me to get around this communication issue without having to respecify the domain name in the forward name and port? For example, I thought forwarding to 127.0.0.1: would work as it would go to the host, and then connect to the appropriate port but that didn't work. docker/podman compose tailored answers are welcome as that's how I'm configuring my services.

I want to use a vsftp-Server in my Docker Swarm but having some network issues. I have the following compose-file: services: vsftpd: container_name: vsftpd image: million12/vsftpd restart: always volumes: - /whatever:/var/ftp/:ro environment: - ANONYMOUS_ACCESS=true - LOG_STDOUT=true - CUSTOM_PASSIVE_ADDRESS="" ports: - 20-21:20-21 - 21100-21110:21100-21110 #network_mode: "host" If i run this compose file with docker compose and network mode "host" then i can easily connect to it. If i run it without network mode "host" i have the following output: Status: Angemeldet Status: Empfange Verzeichnisinhalt... Befehl: PWD Antwort: 257 "/" is the current directory Befehl: TYPE I Antwort: 200 Switching to Binary mode. Befehl: PASV Antwort: 500 OOPS: invalid pasv_address Befehl: PORT 10,10,10,102,10,187 Fehler: Verbindung vom Server geschlossen I can see the login in the output of the vsftp-Server: vsftpd | [VSFTPD 16:46:41] VSFTPD daemon starting vsftpd | Tue Nov 19 16:46:45 2024 [pid 33] CONNECT: Client "" vsftpd | Tue Nov 19 16:46:45 2024 [pid 32] [ftp] OK LOGIN: Client "", anon password "anonymous@example.com" I have the same results if i run it on swarm. It seems like im having issues with the "passive ports" and the docker network. Unfortunately there is no "network_mode: host" in swarm (even if i pin the service to a host) Is there a possibility to run vsftpd in passive Mode in Swarm like Compose with Network-Mode ? Or are there other ways to bring up a *working* ftp-Server in Swarm?

After upgrading Fedora 40 to 41, docker-compose was no longer available. When I try to re-install with sudo dnf install docker-compose, it raises the following conflicts:

- installed package docker-compose-plugin-2.29.7-1.fc41.x86_64 conflicts with docker-compose-plugin provided by docker-compose-2.29.7-1.fc41.x86_64 from fedora
- package docker-compose-2.29.7-1.fc41.x86_64 from fedora conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.2-1.fc41.x86_64 from docker-ce-stable
- package docker-compose-2.29.7-1.fc41.x86_64 from fedora conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.6-1.fc41.x86_64 from docker-ce-stable
- package docker-compose-2.29.7-1.fc41.x86_64 from fedora conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.7-1.fc41.x86_64 from docker-ce-stable

- installed package docker-compose-plugin-2.29.7-1.fc41.x86_64 conflicts with docker-compose-plugin provided by docker-compose-2.29.7-3.fc41.x86_64 from updates
- package docker-compose-2.29.7-3.fc41.x86_64 from updates conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.2-1.fc41.x86_64 from docker-ce-stable
- package docker-compose-2.29.7-3.fc41.x86_64 from updates conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.6-1.fc41.x86_64 from docker-ce-stable
- package docker-compose-2.29.7-3.fc41.x86_64 from updates conflicts with docker-compose-plugin provided by docker-compose-plugin-2.29.7-1.fc41.x86_64 from docker-ce-stable

Running the command line with --allowerasing or --skip-broken does not work either. I also tried (but to no avail): - removing/re-installing docker-compose-plugin - removing/re-installing Docker as per the documentation (https://docs.docker.com/engine/install/fedora) What more can I try to get docker-compose work again?

I have a docker compose project where one of the services launches several replicas using the replicas directive. The replicas have automatically enumerated names, which also serve as hostnames. Note, this different to container id. I want to find out what the replica number of a container is from inside itself, without installing special tools (like dig, etc), or running runtime commands with exec. I'm okay with passing in an environment variable, but I don't know how to access that number in order to make it an environment variable. The solutions I've seen either involve bloating the container with tools, confuse the concept of the id with the name, or suggest stuff like cat'ing /proc/self/cgroup, which just doesn't work for me.

