several instances of netbox on the same server

[quentindavid]

Hi, I have 2 instances of netbox on a server. It seems rack information is mixed up since last update (I updated from an old version to the latest).

Is it possible that information mix-up between the two instances due to the use of redis ?
Is there a way to prevent this ?

Or maybe it's only a cache issue ?

Example of what I get : when I list racks, I can see racks of the other instance of netbox, and when I click if the ID is existing it's showing the corresponding ID, but if it does not it throws an error "this id not exist".

[candlerb]

Is it possible that information mix-up between the two instances due to the use of redis ?

Yes, if they are both pointing to the same redis instance (port and database number).

Or maybe it's only a cache issue ?

Caching is what you see, but it will also mix up background jobs like webhooks which also are queued via redis.

You should point your second netbox instance to different database numbers within redis.

REDIS = {
    'tasks': {
        ...
        'DATABASE': 0,  # change to 2
    },
    'caching': {
        ...
        'DATABASE': 1,  # change to 3
    }

[quentindavid]

Ok I understand ! It seems to work much better now ! Thanks a lot :)
For the instance where I use the new database number it seems fine, but how can I reset the databases 1 and 2 ?

[candlerb]

Caches expire after 15 minutes anyway. You can use redis-cli and the flushdb command; or netbox manage.py has a command to invalidate caches.

Note that each instance uses two databases, and by default these are 0 and 1, not 1 and 2.

[quentindavid]

Yes, my bad, it's fine then it looks good now :)
and yes it's 0/1 and 2/3.

Thanks a lot for the explanations ! Everything seems fine now !