This issue concerns the self-hosted sync server (git hub dot com/ankitects/anki/tree/main/docs/syncserver).
After the update to 24.11, the healthcheck endpoint should work. However, building using the provided Dockerfile or the docker hub images (hub dot docker dot com/r/jeankhawand/anki-sync-server) creates a healthcheck endpoint that still returns unhealthy no matter what.
I was able to fix it making the following change to the Dockerfile:
- HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
- CMD wget -qO- http://localhost:${SYNC_PORT}/health || exit 1
+ HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+ CMD wget -qO- http://0.0.0.0:${SYNC_PORT}/health || exit 1
Perhaps this is due to my configuration, but it may be worth documenting this possible fix it if not done already.
My configuration:
- Anki sync server running in docker
- Nginx running on Ubuntu 24.04.1 LTS host (not containerized)
docker-compose.yml:
services:
anki-server-sync:
build:
context: .
args:
ANKI_VERSION: 24.11
image: anki-server-sync:latest
environment:
SYNC_PORT: 5805
SYNC_USER1: ***:***
SYNC_USER2: ***:***
ports:
- 5805:5805
volumes:
- ./syncserver:/home/anki/.syncserver
restart: unless-stopped
Dockerfile: see change above relative to the source file (git hub dot com/ankitects/anki/blob/main/docs/syncserver/Dockerfile)
Nginx config:
upstream anki-livesync {
server 127.0.0.1:5805;
}
server {
listen 443 ssl;
server_name ***;
ssl_certificate cert/domain.cert.pem;
ssl_certificate_key cert/private.key.pem;
ssl_protocols TLSv1 TLSv1.1 TLSv1.2 TLSv1.3;
ssl_ciphers HIGH:!aNULL:!MD5;
location / {
proxy_pass http://anki-livesync;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-for $remote_addr;
proxy_set_header X-Forwarded-Proto $scheme;
client_max_body_size 100M;
}
}