Hacker News Reader

Comment by Sodman | Hacker News Reader

I'll still take a cloud service with rare outages over an on-premise solution that I need to manage, host, and maintain myself any day.

View on HN

I had never once used Slack until I started my current job in August 2017. I am now in a total of 3 Slack.. "organizations"? "Servers"?

Anyway, this is at least the 7th or 8th time Slack has been totally or partially down during working hours.

This is very, very far from "rare outages".

Fair enough. What about having both? Slack could by default communicate via their backend, but if the client finds peers on the local network, prefer those. When the connection to the backend dies, only use the local connections.

We can have the best of both worlds.

While the redundancy would be nice, I can't see it being worth the effort/money for anything but the most critical of systems (which should probably be self-hosted in their entirety to begin with). If the internet (or some major swath thereof) is down, that's going to cripple a lot of businesses these days regardless.

I just don't see it being worth the extra effort of self-hosting a redundant copy of the system, and then keeping that system up to date with the latest software for security patches / features / bugfixes. Especially when the cloud version has n x 9's of uptime.

Just curious - does Slack offer SLAs? To many businesses, especially remote businesses, Slack is becoming more and more integral to day-to-day operations.

Yes, they do on their Plus plan

>If we fall short of our 99.99% uptime guarantee, we’ll refund customers on the Plus plan 100 times the amount your workspace paid during the period Slack was down.

Source: https://get.slack.help/hc/en-us/articles/204113126-Plus-plan...

Except if you check, Slack claims 100% up time over the past 30 days.

I don't get why people think an sla is any benefit whatsoever hosting or software services are generally so cheap, any refunds are completely superfluous compared to the business and productivity loss.

Because SLAs pay more than the subscription cost during the downtime.

I have never seen an SLA pay out anything more than a small multiple of the hosting cost which still of course doesn't amount to a hill of beans for essentially anyone. Have you?

Wait, why should critical systems be self-hosted? Isn't this one of the problems "the cloud" solves for?

Seems like a great idea really.

Maybe an emergency mode in the Slack application. When users cannot reach the Slack servers (when they're down or bad network connectivity) the application exposes a P2P connection and looks for other P2P instances.

It would still need a centralized tracker (like a torrent tracker) to help P2P connections find each other through the internet.

All this is easier said, then done!

You're just pushing the target aside a bit, comparing an external mothership to running an internal mothership.

Why have a mothership at all? Why couldn't software clients simply talk to each other? We already have hash-based addressing systems that can find known peers anywhere in the world (e.g. Tor).

Because in many corporate networks, it's easier to talk to a server half-way across the world than to talk to a laptop computer on the desk next to yours.

Exactly. And by rare, we're talking 99.99% uptime. Slack is down for me too, and I'm not exactly excited. But this is a mild inconvenience compared to what I'd have to deal with should I have my own self-hosted alternative.

Back in the early 90s I recall using a program that allowed two or more people to chat at the same time, without a server. It was a true peer-to-peer communication program. The current paradigm of Internet usage has blinded us to what is possible.

Didn't AIM have that as an option? You had to "direct connect" to send pictures. Of course, you had to connect to their servers first and the parties had to agree to direct connect.

I'm also pretty sure all of us wrote this in CS 101. I sure did.

Honestly I've heard this before and maybe I'm overly bitter, but this just sounds like an IT department that doesn't want to do their jobs. You're there to manage those internal systems. If they're going off the rails that often, it sounds more to me like you're just shit at your job.

We manage over 900 virtual systems on many, many hosts and file systems and have next to no downtime because we work really hard to make that the case.

Reply on news.ycombinator.com