https://github.com/lldap/lldap logo
Cant log in error message
# troubleshooting
a
message has been deleted
s
I'm trying to get lldap working through Docker on a lightsail instance. And for the 4th time in a row, it seems everything is working but it just refuses to allow me to log in. I'm getting
Could not log in
error message when I first go to the web portal and login with admin and the password specified in the
LLDAP_LDAP_USER_PASS
variable. Using password or trying other passwords gives a invalid user-name or password error message. This has been going on for hours tonight and its about to break me. No matter what i do I can't get it to use a default id and password and if I try and set a password it still just refuses to let me log in.
b
Im in no way a pro, but what are your logs showing? For anyone trying to help you these can be useful
s
Yeah, sorry about that. Working on this as part of a project post surgery and I had to stop last night.
Copy code
> Setup permissions..
> Starting lldap..

Loading configuration from /data/lldap_config.toml
WARNING: A key_seed was given, we will ignore the key_file and generate one from the seed! Set key_file to an empty string in the config to silence this message.
2025-12-11T17:44:18.472357459+00:00  INFO     set_up_server [ 5.37ms | 100.00% ]
2025-12-11T17:44:18.472378946+00:00  INFO     ┝━ i [info]: Starting LLDAP version 0.6.2
2025-12-11T17:44:18.479903539+00:00  INFO     ┝━ i [info]: Starting the LDAP server on port 3890
2025-12-11T17:44:18.480169702+00:00  INFO     ┕━ i [info]: Starting the API/web server on port 17170
...
2025-12-11T17:46:18.701589480+00:00  INFO     LDAP request [ 131µs | 100.00% ] session_id: 20d34098-7bba-4869-bd90-649617889c84
2025-12-11T17:46:18.701996470+00:00  INFO     i [info]: LDAP session end: 20d34098-7bba-4869-bd90-649617889c84
2025-12-11T17:46:19.662420577+00:00  INFO     i [info]: OPAQUE login attempt for "admin"
2025-12-11T17:46:19.754822241+00:00  INFO     i [info]: OPAQUE login successful for "admin"
I don't think its a lldap issue, the logs show it is succeeding with the login, but the webpage still shows `could not log in The container is supposed to be on the same docker network as everything else, including caddy, and it looks like I was trying things out in the Caddyfile last night thinking that was the issue. Here is what I have in there for lldap.
Copy code
# ldap

ldap.acmevnt.net {
    encode zstd gzip

    reverse_proxy lldap:17170 {
        flush_interval -1
        header_up Host {host}
        header_up X-Real-IP {remote}
        header_up X-Forwarded-For {remote}
        header_up X-Forwarded-Proto {scheme}
    }
}
n
Can you post the verbose logs? LLDAP_VERBOSE=true
It's interesting that you get a different error from trying with the correct password than with a wrong one (did I understand that correctly?)
Although yeah, it says "login successful for admin", that's about all we need to know
Can you see if there's an error in your browser in the console? Open it with f12 and then try to log in.
You can try to log in via the command line with ldapsearch, but I expect that would work
You could open the LLDAP http port directly (instead of going through your proxy) to see if that works
That could indicate a proxy error
If it's a frontend problem, you could try incognito mode (or clearing the cache and cookies)
s
correct
just a generic error message from memory. strange but I just went to pull it and I was logged in on refresh
n
Can you try in incognito?
s
yeah, doing that now. I have a theory now.
well, that theroy is just gone, it just works now
n
I'm sorry to say I can't diagnose a non-existent problem 😄
s
As far as I know, nothing changed between last night and today when you responded.
all I did was take the container down and back up using compose to get the log files as startup and tried to login. Maybe it just needed time or something?
welp, thanks anyways
n
No, time doesn't help
You're welcome!
It could be that the frontend was in a bad state with an invalid JWT and then it tried to log in again because the 24h had passed?
Wild guess
s
or something had cached old caddyfile settings and it took until it updated to go through?
n
Could be
s
no telling, logs don't say much since it wasn't a lldap problem in the first place
n
Right. Well, glad you got it working!
s
thanks
I guess mark this as solved, lol