Something weird is happening: I have moved a node. I am pretty sure before moving it was on version 1.53.1. Now after moving it is showing version 1.47.3 in the dashboard and is getting no more ingress. How can that be?
Watchtower is (and has been) running on this node. Now after roughly 30 hours no update yet.
I guess look in the log and see if it is trying to do its new internal update thing
There is nothing in the logs about watchtower, storagenode-updater, attempts to update or contacting version servers.
I believe it has max. 72 hours in which an auto update should happen otherwise something is broken.
Updates happen inside the container now. You should look at the container logs. Even if you have redirected logs to a file, that only redirects the logs for the storagenode, not for the updater.
1 Like
I think version 1.47.3 was watchtower updates only.
What tag are you using for the image?
storjlabs/storagenode:latest
in the node run command and none in the Watchtower.
Have you tried just pulling it manually? Also what architecture are you on?
Not yet, as I wanted to verify that auto-updates are working.
It is now 33 hours in the 72 hours period and I still hope it will update.
However what really puzzles me is that it is showing version 1.47.3 when it was 1.53.1 before the move.
Maybe there was an older image still on the machine you moved it to? If you turn on debug logging for watchtower, the logs will show a line when the next update will happen.
time="2022-05-12T04:30:25Z" level=debug msg="Scheduled next run: 2022-05-12 20:46:39 +0000 UTC"
Just add --debug to the docker run for the watchtower.
1 Like
That’s interesting. I added it, but had to restart watchtower. I shows: time="2022-05-12T06:26:47Z" level=debug msg="Starting Watchtower and scheduling first run: 2022-05-14 08:33:21 +0000 UTC m=+180394.410470515"
So let’s see what will happen in 4 minutes on May 14th?
Edit: It looks like this is UTC time? Then we need to add another 2 hrs?
Yes, we’re at +2 atm. Do you really want to wait that long though? Seems like you’d miss out on quite a bit of ingress. Or do you have other nodes on the same IP getting that ingress now?
Edit: I guess you could also try to restart watchtower a few times and hope the time gets closer to the 12 hours than 72 hours. 
But maybe you’re not as impatient as I am.
Actually I think I can wait. There’s loads of other stuff that keeps me busy currently.
If I restart watchtower it could also move towards the end of the 72 hours period, so I am unsure … 
1 Like
I seem to remember a runonce or one shot flag that got me banned from docker, because it skips the wait
It is updating now… 
Edit: And Uploads are back also. 
3 Likes