But do those specialized hardwares also work with stuff like opnsense?

I hope you are also donating to the projects for uploading multiple copies to different services.

Not an expert.
Assuming it’s internal I’d assume it’s probably as fast as the guest nic allows?

They are already interconnect at various points.
Schools are connected to university networks, university networks to loval government intranets and those are again probably at some point connected to the federal network.
I don’t wanna guess where else they have connections to like the police or legislative network.

AWS Snowflake, ball?
What’s next? Glacier? /s

Maybe NFS share the drive from guest to host?

Probably writing to the whole disk with something like dd and then testing if they can read it.

I heard some advices with “burning them in” by running a stress test over the span of some days with random IO. If they survive, great. If they die, they will be returned.

An agile pot?
Will it run from me while I try to put ingredients in it?

my pleasure if I can help :)

Many (most?) residential internet service providers do not allow self hosting websites on their network and they’d be dynamic IP anyway though you could work around that somewhat with dynamic DNS since you’re going to need to purchase a domain name and point it to somewhere anyway.

That’s what I already do.
cloudflare-ddns, own domain and done. Except for reliability at night where the ISP reboots my DSL connection at 4am.

You can edit comments

What we usually do is using an SMTP relay.
Plus points on having a spam protection service included and in front of the mail server.

As an employee of an MSP it’s sometimes really annoying how some smaller one-man-army + employee MSPs do “crafty” setups on site and meaningless firewall rules to work against.

and email can be hosted sooo cheap these times. My ionos contract just to have my own domain costs me maybe 10€ per month.

• 2 servers for reundancy, preferably 3rd one laying around for a quick swap

Overkill for a small startup. We sell to medical practices and they get (at most) a small tower server from HPE (ML30). Backup servers can either be a small 2 bay NAS or a microserver depending on the needs

Pretty decent UPS setup, again multiple units for reundancy

If they can afford that. Most are already good by just using a decent online UPS.

Routers, network hardware, internet uplinks and everything at least duplicated and configured correctly to keep things running

Not a multi million company.
Routers: BS.
Network: Maybe 2x 16 port switches but too much overhead. 1x 48 Port if you really need that.
Internet uplinks: If you have a satellite office and need the S2S-VPN to be constantly up.
Btw: You didnt mention a duplicated phone line ;P

• A separate backup solution, on at least two different physical locations, so a few more servers and their network, power and other stuff taken care of

Again overkill.
Backup on the server with something to a 2-Bay NAS and a secondary job with rotating external USB disks to take or somewhere external with you.

Monitoring, alerting system in case of failures, someone being on-call for 24/7

Monitoring: Yep
Alerting: Done by monitoring in the best circumstance
On-call 24/7: As this is a small company: LOL

Not fear but you need to have the respect it deserves.

Had a similar issue migrating to my new setup after I started scanning the new archive.
Try increasing your swap disk.
Mine was set at 1GB but increased it later to 8GB. No more freezes of the host.

Dev discontinued the app due to google being difficult to maintain.

I don’t wanna ask if and how much individuals contributed to the ones that host their instances ;)