[Refusing23]

The US bitwarden servers were down - not the european ones accoriding to that little overview.

my workplace selfhosts it, so its not affecting us.

[mmooss]

> my workplace selfhosts it, so its not affecting us.

What is more reliable (ignoring hindsight), Bitwarden's service or your workplace's systems?

[roydivision]

Cached self hosted with Bitwarden as upstream master.

(Edited - added 'upstream')

[linsomniac]

I was just wondering if that was an option... Pointers on how you do that?

[bravetraveler]

Self-hosted things, in my experience, tend to be more reliable. More expensive, too. Worth it. Considerably less churn if done well. That said... reliability/administration is what I do for work. I'm biased.

For instance, skip traversing the internet for in-office deployments/credential use. Also: incentives. PMs or engineers at another company have incentives that don't always align with reliability for yours.

Vaultwarden deserves a mention, happy hoster for years here.

[mmooss]

I agree with those points but there's another side to it, of course: The economy of scale of service providers allows reliability, in important ways, that most people can't match by self-hosting.

[TheNewsIsHere]

That scale also acts as a liability.

When everything becomes hyper concentrated in singular environments, a problem with that environment scales the blast radius to everyone.

For example when AWS US-East-1 occasionally goes down. It’s been a few years since the last time that happened, but when it happens it’s inescapable.

[bravetraveler]

Absolutely. Things that are very data heavy or require extreme planning, for example.

I'll shell out cash for hosted databases... even though I'm fine with building everything else myself. Topology stuff alone usually makes this worth it.

Virtual machines and snapshots can take you a long way, though :)

[gpi]

We self host too for about 60 users and it has been very reliable