Not sure I understand why you’d want to self host a password manager. Bitwarden has never been breached AFAIK. How is it better or safer to keep if self hosted?
It’s good if you like self-hosting stuff.
However, what I tell people is this:
If you know jack about security and how to lock down a machine that is running Vaultwarden, then it’s useless. You should go with Bitwarden.
If you’re looking to install it just to play around with, I would be very cautious about what you store there, unless you can lock the system down to where it’s not accessible by the outside internet and localized only to your network.
And I have redundant backups in place in case one decides to fail, which are all encrypted with GPG and a few other measures.
If you have it installed and not accessible to anyone else but you, it’s a fun project. I like using VW and BW.
The other bonus would be no one is going to look to target you specifically unless you’re turned into a target.
Whereas if BW were to be breached, it wouldn’t have anything to do with you.
However, BW utilizes encryption, so even if they did somehow manage to get in, they can’t read your passwords.
Alright, what minimal security do you need to lock down your vaultwarden? Wireguard, firewall, fail2ban? I’m trying to learn good security practices for my server
I literally just had the exact opposite question! I’ve been wondering why you’d want to pay for a password manager service when you could self host it. The only reason I could think of is guaranteed high uptime, but to me (and at least in my personal use case) that seems a bit pointless, since you can have a copy of your password manager on each device, which is being synced through your server
A few reasons.
-
Privacy, you control your data. It doesn’t go to someone else’s server to sit.
-
Security. It’s on your server. Password managers are primarily targets for hackers, i don’t want to name names, cause I’m not 100% sure of the name. But, one pw manager was hacked like 3x in the past year or something. It’s on your server, you are less likely to be targeted for a huge data breach, and you get to manage your data. Not someone else who fucks up.
-
You can’t be banned, or have the provider suddenly change access to the server, thus losing your data. I will name names here. MyQ garage door opener by Chamberlain suddenly removed the smart home integration, since the whole system ran on their servers. Removing the functionality users paid for. But they don’t own it, so they just got fucked. Your data/service on someone else’s server, is actually their data/service, you are just a visitor.
-
I don’t self host anything where it would impact me unduly if it went down while I was on holiday to the point where I’d have to break state and fix stuff.
A password manager falls in that camp so it’s paid-for Bitwarden every night every day every possible way for me.
Sure Vaultwarden suits others - generally those who either want control of their data, smaller target on their back than a public instance user, watching their pennies etc.
Bitwarden has never been breached AFAIK.
What you mean is it hasn’t been breached *yet*.
All commercial password managers have a huge, fuck off, target on their backs
Nobody is going to come after some random blokes self-hosted password manager to get access to their Sonarr (I’m trivialising to make the point) as long as if a similar effort would get them into Bitwarden.
It’s the same principal as bears in the wood - nobody needs to outrun a bear, just your companion
OK, thanks for the solid answer. I suppose the core of my question was that pretty much: is it just as secure AND a less likely target than bitwarden. That makes a lot of sense to me. I would probably still worry about the strength of the code , though. Do we know if/how it’s been audited?
OK, thanks for the solid answer. I suppose the core of my question was that pretty much: is it just as secure AND a less likely target than bitwarden. That makes a lot of sense to me. I would probably still worry about the strength of the code , though. Do we know if/how it’s been audited?
I mean, your best having a look at the official Git but, i’d say, access/visibility is the most important.
Is it on your LAN/not open then even if it was less secure, it’d still be more secure if you know what I mean.
I host mine on a VPS but it’s behind traefik with authelia (and 2FA). Plan is to get fail2ban setup over the next couple of evenings. SSH is cert only, probably going to change the port too but not sure if that’s really necessary. I’m comfortable exposing on that basis.
change ssh port, put an ssh tarpit on the default