Watchtower: Max --interval is 300s?

[RyanRamchandar]

My docker-compose.yml file sets the --interval to 3600s (1 hour) but it appears the interval is still 300s (5 min) based on debug logs from the container. Is the max interval 300?

docker logs app_watchtower_1

time="2018-07-12T20:40:10Z" level=info msg="First run: 2018-07-12 20:45:10 +0000 UTC" 
time="2018-07-12T20:45:20Z" level=info msg="Unable to update container /dashboard, err='Error response from daemon: Get https://registry-1.docker.io/v2/: dial tcp: lookup registry-1.docker.io on 127.0.0.53:53: read udp 127.0.0.1:60127->127.0.0.53:53: i/o timeout'. Proceeding to next." 
time="2018-07-12T20:50:20Z" level=info msg="Unable to update container /dashboard, err='Error response from daemon: Get https://registry-1.docker.io/v2/: dial tcp: lookup registry-1.docker.io on 127.0.0.53:53: read udp 127.0.0.1:43078->127.0.0.53:53: i/o timeout'. Proceeding to next." 

Note: network errors due to no internet connection during the test

docker inspect --format='{{.Config.Cmd}}' app_watchtower_1

[dashboard --interval 3600 --cleanup --debug]

docker-compose.yml

version: "3"
services:
  watchtower:
    image: v2tec/watchtower:armhf-0.3.0
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    command: ["dashboard", "--interval", "3600", "--cleanup"]

[cnrmck]

Hi @RyanRamchandar, I'll try running mine with a 10 minute timer to see if i'm having any trouble. I'm also running an armhf image. One sec

[cnrmck]

Here's what I'm running:

docker run -it   --name watchtower   -v /home/pi/.docker/config.json:/config.json   -v /var/run/docker.sock:/var/run/docker.sock   --env-file ./env.list v2tec/watchtower:armhf-latest --interval 600 --debug

It seems like it's working for me:

DEBU[0000] Retrieving running containers                
INFO[0000] First run: 2018-07-16 17:04:33 +0000 UTC     
DEBU[0599] Checking containers for updated images       
DEBU[0599] Retrieving running containers                
DEBU[0599] Pulling repo/name:latest for /thirsty_easley 
DEBU[0599] Loaded auth credentials { } for repo/name 
DEBU[0601] No new images found for /thirsty_easley      
DEBU[0601] Pulling v2tec/watchtower:armhf-latest for /watchtower 
DEBU[0601] Loaded auth credentials { } for v2tec/watchtower:armhf-latest 
DEBU[0602] No new images found for /watchtower          
DEBU[0602] Scheduled next run: 2018-07-16 17:14:33 +0000 UTC 

[RyanRamchandar]

Okay, thanks for testing. I'll check again and report back.

[cnrmck]

DEBU[0602] Scheduled next run: 2018-07-16 17:14:33 +0000 UTC 
...
DEBU[1202] Scheduled next run: 2018-07-16 17:24:33 +0000 UTC 

checked further, seems to work.

Maybe try using v2tec/watchtower:armhf-latest. I don't know how the versioning works exactly, but it's working for me.

Could you give me your uname -a?

Also, if it still doesn't work for you, you can always use a chron instead. For hourly check the syntax is:
--schedule "0 0 * * * *" in place of the --interval flag

[RyanRamchandar]

# uname -a
Linux umbrela-bridge 4.15.0-23-generic #25-Ubuntu SMP 
Wed May 23 17:59:52 UTC 2018 aarch64 aarch64 aarch64 GNU/Linux

If --interval doesn't work I'll try --schedule. Thanks for the pro tip!

[RyanRamchandar]

Hmm seems like neither command is working for me now even when I reduce the interval to 30s. Also, adding --debug doesn't seem to output any debug messages.

Reduced interval to 30s w/ --debug

version: "3"
services:
  watchtower:
    image: v2tec/watchtower:armhf-latest
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    command: ["dashboard", "--interval", "30", "--cleanup",  "--debug"]

Trying --schedule

version: "3"
services:
  watchtower:
    image: v2tec/watchtower:armhf-latest
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    command: ['dashboard', '--schedule', '"30 * * * * *"', '--cleanup',  '--debug']

In both cases, I see a similar log:

# docker logs app_watchtower_1 -f
time="2018-07-17T20:41:51Z" level=info msg="First run: 2018-07-17 20:46:51 +0000 UTC" 

Could this be an error in my docker-compose.yml formatting?

Or, could this be because my device is arm64v8 yet the image is armhf. Maybe there is some incompatibility here. See #206

[cnrmck]

I’m not even sure what the dashboard command is. Is that a shell of some
sort?

On Tue, Jul 17, 2018 at 2:54 PM Ryan Ramchandar [e-mail]
wrote:

Hmm seems like neither command is working for me now even when I reduce
the interval to 30s. Also, adding --debug doesn't seem to output any
debug messages.

Reduced interval to 30s w/ --debug

version: "3"
services:
watchtower:
image: v2tec/watchtower:armhf-latest
volumes:
- /var/run/docker.sock:/var/run/docker.sock
command: ["dashboard", "--interval", "30", "--cleanup", "--debug"]

Trying --schedule

version: "3"
services:
watchtower:
image: v2tec/watchtower:armhf-latest
volumes:
- /var/run/docker.sock:/var/run/docker.sock
command: ['dashboard', '--schedule', '"30 * * * * *"', '--cleanup', '--debug']

In both cases, I see a similar log:

docker logs app_watchtower_1 -f

time="2018-07-17T20:41:51Z" level=info msg="First run: 2018-07-17 20:46:51 +0000 UTC"

Could this be an error in my docker-compose.yml formatting?

Or, could this be because my device is arm64v8 yet the image is armhf.
Maybe there is some incompatibility here. See #206
https://github.com/v2tec/watchtower/issues/206

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/v2tec/watchtower/issues/207#issuecomment-405724410,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AVCjgxeFdgWV9l2xmgUx8ml7eSOPuwRIks5uHk8MgaJpZM4VNrYR
.

[RyanRamchandar]

dashboard is the name of the container that should be kept up to date.

[cnrmck]

Gotcha, have you tried '--interval 30' as one string rather than 2?

On Tue, Jul 17, 2018 at 10:52 PM Ryan Ramchandar [e-mail]
wrote:

dashboard is the name of the container that should be kept up to date.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/v2tec/watchtower/issues/207#issuecomment-405808259,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AVCjgzrTBbvP1XRGtFw9A0vP3OkVLWrAks5uHr8AgaJpZM4VNrYR
.

[simskij]

Closing due to lack of activity. As @cnrmck said, it should likely be entered as one string.

With that said, I can't see anything in either our code or in github.com/robfig/cron, doing a quick read through, suggesting that it shouldn't support larger numbers than 300.