hyperbart
commented
2 years ago Although the container in some way seems to be up it is not listening on 7070 anymore:
bart@nas:~$ sudo lsof -i -P -n | grep LISTEN | grep 7070
bart@nas:~$Closed hyperbart closed 2 years ago
hyperbart
commented
2 years ago Although the container in some way seems to be up it is not listening on 7070 anymore:
bart@nas:~$ sudo lsof -i -P -n | grep LISTEN | grep 7070
bart@nas:~$Images are up to date:
bart@nas:~$ docker image ls | grep andi
andig/evcc latest 009715c90545 8 hours ago 53.1MB
andig/evcc <none> c5826aa62924 18 hours ago 53.1MB
andig/evcc <none> 063f8a351aac 2 weeks ago 49.7MB
rubinho76
commented
2 years ago Hab das gleiche Problem. Nach dem Upgrade auf 0.105.1 wird mir der Container als unhealthy angezeigt. Der native Webzugang funktioniert jedoch, allerdings über Traefik nicht mehr, was am unhealthy liegen könnte. Das Evcc Log sieht unspektakulär aus. Mein Docker läuft auf einer Synology DS916+ mit x86_64 SOC.
lg Rubinho
rubinho76
commented
2 years ago --Update-- Das Problem ist auch in der aktuellen Nightly und in der 0.105.0 vorhanden. Ein Downgrade auf 0.104.2 behebt vorerst das Problem. @hyperbart The reason why your webinterface is not working could be Traefik (Do you have Traefik in use?) Traefik does a healthcheck and if it is negative, no proxy will be established.
premultiply
commented
2 years ago Was genau tut denn der "Healthcheck" technisch?
hyperbart
commented
2 years ago I don’t have a Traefik running but I do have Cloudflare tunnel. But that does normally nog intervene with other containers. EVCC is only attached to an extra network where the cloudlared container also resides so it can tunnel traffic outwards.
i will try later today what happens if I take away the network, then it should run standalone.
@premultiply : no idea about the healthcheck.
rubinho76
commented
2 years ago @premultiply Ich kann dir das auch nicht wirklich sagen. Manche Images haben einen, manche nicht. Ich glaube @andig muss ihn bewusst eingebaut haben, oder hat sich einem Basisimage bedient, der ein Healthcheck integriert hat. Was der genau prüft, ist mir noch nicht klar. Im Container-Log sehe ich zumindest nichts.
premultiply
commented
2 years ago Was gibt denn http://evcc.local:7070/api/health (bzw. mit der IP) zurück im Browser?
rubinho76
commented
2 years ago Das hier...
SyntaxError: JSON.parse: unexpected character at line 1 column 1 of the JSON data
Keine Ahnung, ob das der Fehler ist oder der Healthcheck kaputt ist.
rubinho76
commented
2 years ago Kommando zurück. Wenn ich im Browser auf Rohdaten klicke, steht da "ok"
rubinho76
commented
2 years ago Das hier spuckt mir Docker inspect aus...
Output [main ] INFO 2022/10/15 08:56:56 evcc 0.105.1 [main ] ERROR 2022/10/15 08:56:56 health check failed
--Edit-- Ich hab den Container mal auf eine andere Maschine umgezogen, das Ergebnis ist das gleiche.
hyperbart
commented
2 years ago @premultiply : URL gives nothing because the port is not yet open. @andig : I just deployed but it still exits with the following error. See below
bart@nas:~$ docker stack ps EVCC --no-trunc
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS
4z4qjj2bp8zp4xh0azkaho4zz EVCC_evcc.1 andig/evcc:nightly@sha256:eb229e21cfbf01f361d71cff98d43583cbf604e29db26cd6258df6d3acd5bfd8 nas Running Starting about a minute ago
rjheacxcjlf61ut55dju6ze55 \_ EVCC_evcc.1 andig/evcc:nightly@sha256:eb229e21cfbf01f361d71cff98d43583cbf604e29db26cd6258df6d3acd5bfd8 nas Shutdown Failed about a minute ago "task: non-zero exit (1): dockerexec: unhealthy container"@hyperbart It probably doesn't help you, but for me the new version 0.105.2 works again. Did you delete the old image?
hyperbart
commented
2 years ago @rubinho76
By switching from nightly to latest it appears to be working again. Took some time to start the container apparently. (around a minute or so).
I'll check if switching to nightly works too.
hyperbart
commented
2 years ago Nightly still has the issue (I am not a developer, so don't know if that is to be expected). I once switched to nightly's to be able to enjoy the new GUI by @premultiply 👍 .
Switched back to latest and exactly after a minute the container comes up.
rubinho76
commented
2 years ago @hyperbart At the moment the "latest" is the newest version. The nightly is 6 hours old and still has the bug If you are happy with the stable versions, leave it on latest.
Describe the bug
I try to do weekly updates of EVCC. Running EVCC in a Docker Stack. After tonights "docker stack rm EVCC" and "docker stack deploy compose.yml EVCC" I noticed EVCC did not report healthy in my monitoring. The web interface is down. Weirdly it seems to be running 2 dockers at the same time?
Steps to reproduce
Remove the docker stack, pull in new images and redeploy.
Configuration details
Log details
Tried to remove the stack and redeploy:
Weirdly enough it says it is still running 104.2 ?
What type of operating system are you running?
Linux
Version
Linux nas 5.15.0-50-generic #56-Ubuntu SMP Tue Sep 20 13:23:26 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 22.04.1 LTS Release: 22.04 Codename: jammy