tutfbhuf

You exchange network access against filesystem access, but what if your filesystem is a network device?

Many big things started out in a garage, with very simple solutions like your little UPS powered NUC.

Yup. Setting up k8s with kubeadm on bare-metal is very straight forward and can be done within a few minutes on any linux host that is supported by kubeadm and docker + ssh access.

Yes, but Google Spanner is not open source, so it's not so easy to actually look at it. But what is known to public is that Spanner relies on perfectly synchronized atomic clocks within their datacenters, called TrueTime. This solution is very special and does not work for other databases in general.

Eventual consistency happens when you allow your write transaction to finish, before the data has been spread across all masters. It allows you to have fast writes, but it's actually cheating (in my opinion), because only one master (the nearest) has seen your write, while the others don't know about it yet.

If you want to have a strong consistent multi-region database (or other form of storage), then you have to deal with slow writes, because your transaction is only allowed to terminate successfully when all masters have written the new data and are in a synchronous state.

I very much prefer slow strong consistent writes instead of fast eventual consistent writes.

I think at a certain point up in the stack you don't want to have another orchestrator for the orchestrator. You would just tell each team to keep their cluster up-to-date without creating a (dangerous) master button to update them all at once.

While it is certainly possible to have mixed region clusters, from my experience it is better to have one cluster per region. Do your synchronous communication within the cluster / within the datacenter with sub millisecond latencies and do asynchronous communication across cluster boundaries with up to triple digit millisecond latencies.

The current trend goes to multi-cluster environments, because it's way too easy to destroy a single k8s cluster due to bugs, updates or human mistake. Just like it's not an very unlikely event to kill a single host in the network e.g. due to updates/maintenance.

For instance, we had several outages when upgrading the kubernetes version in our clusters. If you have many small cluster it's much easier and more save to apply cluster wide updates, one cluster at a time.

Does Flutter (Dart) generate native binaries like Go?

Use hCaptcha for sign up and other sensitive parts of your application.

Okay, but what I really wanted to say is that I want a very small and low powered NAS without a big PSU. The PI is powered over USB, that's fine for me.

HN, please read: https://github.com/jasperla/openbsd-wip/issues/86

before you consider to use Pale Moon.

It's post like these that bring up emotional flame wars.

If someone says that Wayland sucks, and tells why he thinks so, then this is a very legitimate opinion that has nothing to do with 9/11.

RAID-Z2 on 4x2TB 2.5 inch HDD (without SMR) on a PI over 1 GbE sounds fine to me.

It's not super much space, but any two drives can fail, it's rather cheap and low powered + passive cooling should be possible.

50W is way more than I am willing to operate. I want it to be extremely low powered. I think that a 2.5 inch SSD might be a better fit for me then, maybe M.2 if there's a adapter for the PI.

SSD and low cost is still possible, but one has to make compromises in terms of storage size. A 1 TB network drive would be enough for my use case, be it 4 x 256 SSD's or one big.

> speeds will be lower and reliability may suffer

1Gbps is the max. anyways for my home setup, no matter how fast the drive is and I have a backup for important stuff, I'm not dependent on enterprise NAS drives.

I want: extremely small, low cost and very low watt usage.

> Regarding the PSU, but you'd need a beefy adapter anyway for running 4 drives

How about 2.5 inch drives? I'm thinking about: https://cdn-reichelt.de/bilder/web/xxl_ws/A300/RPI_NAS_4XSAT...

SoftBank is the $100B vision fund. $100M is just 1/1000 of that. This is like if you'd own $100k and invest 100$ in something, that's not even worth a news.

No, they shouldn't. Centralized App Stores isn't the way to go, look at recent examples of (temporarily) removed Apps like Matrix from Play Store. A decentralized App Store (e.g. based on IPFS) would be great.

This is a very synthetical fio benchmark, I would like to see how actual applications like a postgres databases would perform on such a tuned machine.

> Does this come up in practice? What kind of use cases suffer from disk latency?

One popular example is HFT.

And from my experience on a desktop PC it is better to disable swap and have the OOM killer do his work, instead of swapping to disk, which makes my system noticeable laggy, even with a fast NVMe.

This article focuses on IOPS and throughput, but what is also important for many applications is I/O latency, which can be measured with ioping (apt-get install ioping). Unfortunately, even 10x PCIe 4.0 NVMe do not provide any better latency than a single NVMe drive. If you are constrained by disk latency then 11M IOPS won't gain you much.

Would be an interesting case study to find out if remote developers from Mars are actually more productive due to lack of synchronous communication.

But then it's rather a problem with the interview itself.

> stack overflow didnt package so easily

You just need some copy of something like: the way back machine, google cache or archive.is and then do asynchronous mirroring and you are not so far behind.

Zoom meetings might be a problem though.

Just go back to Xorg with peace in mind like me. Wait until you hear from others that Wayland is finally stable then switch, or if that is never going to happen, stay on Xorg until the next community replacement attempt.

Either pre-IPO trading or wait until the dust settles and then decide whether to buy or not.

I think lenovo could easily solve this by selling products like the Nano AND selling modern much thicker, heavier fully repairable devices like the T430/T530. Let the customer decide what they prefer.

Completely unrelated, but I somehow thought https://lobste.rs had invented a new shell (script) pattern at first.

One time email domains and email forward services are usually blocked, there are very long block lists for such domains.

From my personal experience it is best to have a secondary email account on a provider that is usually not blocked (like gmail), to keep your primary email account clean.