I have my nextcloud server running version `Nextcloud Hub 25 Autumn (32.0.3)` and my android client is up to date version 3.35.0.
In the last few days I’ve not been able to upload any files using the auto upload feature. I’ve tried clearing the app cache but that hasn’t changed anything. Everything that needs to upload gives a generic unknown error message.
Is there any suggestions on how I can troubleshoot this issue? Thanks a ton
PS. Ignore the huge number of pending files. Most of those are existing images after clearing my cache.
There is a huge list of various failure messages; that one is just a generic fallback.
Can you look a bit at your web server, reverse proxy,.and Nextcloud logs to see if the transactions are getting that far? And, if so, what their return codes/etc are?
Thanks for the tips. I’m running aio docker. The docker container logs (from Apache, nextcloud, database and mastercontainer) have no errors logged. I’m running the aio behind nginx proxy manager. NPM doesn’t seem to have any errors logged either.
The /settings/admin/logging shows a lot of errors with the antivirus (`Tried to scan non file`). I don’t think it is related but it might be causing some issues. The only error level log is a connection drop during curl ``https://nextcloud.com
This seems a broader issue because I have a very similar one. Since couple of days the auto photo upload does not work on either of our (my wife’s and mine) Samsung Galaxy (android) phone. I did not changed anything on the server side, which is a self-hosted one on TrueNAS or on the phone.
Nextcloud should check the issue with the phone app.
When I tried to upload a photo manually, that worked. I tried to switch off and on the auto upload but it did not help. Tried many things, nothing helped.
I wouldn’t expect so many upstream prematurely closed connection while reading while reading response header from upstream (particularly for /index.php/204 transactions which are the lightest weight transactions possible essentially)… unless something weird going on server-side. Unfortunately that can have multiple causes ranging from resource usage to protocol mismatches.
It’s a bit challenging to determine how relevant your various log entries are and how they correlate with the original transactions though so not sure what’s even relevant.
Are you using the exact configuration for NPM+AIO described here?
Are you able to correlate (by timestamp) a specific single upload attempt (a failing one) end with a single NPM access log entry?
Also, if you’re comfortable doing so, I’d generally kill off virus scanning first thing while troubleshooting anything like this to assist with process of elimination since it’s both a major intermediary in the transaction path and a resource hog.
Those of you that chimed in on this thread with similar problems, are you specifically referring to the “Unknown error” in the Uploads tab? Or just general auto uploading problems?
Yes I’m using that AIO nginx proxy manager config.
I’m not sure how to disable the virus scanning. If possible I’d like to be able to disable it always except for file drops (since this is basically a single user install of nextcloud so I know everything I’m uploading)