How to take a full backup and restore of Docker Desktop. I see that there is a way to save the Docker image as a tar file in the local filesystem. I was looking to dump all the current images into one single tar file as a backup for the current status of the docker desktop To save a docker image

% docker save image-name | gzip > image-name.tar.gz

I am not aware how to restore it, again this is only one image I have almost ~30 odd images, can I take a backup in a single file

Sorry I don't know if this is a docker issue or a dig issue on rockylinux 9. Everything works as expected on rockylinux 8. I have a docker-compose.yml file below with a service named https. That allows the container to be referenced by the hostname https. While ping https works, for some reason dig https (DiG 9.16.23-RH) does not work on rockylinux 9. It does work on rockylinux 8 (DiG 9.11.36-RedHat-9.11.36-16.el8_10.2). If I change the service name to httpsx then dig httpsx works.

services:
  https:
    image: "rockylinux:${RL_VERSION}"
    command: bash -c "yum install -y iputils bind-utils && echo '=====dig version output====' && dig -v && echo '=====ping https output====' && ping -c 3 https && echo '=====dig https output====' && dig +short https"
    environment:
       - RL_VERSION

Working 8:

% RL_VERSION=8 docker-compose up
Attaching to https-1
https-1  | Rocky Linux 8 - AppStream                       5.7 MB/s |  11 MB     00:01    
...
https-1  | Complete!
https-1  | =====dig version output====
https-1  | DiG 9.11.36-RedHat-9.11.36-16.el8_10.2
https-1  | =====ping https output====
https-1  | PING https (172.21.0.2) 56(84) bytes of data.
https-1  | 64 bytes from c3f0c7a6613c (172.21.0.2): icmp_seq=1 ttl=64 time=0.558 ms
https-1  | 64 bytes from c3f0c7a6613c (172.21.0.2): icmp_seq=2 ttl=64 time=0.051 ms
https-1  | 64 bytes from c3f0c7a6613c (172.21.0.2): icmp_seq=3 ttl=64 time=0.040 ms
https-1  | 
https-1  | --- https ping statistics ---
https-1  | 3 packets transmitted, 3 received, 0% packet loss, time 2025ms
https-1  | rtt min/avg/max/mdev = 0.040/0.216/0.558/0.241 ms
https-1  | =====dig https output====
https-1  | 172.21.0.2

Failing 9:

% RL_VERSION=9 docker-compose up
[+] Running 1/1
 ✔ Container testhttps-https-1  Recreated                                                                                                    0.2s 
Attaching to https-1
https-1  | Rocky Linux 9 - BaseOS                          2.4 MB/s | 2.4 MB     00:00    
...
https-1  | Complete!
https-1  | =====dig version output====
https-1  | DiG 9.16.23-RH
https-1  | =====ping https output====
https-1  | PING https (172.21.0.2) 56(84) bytes of data.
https-1  | 64 bytes from 4a2841b5dac9 (172.21.0.2): icmp_seq=1 ttl=64 time=0.404 ms
https-1  | 64 bytes from 4a2841b5dac9 (172.21.0.2): icmp_seq=2 ttl=64 time=0.117 ms
https-1  | 64 bytes from 4a2841b5dac9 (172.21.0.2): icmp_seq=3 ttl=64 time=0.088 ms
https-1  | 
https-1  | --- https ping statistics ---
https-1  | 3 packets transmitted, 3 received, 0% packet loss, time 2009ms
https-1  | rtt min/avg/max/mdev = 0.088/0.203/0.404/0.142 ms
https-1  | =====dig https output====
https-1  | c.root-servers.net.
https-1  | l.root-servers.net.
https-1  | e.root-servers.net.
https-1  | d.root-servers.net.
https-1  | i.root-servers.net.
https-1  | b.root-servers.net.
https-1  | g.root-servers.net.
https-1  | m.root-servers.net.
https-1  | a.root-servers.net.
https-1  | f.root-servers.net.
https-1  | h.root-servers.net.
https-1  | j.root-servers.net.
https-1  | k.root-servers.net.

I have a permissions problem. I am running Photoprism inside a Docker container on Ubuntu 22.04. I want to use Intel QuickSync hardware transcoding. To do this, the app needs to access the /dev/dri device. I am attempting to get the app running without using priviledged: true in the docker-compose file. It works (ie, the app is able to use /dev/dri) when using priviledged: true. When I remove priviledged: true, the app reports this, even though devices:\n - /dev/dri is in the docker-compose.yml file:

$ docker-compose up -d
...
 ⠿ Container photoprism-photoprism-1  Starting                                                                                                                             1.9s 
Error response from daemon: error gathering device information while adding custom device "/dev/dri": no such file or directory

Currently I have Plex installed natively (ie, not using Docker), and it works with /dev/dri just fine. Here are the permissions on /dev/dri:

