Immich-Proxy timing out for 1st time setup Android app "Building Timeline"
It seems that the nginx timing out for a first time setup of the Immich Android app.
Getting these in the Immich-Proxy logs:
2023/05/27 02:10:14 [error] 40#40: *196 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.147.17.20, server: , request: "GET /api/asset HTTP/1.1", upstream: "http://172.18.0.7:3001/asset", host: "10.147.17.2:2283"
I tried to update the /etc/nginx/conf.d/default.conf file but I can't as its not writable with the nginx user.31 Replies
nginx timing out for a first time setup of the Immich Android app.I think this is unrelated to the android app Are you seeing the Android app's building timeline keep going on?
Yep. Immich-Web is fine.
lemme try to get my wife's iPad Prod...
Have you choose an album/folder in the backup page yet?
yep
to an album that only has 1 image in the backup
that show up after a while
and only that 1 image
Which phone model is this?
Samsung Galaxy S22
Stack Overflow
NGINX: upstream timed out (110: Connection timed out) while reading...
I have Puma running as the upstream app server and Riak as my background db cluster. When I send a request that map-reduces a chunk of data for about 25K users and returns it from Riak to the app, ...
So you are seeing the local assets sometimes show up but not the remote assets that were on the server, correct?
i do have a lot in my libarary - at almost 200K+ obj
local assets always shows up but not the remote assets
Are you accessing with your local IP or over DNS?
i'm actuallly over Zerotier VPN as I wanted to access it outside
lemme try a to switch to a the local network
Yes, let's try that first
same thing: 2023/05/27 02:30:54 [error] 41#41: *103 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.86.199, server: , request: "GET /api/asset HTTP/1.1", upstream: "http://172.18.0.7:3001/asset", host: "192.168.86.7:2283"
@zody probably we run into resources management here š¤
hmm... lemme try to override:
20-envsubst-on-templates.sh: Running envsubst on /etc/nginx/templates/default.conf.template to /etc/nginx/conf.d/default.conf
to add proxy_read_timeout 3600;
You probably have to rebuild the image
yeah.. getting permission denied as its login as nginx by default
can't su or sudo
lemme see if I can build it... i don't really qualify as a "dev" but more of a "hacker" lol
I think you will need to clone the Immich repo, change the content then build for production
found a workaround...
added this to the docker-compose:
immich-proxy:
container_name: immich_proxy
image: ghcr.io/immich-app/immich-proxy:release
environment:
# Make sure these values get passed through from the env file
- IMMICH_SERVER_URL
- IMMICH_WEB_URL
ports:
- 2283:8080
depends_on:
- immich-web
labels:
- "com.centurylinklabs.watchtower.monitor-only=true"
volumes:
- /mnt/immich/proxy/templates:/etc/nginx/templates
restart: unless-stoppedah nice
created the conf in my host and mounted it
let's see how it goes
my library is huge!
let me know, we can increase the timeout
yeah it is not your fault, we need to find a way to handle large library like yours
woohoo!! works!!!
Wooh nice!
got the timeline now
yeah its default timeout... not sure to what
added it here at this point:
proxy_buffering off;
proxy_request_buffering off;
proxy_buffer_size 16k;
proxy_busy_buffers_size 24k;
proxy_buffers 64 4k;
proxy_force_ranges on;
** proxy_read_timeout 3600;**
in default.conf.template and restarted the docker-compose after mounting itThis is very valuable
thank you for helping us resolve this
yeah at 200K images.. am I the biggest yet? lol
probably not
we have people with like terabytes of assets
š
i'm sure
anyways.. yeah.. if we can increase the timeout settings at the proxy
loving it!!!
thanks man!
so in the article above: keepalive is better than timeout for better resource management?
have not much of experience with this, so gonna try keepalive then...
since its me and wife using I'm good with what I have. it would depend on your use case then
might be best to just document to add to their docker-compose:
volumes:
- /mnt/immich/proxy/templates:/etc/nginx/templates
and let folks update based on their use case at their default.conf.template file
nope. that method suggested to keepalive didn't work at all. reverting back to proxy_read_timeout 3600;