devicenull

You also can't ctrl-f within a page, which makes finding things really annoying.

Yes, I have to do this maybe a couple times a year (tracking down weird KVM bugs, or figuring out poorly documented features)

Who said anything about average linux users?

We're talking about a product designed to be run on multiple servers, in a datacenter. At that level, you're generally talking about professionals.

Not quite... when I run into an issue with my Linux machine, I can trace the code all the way down to the kernel.

When I run into a problem with my Windows machine, I have to resort to random searches and hoping that someone else has seen it.

I'd guess all of them are VPS instances. When you order a VPS, most hosts will run a full OS update before delivering it to you. (this is a feature of cloud-init, which is fairly standard) So, if you ordered all Debian instances at the same time, they'd all have the same SSH version string even if you didn't do anything special.

Keep in mind you can generally only anycast a /24's worth of IP addresses, so it's very unlikely they're doing this with single IPs.

Why are MILLION and ONE capitalized?

> That's weird. Since it defeats the purpose of expanding their WiFi hotspots coverage. If people have capped plans, they are likely to disable that hotspot altogether and use their own routers to begin with, to avoid anyone switching it back on. If they wanted such thing to work, it had to be excluded from both monthly data caps and active bandwidth limit.

I think what he's trying to say is that as an individual using the xfinitywifi network the traffic counts towards your data cap. As the person hosting the access point, other people using it does not count towards your cap.

> UDP echo: Echoes any request sent to 54.175.103.105:40000

> UDP info: Get request info at 54.175.103.105:40001

That seems a little worrying... it liiks like it could be abused as a reflector in a DDOS attack ( http://en.wikipedia.org/wiki/Denial-of-service_attack#Reflec... )

> It's a double edged sword. The manufacturer will be happy if he sell a car. The dealer needs to first buy the car from the manufacturer and then sell it for a profit. Thus the manufacturer have a competetive advantage, that they could use to remove all competition, and then charge whatever the fuck they want.

I don't think that's it... having dealerships does nothing to stop the manufacturer from changing prices to whatever they want.

> I agree, it is like when Microsoft made KMS activation servers for companies that had volume licenses and no Internet connection to Microsoft to activate systems. Only now it can download Windows 10 updates and then have each machine on the LAN connect to it to download updates.

Microsoft has had WSUS for quite some time now, which lets you get all your office PCs downloading from one server. IIRC, you can even do offline updating of your WSUS server (download the updates elsewhere, transfer to WSUS via USB, install on all your clients via the network)

50 ms? NTP is far more accurate then that.

The last mile costs are significant even for running some wire to every house. Can you imagine what they'd be to run a multiple inch tube?

It takes basically nothing to run a blacklist. You're pretty much free to do whatever the hell you want with one.

You'd be hard-pressed to find an IP that isn't listed in at least one blacklist... for example, my latest favorite is the blacklist that lists anyone sending commercial IPs. For example, send your customers an invoice? They'll list you. Send them an order confirmation? Listed again.

The real test, is are you seeing email being rejected? If not, then you're probably just listed on some very low quality blacklists.

Libvirt has serious issues with this https://www.redhat.com/archives/libvirt-users/2014-November/...

I'm very confused by your reply.. are you trying to say you have three separate groups? ops, devops, and developers? With devops existing to communicate between ops and developers?

> It seems, especially for major corporations, that on-call/pager duty is quickly becoming the norm for software development teams. I do agree that pager duty is a symptom of a fundamental flaw within the system/architecture. I think it would be in a company's best interest to devote time in improving the reliability and stability of their infrastructure, instead of relying on the band-aid approach that pager duty seems to be.

I can't see there ever being a time where there is no on-call requirement. You always need someone standing by in case of some terrible disaster that cannot be handled automatically. Better to have this a formal responsibility that never gets used, then to not have it and end up with an extended downtime because you can't contact anyone.

That being said, if you're getting paged continuously during on-call, then there's a bigger problem that needs to be resolved.

Don't forget having to deal with billing nightmares even after you're no longer using an insurance company. You still could end up fighting with them over their failure to pay for something.

> Q: How does MassiveMesh work?

> A: We use fiber optic connections that form the backbone of our network. Our internet data centers use cutting edge wireless radio transmitters to transmit high speed signals to repeater sites throughout the city. These sites act as the go-between and deliver the internet directly to homes and businesses. Homes and businesses connect to those repeater sites through a small radio modem, which we provide, that creates your personal high speed network.

So, basically this is a standard WISP, not some special mesh network.

That's not really an excuse. This is a pretty trivial thing to fix (update the cert, or force a redirect).

Most modern routers do sflow collection in hardware. There's no CPU or memory overhead for having it on.

The main reason why it's not an official image is OpenBSD doesn't seem to support any sort of online disk resizing.

Operating systems get deployed from an image, and typically expand the own partitions to fill up the disk... OpenBSD can't do this, which makes things pretty tricky.

Not to be confused with the Android FastBoot.

How do you deal with validating a sender? That is, how do you ensure that spammer's don't just create tons of fake new senders?

This is also the default behavior for ASAs, so it's really just not bothering to configure them. Probably because email works, so why screw with it?

I have the new X1, and I find myself continuously hitting the function keys, because they're on some kind of touch screen... it's probably the only complaint I have with the laptop.

The important part is you can't read your private key out, nor update the firmware to something that allows you to read the private key out. The customization you can do is unrelated.

We have customers that try to run exits from time to time. We can usually tell pretty quickly, because the abuse complaints start coming in within a few days.

It seems that if you don't run the limited exit policy, you'll be hit with DMCAs pretty much instantly.

If you do run the limited exit policy, you'll get hit with abuse notifications (you hacked my site, you're posting spam, etc) pretty quickly.

So then you have two places to look to figure out where a scheduled task is running?