$ ls -al /dev/dri
total 0
drwxr-xr-x   3 root root        100 May 29 14:09 .
drwxr-xr-x  19 root root       5200 May 29 14:09 ..
drwxr-xr-x   2 root root         80 May 29 14:09 by-path
crw-rw----+  1 root render 226,   0 May 29 14:09 card0
crw-rw----+  1 root render 226, 128 May 29 14:09 renderD128

(render is a group name, but what is the meaning of 226, in the listing output?) Here are the details of the plex user (which is the user which is running plexmediaserver, which is working with /dev/dri):

$ id plex
uid=998(plex) gid=998(plex) groups=998(plex),44(video),109(render)

...and the user which is running docker-compose up -d

$ id myuser
uid=1000(myuser) gid=1000(myuser) groups=1000(myuser),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),120(lpadmin),131(lxd),132(sambashare)

I thought that since the plex user is working fine with /dev/dri, maybe I could use the plex user with Photoprism as well. But I was unable to get it working:

# inside docker compose
   user: "998:998"

The relevant parts of the docker-compose.yml file:

services:
  photoprism:
    image: photoprism/photoprism:latest
    ...
    ## Start as non-root user before initialization (supported: 0, 33, 50-99, 500-600, and 900-1200):
    user: "998:998"
    ## Share hardware devices with FFmpeg and TensorFlow (optional):
    devices:
      - "/dev/dri:/dev/dri"                         # Intel QSV
    ...

TL;DR: Works when I use docker privileged mode, doesn't work otherwise. Also, I could use some help understanding ls output which mysteriously includes an unexplained 226, in it (See above ls output).

I have a service defined via docker compose (see definition below). When I tried to start this service via docker-compose -f up --wait -d my_service, I get the error

Error response from daemon: failed to create task for container: failed to initialize logging driver: dial unix /dev/log: connect: protocol wrong type for socket

On my host server where I'm executing the docker compose cmd, I see the socket exists and my user has write perms:

srw-rw-rw-. 1 root root 0 Aug 29  2023 /dev/log

service definition:

my_service:
    command: 
    image: 
    volumes:
      - "/dev/log:/dev/log"
    logging:
      driver: "syslog"
      options:
        syslog-address: "unix:///dev/log"
        tag: "my_service"

Does anyone know what could be causing this error?

Reposting from [here](https://forums.docker.com/t/docker-compose-not-synchronising-file-changes-in-volume/79177) as I don't quite understand how the "solution" works. **Symptom:** As reported [here](https://forums.docker.com/t/docker-compose-not-synchronising-file-changes-in-volume/79177) : I mount my local files into the container for development. My docker-compose.yml file is as follows:

version: '3.7'
services:
  node:
    build: .
    command: npm run dev
    image: node
    ports:
      - "4000:3000"
    volumes:
      - .:/code
    working_dir: /code

when I run the Next JS server in a container, it will initially load fine, but any changes made afterwards will not be shown. This is same as my observation as well. **Answers:** There has been discussion on the issues of running docker in Windows 10 Pro, but I'm hosting in Linux for Linux, and the "solution" mentioned nothing about Windows either. **Solution:** The reported working "solution" is:

version: '3.7'
services:
  node:
    build: .
    command: npm run dev
    image: node
    ports:
      - "4000:3000"
    volumes:
      - type: bind
        source: .
        target: /code
    working_dir: /code

This, to my knowledge, would be the same as what used by OP, i.e.:

volumes:
      - .:/code

I.e., I think the question of why _"docker Compose not synchronising file changes in volume"_ was not clearly answered. And neither of the following pages: - https://stackoverflow.com/questions/44678042/named-docker-volume-not-updating-using-docker-compose - https://stackoverflow.com/questions/44251094/i-want-to-share-code-content-across-several-containers-using-docker-compose-volu/44265470 - https://stackoverflow.com/questions/42958573/docker-compose-recommended-way-to-use-data-containers My case: I have a php site hosted on nginx:

version: '3.7'
services:
    web:
        image: nginx:latest
        ports:
            - "80:8081"
        volumes:
            - ./nginx.conf:/etc/nginx/conf.d/default.conf:delegated
            - ./app:/app:delegated
    php:
        build:
            context: .
            dockerfile: PHP.Dockerfile
        volumes:
            - ./app:/app:delegated
        restart: always

This is my observation: - I do docker compose up - Then ^-C to stop it - Update code in host, - Re- do / start docker compose up - Check the modified volume-mounted file within docker compose container, - The file stays the same, as before the change.