Runtipi•2d ago

Apps containers not starting after 4.0.0 upgrade

Hello, After running the migration script to 4.0.0, it looks like none of my applications are starting. The upgrade script went through successfully. I do have these errors in the logs though :

2025-05-01T08:36:00.882Z - info >       Logs flushed
2025-05-01T08:36:00.883Z - info >       Running version: v4.0.0
2025-05-01T08:36:00.884Z - info >       Generating system env file...
2025-05-01T08:36:01.979Z - info >       Creating traefik folders
2025-05-01T08:36:01.981Z - info >       Copying traefik files
2025-05-01T08:36:01.985Z - info >       Creating base folders
2025-05-01T08:36:01.986Z - info >       Creating media folders
2025-05-01T08:36:02.020Z - info >       TLS certificate for tipi.local already exists
2025-05-01T08:36:02.727Z - error >      RabbitMQ connection error connect ECONNREFUSED 172.18.0.5:5672 Error: connect ECONNREFUSED 172.18.0.5:5672
    at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1636:16)
2025-05-01T08:36:02.727Z - warn >       RabbitMQ connection lost, attempting to reconnect...
2025-05-01T08:36:04.079Z - info >       Connected to the queue
2025-05-01T08:36:04.531Z - info >       Connected to the queue
2025-05-01T08:36:06.775Z - info >       Connected to the queue

2025-05-01T08:36:00.882Z - info >       Logs flushed
2025-05-01T08:36:00.883Z - info >       Running version: v4.0.0
2025-05-01T08:36:00.884Z - info >       Generating system env file...
2025-05-01T08:36:01.979Z - info >       Creating traefik folders
2025-05-01T08:36:01.981Z - info >       Copying traefik files
2025-05-01T08:36:01.985Z - info >       Creating base folders
2025-05-01T08:36:01.986Z - info >       Creating media folders
2025-05-01T08:36:02.020Z - info >       TLS certificate for tipi.local already exists
2025-05-01T08:36:02.727Z - error >      RabbitMQ connection error connect ECONNREFUSED 172.18.0.5:5672 Error: connect ECONNREFUSED 172.18.0.5:5672
    at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1636:16)
2025-05-01T08:36:02.727Z - warn >       RabbitMQ connection lost, attempting to reconnect...
2025-05-01T08:36:04.079Z - info >       Connected to the queue
2025-05-01T08:36:04.531Z - info >       Connected to the queue
2025-05-01T08:36:06.775Z - info >       Connected to the queue

But I think that none of the apps are starting (specially tailscale because I am using it to access the tipi GUI on port 8080 since the machine is on a VPS)

15 Replies

frankynovOP•2d ago

My docker ps is the folllowing :

root@hetzner-arm:~/runtipi# docker ps
CONTAINER ID   IMAGE                             COMMAND                  CREATED         STATUS                   PORTS                                                                                      NAMES
9eeddbf72e11   traefik:v3.2                      "/entrypoint.sh --pr…"   3 minutes ago   Up 3 minutes             0.0.0.0:8080->80/tcp, [::]:8080->80/tcp, 0.0.0.0:8443->443/tcp, [::]:8443->443/tcp         runtipi-reverse-proxy
338b12595221   ghcr.io/runtipi/runtipi:v4.0.0    "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   3000/tcp                                                                                   runtipi
6282949ed690   postgres:14                       "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   5432/tcp                                                                                   runtipi-db
203f005168ec   rabbitmq:4-alpine                 "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   4369/tcp, 5671-5672/tcp, 15691-15692/tcp, 25672/tcp                                        runtipi-queue
1b299a8d52f1   jc21/nginx-proxy-manager:latest   "/init"                  5 weeks ago     Up 10 minutes            0.0.0.0:80-81->80-81/tcp, [::]:80-81->80-81/tcp, 0.0.0.0:443->443/tcp, [::]:443->443/tcp   npm-app-1

root@hetzner-arm:~/runtipi# docker ps
CONTAINER ID   IMAGE                             COMMAND                  CREATED         STATUS                   PORTS                                                                                      NAMES
9eeddbf72e11   traefik:v3.2                      "/entrypoint.sh --pr…"   3 minutes ago   Up 3 minutes             0.0.0.0:8080->80/tcp, [::]:8080->80/tcp, 0.0.0.0:8443->443/tcp, [::]:8443->443/tcp         runtipi-reverse-proxy
338b12595221   ghcr.io/runtipi/runtipi:v4.0.0    "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   3000/tcp                                                                                   runtipi
6282949ed690   postgres:14                       "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   5432/tcp                                                                                   runtipi-db
203f005168ec   rabbitmq:4-alpine                 "docker-entrypoint.s…"   3 minutes ago   Up 3 minutes (healthy)   4369/tcp, 5671-5672/tcp, 15691-15692/tcp, 25672/tcp                                        runtipi-queue
1b299a8d52f1   jc21/nginx-proxy-manager:latest   "/init"                  5 weeks ago     Up 10 minutes            0.0.0.0:80-81->80-81/tcp, [::]:80-81->80-81/tcp, 0.0.0.0:443->443/tcp, [::]:443->443/tcp   npm-app-1

Stavros•2d ago

Alright The issue is that the queue probably didn't have enough time to start up so it shows as unhealthy

frankynovOP•2d ago

I did restart the machine & perform a ./runtipi restart

Stavros•2d ago

We can manually increase the timeout You will need to create the user-config/tipi-compose.yml file And add

services:
  runtipi-queue:
    healthcheck:
      test: rabbitmq-diagnostics -q ping
      interval: 5s
      timeout: 15s
      retries: 20
      start_period: 15s

services:
  runtipi-queue:
    healthcheck:
      test: rabbitmq-diagnostics -q ping
      interval: 5s
      timeout: 15s
      retries: 20
      start_period: 15s

