undefined | Better HN

0 pointsnijave1y ago0 comments

Not only that, some things like monitoring you get for free on cloud providers and setting up your own HA stack is going to be significantly more expensive.

There's almost no testing and validation needed for something like AWS RDS Postgres backups. Occasionally you store an instance and that's it.

Other things like Postgres out-of-disk-space is a 10 min fix on AWS to increment the assigned space. If your VM provider is offering SAN/NAS you may be in luck, otherwise hopefully you have a balast file or some logs you can delete to free enough space to get things running long enough to fix the problem.

0 comments

1 comments · 1 top-level

sgarland1y ago

Setting up HA is approximately the same cost as with anything else: N x $cost. Web servers? Run two with HAProxy fronting them. Want HA HAproxy? Run two with a virtual IP. Etc. Not hard to calculate, not hard to set up. These are all extremely solved problems with tons of documentation available.

I mentioned monitoring being required and having a wide variety of options and price points.

If your DB runs out of space in a VM that’s on you for not having adequate monitoring and alerting. If you don’t have networked storage available and somehow fail to see the predicted eventual out of disk, again, that’s on you.

j / k navigate · click thread line to collapse