Hacker News new | past | comments | ask | show | jobs | submit | cgeier's comments login

IMO retirement isn't a long holiday but it's for people who cannot contribute to society anymore.

That point might be reached earlier for some people, e.g. those who have a hard physical job, than others and a general retirement age is an approximation for when that point is reached for the general population.


Retirement doesn't mean I won't contribute to society anymore. It'll mean that I can contribute in ways that I am most interested in, without having to make money doing it.

This is huge news for ZFS users (probably mostly those in the hobbyist/home use space, but still). raidz expansion has been one of the most requested features for years.


I'm not yet familiar with zfs and couldn't find it in the release note: Does expansion only works with disk of the same size? Or is adding are bigger/smaller disks possible or do all disk need to have the same size?


You can use different sized disks, but RAID-Z will truncate the space it uses to the lowest common denominator. If you increase the lowest common denominator, RAID-Z should auto-expand to use the additional space. All parity RAID technologies truncate members to the lowest common denominator, rather than just ZFS.


Is it definitely the LCD? Given drive of size 15 and 20 the LCD would be 1, no? I had assumed it would just use the size of the smallest drive on every drive (so 15+20->15+15=30). When I first read your comment I was thinking of GCF but even that would be fairly inefficient (GCF(15,20) = 5, so 15+20->5+5=10).


That's not entirely true, Unraid has mechanisms for unbalanced disks, but they come at a high cost in terms of usability by standard workloads.


Unraid is not a RAID technology:

> Unraid saves data to individual drives rather than spreading single files out over multiple drives

https://en.wikipedia.org/wiki/Unraid#Software-defined_NAS

At least, it is not one in the sense of the original RAID paper that coined the term:

https://web.mit.edu/6.033/2015/wwwdocs/papers/Patterson88.pd...


As far as I understand, ZFS doesn't work at all with disks of differing sizes (in the same array). So if you try it, it just finds the size of the smallest disk, and uses that for all disks. So if you put an 8TB drive in an array with a bunch of 10TB drives, they'll all be treated as 8TB drives, and the extra 2TB will be ignored on those disks.

However, if you replace the smallest disk with a new, larger drive, and resilver, then it'll now use the new smallest disk as the baseline, and use that extra space on the other drives.

(Someone please correct me if I'm wrong.)


> As far as I understand, ZFS doesn't work at all with disks of differing sizes (in the same array).

This might be misleading, however, it may only be my understanding of word "array".

You can use 2x10TB mirrors as vdev0, and 6x12TB in RAIDZ2 as vdev1 in the same pool/array. You can also stack as many unevenly sized disks as you want in a pool. The actual problem is when you want a different drive topology within a pool or vdev, or you want to mismatch, say, 3 oddly sized drives to create some synthetic redundancy level (2x4TB and 1x8TB to achieve two copies on two disks) like btrfs does/tries to do.


This is the case with any parity based raid, they just hide it or lie to you in various ways. If you have two 6TB dives and two 12TB drives in a single raid-6 array, it is physically impossible to have two drive parity once you exceed 12TB of written capacity. BTRFS and bcachefs can’t magically create more space where none exists on your 6TB drives. They resort to dropping to mirror protection for the excess capacity which you could also do manually with ZFS by giving it partitions instead of the whole drive.


IIRC, you could always replace drives in a raidset with larger devices. When the last drive is replaced, then the new space is recognized.

This new operation seems somewhat more sophisticated.


You need to buy the same exact drive with the same capacity and speed. Your raidz vdev be as small and as slow as your smallest and slowest drive.

btrfs and the new bcachefs can do RAID with mixed drives, but I can’t trust either of them with my data yet.


It doesn't have to be the same exact drive. Mixing drives from different manufacturers (with the same capacity) is often used to prevent correlated failure. ZFS is not using the whole disk, so different disks can be mixed, because the disk often have varying capacity.


You can run raid-z across partitions to utilize the full drive just like synology does with their “hybrid raid” - you just shouldn’t.


> You need to buy the same exact drive

AFAIK you can add larger and faster drives, you will just not get any benefits from it.


You can get read speed benefits with faster drives, but your writes will be limited by your slowest.


