Please, update your watchtower command as soon as possible

k, so I’ll wait for the next update and see if it will work then. thx…

1 Like

Hi @Alexey!

I would like to suggest adding --cleanup to our auto-update documentation, the main reason is clean up old images from host after update. Here is the log:

Complete command:
docker run -d --restart=always --name watchtower -v /var/run/docker.sock:/var/run/docker.sock storjlabs/watchtower storagenode watchtower --stop-timeout 300s --interval 21600 --cleanup

Log:

Summary
2019-10-16T22:46:12.681371985Z time="2019-10-16T22:46:12Z" level=info msg="Found new storjlabs/storagenode:beta image (sha256:694739e41afd626821a7747c9a74259b6d6000a71baa37767fbfe2ae3e1c33ef)"
2019-10-16T22:46:12.681500166Z time="2019-10-16T22:46:12Z" level=debug msg="No pre-update command supplied. Skipping"
2019-10-16T22:46:12.681524151Z time="2019-10-16T22:46:12Z" level=info msg="Executing pre-update command."
2019-10-16T22:46:13.078985079Z time="2019-10-16T22:46:13Z" level=info msg="Stopping /storagenode (4371061a6cbed03c35e09300548fa420e8c576925399971d3ad8a891252ec160) with SIGTERM"
2019-10-16T22:46:21.231769364Z time="2019-10-16T22:46:21Z" level=debug msg="Removing container 4371061a6cbed03c35e09300548fa420e8c576925399971d3ad8a891252ec160"
2019-10-16T22:46:21.857981063Z time="2019-10-16T22:46:21Z" level=debug msg="This is the watchtower container /watchtower"
2019-10-16T22:46:21.858153000Z time="2019-10-16T22:46:21Z" level=info msg="Creating /storagenode"
2019-10-16T22:46:23.313132766Z time="2019-10-16T22:46:23Z" level=debug msg="Starting container /storagenode (97f67b1ab3da3e5a85f658b4a5c188986f353df8db3f9783e67caa21f24a734d)"
2019-10-16T22:46:24.043921625Z time="2019-10-16T22:46:24Z" level=info msg="Removing image sha256:0e41d38870866613ddfcd5ce970fef1651d0c3cb2759a20b4f18fb817b6a1704"
2019-10-16T22:46:25.301353073Z time="2019-10-16T22:46:25Z" level=debug msg="Scheduled next run: 2019-10-17 04:46:00 +0000 UTC"

Not needed. If one wants to cleanup they could run a docker image prune in any time.
I do not like automated deletes

1 Like

@Alexey

Update: I noticed this evening that the latest version was v0.23.5, however, even though I still had watchtower running I was still at v0.23.2. So, I was still 3 updates behind. The docker logs -t watchtower command returned nothing. I’ve now manually updated my storagenode to the latest version but it still looks like watchtower isn’t working for me.

Afaik, the latest instance of watchtower now has a random poll interval in the range of 12 to 72 hours. This is intentional to avoid that almost all nodes update at the same time (and be offline during that time). So please do not try to force your node to update immediately after each release. Also note that not all releases are updates for SNOs. If your node is not updated automatically after 72 hours, you can always use the manual update method as described in https://documentation.storj.io/setup/cli/software-updates#manual-updates

1 Like

The point is the last several updates I have had to do manually. The watchtower has not updated my storagenode for quite some time now. There is nothing in the watchtower log to indicate there was even an attempt to update the new version.

Please, stop and remove the watchtower container and run it back: https://documentation.storj.io/setup/cli/software-updates#automatic-updates

@Alexey

Ok, I’ve just done that, however, let me just say that I did exactly the same thing about a week ago.

Please, try to pull the latest image before the run

@Alexey

Did this, is that what is expected with the pull?

storj@pine64 : ~ $ docker pull storjlabs/watchtower --platform aarch64
Using default tag: latest
latest: Pulling from storjlabs/watchtower
Digest: sha256:6accfdce9f39eedb957dd8ce20db3eff0872b33a4708b897ecaddfe967b2b941
Status: Image is up to date for storjlabs/watchtower:latest

Just want to clarify before moving on.