Using watchtower to monitor without pulling new images

[latimeria68]

Hello,
I have started today using watchtower through portainer stack with compose file below; everything seems working except that watchtower does not detect newer images on docker hub. Is this some kind of known problem with portainer?
The scope is to just have an email notification when a new image is released.
Thank you

version: "2"
services:
  watchtower:
    image: containrrr/watchtower
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    environment:
      - TZ=Europe/Rome
      - WATCHTOWER_DEBUG=true
      - WATCHTOWER_NOTIFICATIONS_LEVEL=debug
      - WATCHTOWER_CLEANUP=false
      - DOCKER_API_VERSION=1.24
      - WATCHTOWER_INCLUDE_STOPPED=true
      - WATCHTOWER_POLL_INTERVAL=3600
      - WATCHTOWER_NO_PULL=true
      - WATCHTOWER_MONITOR_ONLY=true
      - WATCHTOWER_NOTIFICATIONS=email
      - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_TLS_SKIP_VERIFY=false
      - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_PORT=587
      - WATCHTOWER_NOTIFICATION_EMAIL_FROM=myemail@gmail.com
      - WATCHTOWER_NOTIFICATION_EMAIL_TO=abc@xyz.com
      - WATCHTOWER_NOTIFICATION_EMAIL_SERVER=smtp.gmail.com
      - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_USER=myemail@gmail.com
      - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_PASSWORD=myapppwd
      - WATCHTOWER_NOTIFICATION_EMAIL_DELAY=2
    restart: unless-stopped

from the log:

time="2020-05-27T23:56:47+02:00" level=debug msg="Scheduled next run: 2020-05-27 23:59:47 +0200 CEST"
time="2020-05-27T23:59:47+02:00" level=debug msg="Checking containers for updated images"
time="2020-05-27T23:59:47+02:00" level=debug msg="Retrieving containers including stopped and exited"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /watchtower_watchtower_1"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /diun"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /nextcloud"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /letsencrypt"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /mariadb"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /droppy"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /transmission"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /sickchill"
time="2020-05-27T23:59:47+02:00" level=debug msg="Skipping image pull."
time="2020-05-27T23:59:47+02:00" level=debug msg="No new images found for /portainer"
time="2020-05-27T23:59:47+02:00" level=debug msg="Scheduled next run: 2020-05-28 00:02:47 +0200 CEST"

[simskij]

Hi, and welcome to the community! 🎉

Watchtower works great with portainer, I've been using it this way myself for the last couple of years. I noticed though that you've got both WATCHTOWER_NO_PULL and WATCHTOWER_MONITOR_ONLY enabled.

To quote the docs for --no-pull:

Do not pull new images. When this flag is specified, watchtower will not attempt to pull new images from the registry. Instead it will only monitor the local image cache for changes. Use this option if you are building new images directly on the Docker host without pushing them to a registry.

What this means is that; as the only way watchtower is able to check for new remote images is to pull them, you can't really disable pulling new images and still getting notifications about whether a new remote image is available.

https://docker-notify.com/ for instance seems to send an email digest with updated images. Maybe this could be what you're looking for? Do note though that I've not tried this service, so I can't really vouch for it's quality.

[piksel]

Maybe we should warn/fatal on doing this combination of options, since Watchtower will essentially be useless with both those specified?

[simskij]

Good suggestion. Created #570 to track this.

[latimeria68]

Hello,
so which option set alone will make watchtower just send an email alert when a new image version is released, without pulling and updating? or maybe watchtower simply cannot do it.
Thank you

[piksel]

Not without pulling, no.

[simskij]

WATCHTOWER_MONITOR_ONLY will pull but not update. This is as close as you'll get with watchtower I'm afraid.

[breeegz]

Does "Monitor Only" generate a report for Notifications (for instance, Discord/Slack)?

[piksel]

@breeegz Yes. You probably want to enable --notification-report, though.

[chris4prez]

@breeegz Yes. You probably want to enable --notification-report, though.

I'm looking to do similar to what's being described in this thread, and want to augment the following notification-report template to provide:

Current Output
7 Scanned, 0 Updated, 0 Failed

• /watchtower (containrrr/watchtower:latest): Fresh

Desired Output
7 Scanned, 1 Found New, 0 Updated, 0 Failed

• /watchtower (containrrr/watchtower:latest): Fresh
• /heimdall (heimdall:latest):Found New

However, I'm unable to find a list of the log entries or notification.data structs I can reference to augment below template I'm using to add the ability to reference a new image found but not deployed due to monitoring only.

{{- if .Report -}}
{{- with .Report -}}
{{len .Scanned}} Scanned, {{len .Updated}} Updated, {{len .Failed}} Failed
{{- range .Updated}}

• {{.Name}} ({{.ImageName}}): {{.CurrentImageID.ShortID}} updated to {{.LatestImageID.ShortID}}
  {{- end -}}
  {{- range .Fresh}}
• {{.Name}} ({{.ImageName}}): {{.State}}
  {{- end -}}
  {{- range .Skipped}}
• {{.Name}} ({{.ImageName}}): {{.State}}: {{.Error}}
  {{- end -}}
  {{- range .Failed}}
• {{.Name}} ({{.ImageName}}): {{.State}}: {{.Error}}
  {{- end -}}
  {{- end -}}
  {{- else -}}
  {{range .Entries -}}{{.Message}}{{"\n"}}{{- end -}}
  {{- end -}}

[unstressable]

here is my compose file, with some advantages for the same functionality:

version: "2"
services:
watchtower:
image: containrrr/watchtower
container_name: watchtower

volumes:
  - /var/run/docker.sock:/var/run/docker.sock
environment:
  - TZ=Europe/Rome
  - WATCHTOWER_DEBUG=true
  - WATCHTOWER_NOTIFICATIONS_LEVEL=debug
  - WATCHTOWER_CLEANUP=false
  - DOCKER_API_VERSION=1.24
  - WATCHTOWER_INCLUDE_STOPPED=true
  - WATCHTOWER_POLL_INTERVAL=36000 #Check every 10h
  - WATCHTOWER_MONITOR_ONLY=true
  - WATCHTOWER_NOTIFICATIONS=email
  - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_TLS_SKIP_VERIFY=false
  - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_PORT=587
  - WATCHTOWER_NOTIFICATION_EMAIL_FROM=my.mail@gmail.com
  - WATCHTOWER_NOTIFICATION_EMAIL_TO=my.mail@gmail.com
  - WATCHTOWER_NOTIFICATION_EMAIL_SERVER=smtp.gmail.com
  - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_USER=my.mail@gmail.com
  - WATCHTOWER_NOTIFICATION_EMAIL_SERVER_PASSWORD=my-app-password #can be found at google account under security->2-factor-authentication
  - WATCHTOWER_NOTIFICATION_EMAIL_DELAY=2
restart: unless-stopped

[unstressable]

there is just one more thing:
I get emails like:
"Making sure everything is sane before starting"
"Retrieving running, stopped and exited containers"
"There are no additional watchtower containers"

Everytime I update my compose/stack and redeploy.
Is there a way to prevent this?