[jodah]

Like anything, it depends on what you're using it for. I wouldn't put a distributed lock into some massively high volume request path, or where absolute availability is required, but it's perfectly fine for some scenarios.

As Martin points out though, many (most?) distributed lock implementations are or can be broken in various ways. He hints at one of the fundamental problems - consensus. Many distributed lock implementations fail simply because they cannot achieve reliable consensus, or don't even try to.

That said, distributed locks can be safe and handle reasonably high throughput. The Atomix distributed lock is one example:

http://atomix.io/atomix/

http://atomix.io/atomix/docs/coordination/#distributedlock

Since consensus requires quorum and quorum requires availability, there is a risk that your ability to obtain a lock or learn about lock related events could be effected by availability, but at least in the case of Atomix, the system is fairly resilient with auto-failovers to passive or inactive nodes as needed (as compared to, say, a ZooKeeper based lock).

[mavelikara]

Thanks for posting about Atomix; I had not heard about it before. This seems to be based on Copycat, discussed previously on HN here: https://news.ycombinator.com/item?id=8180360

[jodah]

It is indeed built on Copycat. We're updating the docs right now ahead of an RC, this week or next.