Then restart and this time the queue should have enough time to report as healthy

frankynovOP•2d ago

Should it be created in the new subfolder "migrated" under user-config folder ? I did put it here, no more error in the log but still the app containers don't start :

root@hetzner-arm:~/runtipi/user-config# ls -la
total 16
drwxrwxrwx  3 root root 4096 May  1 10:50 .
drwxr-xr-x 14 root root 4096 May  1 10:32 ..
drwxrwxrwx  7 root root 4096 May  1 10:32 migrated
-rwxrwxrwx  1 root root  165 May  1 10:50 tipi-compose.yml

root@hetzner-arm:~/runtipi/user-config# ls -la
total 16
drwxrwxrwx  3 root root 4096 May  1 10:50 .
drwxr-xr-x 14 root root 4096 May  1 10:32 ..
drwxrwxrwx  7 root root 4096 May  1 10:32 migrated
-rwxrwxrwx  1 root root  165 May  1 10:50 tipi-compose.yml

Stavros•2d ago

No, it should be created inside the user config folder not in subfolders, the migrated one contains all of the user config for the default repo you have which is called migrated

frankynovOP•2d ago

root@hetzner-arm:~/runtipi# cat logs/app.log
2025-05-01T08:51:00.429Z - info >       Logs flushed
2025-05-01T08:51:00.430Z - info >       Running version: v4.0.0
2025-05-01T08:51:00.431Z - info >       Generating system env file...
2025-05-01T08:51:00.959Z - info >       Creating traefik folders
2025-05-01T08:51:00.961Z - info >       Copying traefik files
2025-05-01T08:51:00.962Z - info >       Creating base folders
2025-05-01T08:51:00.963Z - info >       Creating media folders
2025-05-01T08:51:00.989Z - info >       TLS certificate for tipi.local already exists

root@hetzner-arm:~/runtipi# cat logs/app.log
2025-05-01T08:51:00.429Z - info >       Logs flushed
2025-05-01T08:51:00.430Z - info >       Running version: v4.0.0
2025-05-01T08:51:00.431Z - info >       Generating system env file...
2025-05-01T08:51:00.959Z - info >       Creating traefik folders
2025-05-01T08:51:00.961Z - info >       Copying traefik files
2025-05-01T08:51:00.962Z - info >       Creating base folders
2025-05-01T08:51:00.963Z - info >       Creating media folders
2025-05-01T08:51:00.989Z - info >       TLS certificate for tipi.local already exists

Stavros•2d ago

If for example you add a new appstore called myapps and want to customize an app using user config then you will go to the user config directory, make a myapps directory and for every app add your user config as you would normally do That's good It means tipi started Can you check if you can access it?

frankynovOP•2d ago

indeed, internally it is started, I can see the login page with curl on localhost :

root@hetzner-arm:~/runtipi# curl localhost:8080
<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8" />
    <script src="/js/tabler.min.js" async></script>
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
    <link
      rel="apple-touch-icon"
      sizes="180x180"
      href="/icons/apple-touch-icon.png"
    />
    <link
      rel="icon"
      type="image/png"
      sizes="32x32"
      href="/icons/favicon-32x32.png"
    />
    <link
      rel="icon"
      type="image/png"
      sizes="16x16"
      href="/icons/favicon-16x16.png"
    />
    <link rel="manifest" href="/icons/site.webmanifest" />
    <title>Runtipi</title>
    <script type="module" crossorigin src="/assets/index-DGka_Jet.js"></script>
    <link rel="stylesheet" crossorigin href="/assets/index-DL4jEUj1.css">
  </head>

  <body>
    <div id="root"></div>
  </body>
</html>

root@hetzner-arm:~/runtipi# curl localhost:8080
<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8" />
    <script src="/js/tabler.min.js" async></script>
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
    <link
      rel="apple-touch-icon"
      sizes="180x180"
      href="/icons/apple-touch-icon.png"
    />
    <link
      rel="icon"
      type="image/png"
      sizes="32x32"
      href="/icons/favicon-32x32.png"
    />
    <link
      rel="icon"
      type="image/png"
      sizes="16x16"
      href="/icons/favicon-16x16.png"
    />
    <link rel="manifest" href="/icons/site.webmanifest" />
    <title>Runtipi</title>
    <script type="module" crossorigin src="/assets/index-DGka_Jet.js"></script>
    <link rel="stylesheet" crossorigin href="/assets/index-DL4jEUj1.css">
  </head>

  <body>
    <div id="root"></div>
  </body>
</html>

But I don't know why the apps won't start 😦 I do have a recent docker : Docker version 28.1.1, build 4eba377

Stavros•2d ago

The first time you need to start them manually Not through the cli In v4 the cli commands are not supported, we will include them in a future release

frankynovOP•2d ago

Aaaah ok ! So, I did briefly open the firewall to 8080, connected to tipi GUI from the WAN and started tailscale. From there I disabled again the firewall and started all the other containers and apps in the GUI. All seems to be good, and I validated with a reboot of the machine, all apps are starting again

Stavros•2d ago

Cool

frankynovOP•2d ago

Thanks a lot for your help ! I hadn't seen that manual restart step needed in the breaking update web page so maybe worth to add it there ? 🙂

Stavros•2d ago

Yeah we should probably add it, we will also release a patch fix for the queue timeout so as you don't need the user config

frankynovOP•2d ago

Excellent. Thank you very much ! I'll validate everything is ok and then will delete the migration-backups folder. Cheers !

Gaming

Programming

Apps containers not starting after 4.0.0 upgrade

Did you find this page helpful?