An umbrella community for all things cybersecurity / infosec. News, research, questions, are all welcome!
Enjoy!
I have a backdoor into my home vpn using a series of usernames, passwords, and long obfuscated http paths/subdomains.
In an absolute emergency, I can traverse that maze, retrieve a key+config to connect to openVPN, then reach my vaultwarden vault. No 2fa on that vault as it's not accessible from WAN. (though technically I could add 2fa and still be able to disable just the 2fa from vaultwardens admin console in a pinch)
Do you have all the paths, usernames and passwords committed to memory? My biggest fear is making it so secure that I don't remember it myself, since I'll effectively never use it until the emergency case occurs.
Yeah. They're all human readable but non-obvious instead of random strings. Stuff that's easy to remember but difficult to guess. You've just got to avoid typical patterns like 'randomwords526!!' or 'p00rex@mpl3'.
I do like to exercise that memory now and again, testing that I remember and that everything's functioning as it should. Just in case, theres instructions on paper in a safe place.
Being four separate item's minimum: subdomain, path, username, and password, none of which are published anywhere ofc; makes it pretty secure. The openVPN config/key needs a password as well, so 5 items.
Right, I've taken a similar approach now. Unknown subdomain at an unknown domain which is not accessible from the web, only via ftp. FTP username and password are known only to me, long and obscure but not forgettable. Then a random subfolder-tree down in an outdated cgi-bin script. In the folder I've got a password protected zip archive with dropbox recovery codes, and in the dropbox finally my google codes in yet another password protected archive. All passwords different and never been in any reported breach.
That's gotta do it for now. Thanks a lot for your input!
Keepass db syncd on cloud with multiple devices, 2FA using email instead of mobile number.
How do you get into your email?
I use an email client which saves my password and I don't need to enter it. But the keepass db can also store your email password for you.
That's the problem.
We're discussing accessing your accounts without prior access to a pre-authorized device.
If you don't have a device that's already signed into your email, you can't get into your passwords at all. Email is locked with a password stored in your password db, your password db is locked with your email. Without one or the other signed in already, you've locked yourself out of your own accounts.
Keepass db doesn't use email 2fa, its just a file you store on your device
I store it on all my devices so if I lose one I still have several others.
I use nextcloud to keep them syncd but you can use any cloud (google drive, icloud, one drive, Dropbox, etc)
That's still gaining access through a device that's already signed in/has your password db.
If you do not have access to a device that's already signed into your accounts/has a copy of your password db; how do get in?
Presumably you're smart enough to not have password only auth on a public facing nextcloud instance if it stores your password db...
This is the scenario we are discussing. The fact you store you db on other devices is entirely irrelevant.
My nextcloud instance uses fail2ban and I use a >32bit strong password.
Assuming I lose my phone and my laptop and my personal computer and my nextcloud instance I would be screwed.
Since I host my own mailserver I would be able to create a new mailserver with a new password though and recover any accounts with a new email.
Well that’s kinda the point. Missing your factors is supposed to make it harder.
So — account recovery code written out, that doesn’t say it is for your account?
E.g. Apple allows it as a backup to your Apple ID should you get locked out.
Depends on the syy
Yes exactly, but I still need to be able to access it myself.
Writing the codes out is one thing, but taking a paper with me on international trips actually increases the risk of it getting lost, damaged or stolen as well; and if someone was to figure out what they are for, that could even increase the potential risk. Hence my question what other people are doing :-)