Live data from Hacker News

Parler’s epic fail: A crash course on running your own servers

blog.alexgleason.me

91–100 of 141 posts

Re: Parler’s epic fail: A crash course on running your own servers

#91
post #61

Earlier quoted context omitted.

I wonder how much of this is just the industry maturing/specialising. We don't think it's weird that most people don't mill their own flour when they bake bread, so long as the quality of the flour is good enough we're happy for someone else to do it for us. In the same way most people/companies don't really need to care what hardware their application runs on, only that it meets some bar of quality/cost that's appro…

If you need to have your baked goods constaltly, you will be better prepared with many flour providers in case of that one provider does NOT LIKE YOUR RECIPE and stop selling the ingredients to you... Or get your own flour mill. So, not a ideal analogy...

Pissing off your supplier is something everyone has to be careful of. If you're drawing little swastikas on your cakes while telling everyone how great "Phil's flour" is then don't be surprised when Phil doesn't want to associate himself with you any longer.

The sibling comment pointed out that vendor lock in can be a problem which I agree with, but I think for most of the industry that's a problem of protecting yourself from predatory price hikes/services being deprecated rather than the problem of actively pissing off people you need.

Re: Parler’s epic fail: A crash course on running your own servers

#92
post #26

Earlier quoted context omitted.

That's when I stopped learning how to admin systems so I am very down with this...

Well, it works. So much money is wasted with AWS. (Look at at their margins.) It's useful in development/testing. It's irresponsible to use managed services that don't have an easy migration path to something open.

Them having margins in no way means it's a waste of money. That's not how things work. It would cost just companies so much more money for the devops team needed to achieve parity with the services they provide.

Do you look at a car manufacturer, see that they're making a profit, and assume "I could clearly build a comparable car for less money?"

Re: Parler’s epic fail: A crash course on running your own servers

#93

> freedom minded > tech oligarchs ...here we go. I really struggle to understand how some folks didn’t realize they were using other people’s things until recently. Maybe the fact that they weren’t physically in a space was hiding the proverbial threshold crossing activity that takes place a dozen or so times as you travel to twitter.com. It is free speech to kick someone off your platform. Freedom minded individuals…

The argument isn't against individuals, or even corporations, having the freedom to deny service. The argument is about monopolies, or in the case of Parler: colluding monopolies, being able to do so.

There are two things everyone knows are market failures that need to be regulated: monopolies and third party negative externalities. Right and left wing economists alike. Trust bust and regulate monopolies, and return net neutrality.

Re: Parler’s epic fail: A crash course on running your own servers

#94
post #23

> Most datacenters only lease by the rack. These racks can hold up to 42 servers and are far too expensive. Hurricane Electric will happily lease you an entire rack for $400/mo with a 1Gbps connection. I would argue that, in comparison to the prices the OP mentions, Hurricane Electric's price is quite good. Also, there is one fewer middleman between you and the Internet (and power, and rack space, etc, and the front…

They have some power included in that $400

Re: Parler’s epic fail: A crash course on running your own servers

#95

Earlier quoted context omitted.

I worked at a startup that managed its own bare metal. It was very hard and time consuming. And when things went down it was on you. You needed a really good datacenter partner to make sure that they were on top of things and could drop things at a moment's notice. But you don't know how responsive they will be until you're actually experiencing an outage. The level of convenience that cloud providers give is just or…

I have worked for companies in the past with horrible datacenter partners (one of them did not know that one of the two switches they routed our traffic through was completely dead), so I definitely agree with you on that. On the flip side, I wonder how much better the support from a cloud provider is if it's an isolated problem and not something that's setting twitter aflame. If it takes a cloud provider in the orde…

> On the flip side, I wonder how much better the support from a cloud provider is if it's an isolated problem and not something that's setting twitter aflame.

I’ve had 4w to resolution for major service degradation on a big name provider and ~2d for full site outage (highest support tier)

You said it yourself - aws, goog, msft have very good sres but they’re not your sres. Meaning they dont care if you have a big event/demo/deal close coming up before they start doing network gear upgrade and such...

Re: Parler’s epic fail: A crash course on running your own servers

#96

Earlier quoted context omitted.

I worked at a startup that managed its own bare metal. It was very hard and time consuming. And when things went down it was on you. You needed a really good datacenter partner to make sure that they were on top of things and could drop things at a moment's notice. But you don't know how responsive they will be until you're actually experiencing an outage. The level of convenience that cloud providers give is just or…

I have worked for companies in the past with horrible datacenter partners (one of them did not know that one of the two switches they routed our traffic through was completely dead), so I definitely agree with you on that. On the flip side, I wonder how much better the support from a cloud provider is if it's an isolated problem and not something that's setting twitter aflame. If it takes a cloud provider in the orde…

An enterprise client of a cloud vendor will have dedicated account technicians - response time is extremely fast (minutes or hours). Perhaps things have improved compared to some of the other comments here.

The other consideration is that things generally don't just 'go down' like they do with bare metal (because HA - replication and so on), but if they do, it's likely affecting a large portion of the internet too.

Re: Parler’s epic fail: A crash course on running your own servers

#97
post #4

I grew up in an era where running your own servers was just generally accepted practice. It's been fascinating watching how dramatically that viewpoint has shifted over the years to the point where it is now a novel idea to do so.

It surprised me how well cloud providers have managed to sell "availability" and lack of responsibly for site downtime, and how much of a premium they can charge for it. Most people don't need 5 9's (not that any cloud actually provides that level of reliability) or infinite scalability, just a box that's good enough in both dimensions.

Re: Parler’s epic fail: A crash course on running your own servers

#98

Earlier quoted context omitted.

I worked at a startup that managed its own bare metal. It was very hard and time consuming. And when things went down it was on you. You needed a really good datacenter partner to make sure that they were on top of things and could drop things at a moment's notice. But you don't know how responsive they will be until you're actually experiencing an outage. The level of convenience that cloud providers give is just or…

I have worked for companies in the past with horrible datacenter partners (one of them did not know that one of the two switches they routed our traffic through was completely dead), so I definitely agree with you on that. On the flip side, I wonder how much better the support from a cloud provider is if it's an isolated problem and not something that's setting twitter aflame. If it takes a cloud provider in the orde…

You make me feel so underpaid, is 300k the going rate for being a DevOps Engineer that knows the old ways of CGI/LAMP and can leverage running Exchange/IIS nowadays?

I don't even know if these salaries are truthful now.

Re: Parler’s epic fail: A crash course on running your own servers

#99

The "reason" given why Parler failed is awfully glib and one dimensional. You can't just serve content to millions of people per day and think you can just up and move to another provider. Had Parler just moderated their posts like they were asked to do, they would still be around. You can still have right wing, even extreme right wing views without calling for violence or organizing insurrections. The fact they esse…

>You can still have right wing, even extreme right wing views without calling for violence or organizing insurrections

The insurrectionists did not actually use Parler. They used Twitter and Facebook.

Re: Parler’s epic fail: A crash course on running your own servers

#100
post #42

Does there exist a de-platform proof way of hosting something on the internet? Even if you host your own server on your own premises they could forbid you from using the location, shut off your electricity, etc. Is it possible to host a website entirely on bittorrent? I suppose there's also IPFS but I'm not entirely familiar. Maybe once solar is cheap enough you could launch some sort of array of powered drones into…

Make a dark web[1] site? Basically make it impossible to link what you're doing to who your infra providers/internet connections are. [1] https://en.wikipedia.org/wiki/Dark_web

I bet there won't be many data centres happy if you're doing tons of traffic to the tor network. Many even disallow simple things like IRC.
Post reply on HN