Issue regarding the famous api bug following us from last year;

Hey guys.
Just a question, if anyone had this issue as well./
Namely, Ive been having that
Shows to be no order at all on this error.
Sometimes it goes on (eventho credentials are valid), sometimes it does not.
Now I know that from new authentication method, things went wrong. But what bugs me that , no matter the version of RoS I have rolled on my routers, some same versions show that api error when trying to login. For no reason (nothing has been tampered with).

So basically, I have multiple Tik`s (same build), same working with api calls (without new procedures), some not-working.

Anyone else has same issue?

It’s difficult to tell from your text precisely what your problem is. But from what I can gather, you have some routers showing API errors and some not showing them, yet all are on same RouterOS level.

My first guess would be that some of your units were compromised by the rash of exploits that occurred last year, and some were not.

Try exporting the configuration of one of the problem units, netinstalling it from scratch to destroy any leftover hidden files, then reimporting the configuration, and see if the API problem corrects itself.