1.6.4 rolled out?

Is 1.6.4 fully rolled out yet?
I’ve just stopped, deleted and restarted one of my nodes and docker seems to have pulled 1.5.2 again…

Yes it has. Do you have watchtower ?

“docker rm” only removes the running container “docker rmi” is to remove the images so that it gets re-downloaded. “docker ps -a” and “docker images” show the the two types of … thing (?!)
(I am making some assumptions about your question here obviously)

No, I’ve been reading bad things about it so would rather do it manually…

docker rmi storagenode didn’t work, interestingly…

EDIT: me being an idiot. Worked now. Thank you! :slight_smile:

There is a small club of us “watchtower deniers”. :rofl:

That’s good to know
I only wish there was some form of alerting when a node needs updating…

There is the alert in the dashboard
image

The email notification for node update is under review (as per last update I read).

There are few SNOs that have trouble with watchtower but it works for most. Personally there were a few updates that didn’t go through watchtower and I had to manually update but the latest went through without a hitch (knock on the wood).

1 Like

What’s about the alert ?

It is an example of an alert in the notifications that I was showing to the O.P. Not a problem for my node at the moment

1 Like

Interesting! I never got that on 1.5.2
Maybe you have to be more than one version out of date, but it’s good to know that there is some form of warning.

Thank you for the info! :smiley:

We are going to move back software update notifications to a SNOBoard in one of next releases. PR that doing that already been merged:
https://review.dev.storj.io/c/storj/storj/+/2143

3 Likes

I would advise against that. Watchtower works flawlessly for the vast majority of users, including me. Especially if you’re not that familiar with docker you shouldn’t rely on manual updates. Just let watchtower take care of it for you. If you’re worried about the node going down during an update, set up uptime robot so you get a notification if that happens.

1 Like

Seemed actually very straightforward to do it manually?

I’m going to be posted abroad for a few months, so may fire up Watchtower then, though.

Out of interest, does one instance of Watchtower update all my nodes or do I have to run multiple instances of it? (I really should use the search bar, I know…) :flushed:

Just run one. I think running multiple has actually caused some of the trouble people saw. You have to list all container names you want to update in the command though.

For example
docker run -d --restart=always --name watchtower -v /var/run/docker.sock:/var/run/docker.sock storjlabs/watchtower storagenode storagenode2 storagenode3 watchtower --stop-timeout 300s --interval 21600

I’ll give it a go. Nodes are fairly new so no great loss if it all goes wrong. Only to your credibility :wink: