after some software problems, the docker image of my Seafile server was somehow messed up (seahub start script kept attempting to restart the process although it’s running already), so I decided to just move away from Docker and do a regular local install.
Anyway, it seems that no matter what version (tried current 6.3 first, then 6.2.5 which is what was in the docker image, then 6.2.3, which I had been running natively before switching to the docker image) I used for a fresh install (before even thinking about moving my data to it), I just can’t get it to work. From what I can tell, I followed the manual to the letter. Still, I get loads of errors from seahub, partly from seafile too … e.g.:
==> seafile.log <==
[11/29/2018 07:32:13 PM] …/common/seaf-db.c(425): Error prepare statement SELECT repo_id, commit_id FROM Branch WHERE name=‘master’ AND repo_id IN (‘9ac31c6f-877a-4ef9-8306-907ee0a90899’) LOCK IN SHARE MODE: sqlite3_prepare_v2 failed: near “LOCK”: syntax error.
This error shows up before even logging in, and it repeats every 10 seconds …
Logging in fails, and after a bunch of Django error trace lines, this line shows up:
AttributeError: ‘CcnetThreadedRpcClient’ object has no attribute ‘get_primary_id’
The same error occurred with all the versions I tried.
After trying to get Seafile work again for two days, I’m pretty much frustrated … I’m pretty experienced with Linux and FOSS, but a clean setup with setup script leaving such a mess is pretty annoying … I’m completely out of ideas … and I can’t even switch to any other software, as I can’t get the f’ing files out of the binary mess seafile has on disk unless I get it to work again!
System is Ubuntu 16.04 with all current patches.