Just have backups. I used btrfs and zfs for different purposes. Never had any lost data or downtime with btrfs since 2016. I only use raid 0 and raid 1 and compression. Btrfs does not havr a hungry ram requirement.


Neither does zfs, that’s a widely repeated red herring from people trying to do dedup in the very early days, and people who misunderstood how it used ram to do caching.


Tbh the idea of keeping backups defeats the purpose of using RAIDZ (especially RAIDZ3). I don’t want to buy an LTO drive, so if I backup, it’s either buying more HDDs or S3 Glacier ($$$). I like RAIDZ so I don’t have to buy so many drives. I guess it protects you if your house burns down, but how many people do offsite backups for their personal files? And dormant, unpowered HDDs die a lot faster than live, powered HDDs.


Yes, seriously handling your data is expensive. I am talking about buying new hardrives.


German Eurofighters do not carry nuclear bombs. That's Tornados (and soon F35s).


You are mixing up nuclear bombs with american nuclear bombs. But regardless, i think eurofighters are getting U.S. B61 certification soon.


What other nuclear weapons would German Eurofighters carry? Also with Germany buying F-35s, not sure any certification for German Eurofighters is going ahead (the public discussions there seem to predates the F-35 purchase decision, I think).


Which works really well! Thanks for a great project!

Is there any more information available on how it works?


According to Wikipedia it's about 180 widebodies for Lufthansa group

https://de.m.wikipedia.org/wiki/Lufthansa_Group


How do I then get the passkey for my second device accepted by the service? Do I mail the public part to myself and insert it from my first device?


The first time I log in to a service on a new device it'll prompt me to sign a challenge with a previous passkey. If I've got my yubikey handy I'll just plug it in and sign it and add a new passkey to my new device. If I only have my phone the site will flash up a QR code I scan with my phone which signs and posts back the proof to a callback URL for the site. I only need to do this once per device if I add a passkey to the device.


Is the fact that you need access to an already- enrolled device to create additional passkeys part of the threat model that passkeys resolves, or just an annoying detail? And is this for every site, or just once per device? I can just look it up, this thread has been great to improve my mental model enough to start considering trusting it.


Its per-site. So the first time I log into GitHub on a new device, I need to do the handshake with another device. The first time I sign into Coinbase, I need to do the handshake with the other device.

So this typically means when I get a new device I'll have my Yubikey in a bag or something with me for a while and pull it out from time to time. Eventually practically every site I use gets enrolled on the new device and I never actually need to reach for the Yubikey or my phone or whatever.

I don't really make any concerted effort to go through each and every account when I get a new device, it'll pretty much just happen eventually. When I do sign up for a new account that supports passkeys I do try and make an effort make a passkey on at least two devices though, often at least whatever device I'm using to initially register and my yubikey. Then I'll make a point to log in sometime in the next few weeks on another computer and create a passkey there. Eventually I'll probably end up logging in and making passkeys on most of my devices.

Needing to auth with an existing passkey is a major part of the model. If you could just log in and create a new passkey with just a regular password, what's the point?


According to the article, she is called "Truong My Lan", perhaps the author's muscle memory or autocorrect was kicking in.


No, it's just the silly HN parser that automatically changes the case of some words in titles based on an obscure internal rulebook dictionary and also removes other words that sound click-baity.

I'm still baffled HN hasn't opensourced their stack.


It's Lisp. I suspect there is no copy of the source code, they've just kept adding to the running program.

Just kidding but that does sound like a fun art project to see how complex you could get...


Hacker News auto-"fixes" some keywords in story titles, it could be that.


There is "she asked her 18,000 followers on Facebook to share their opinions about a tomato puree". Is jail time appropriate for that? Probably not, but she did not just review puree, she asked her followers to do it to.


Apparently that wasn’t the trigger.

She seemingly then followed it up with something like “It’s k**ing people” (sic) - and then she was arrested.

Still seems like a mass amount of overreach to call it a CyberCrime though.


I personally like the Economist.

They do have a some bias, but they make that pretty clear most of the time.

The articles don't contain too much fluff or "storytelling", but mostly get very straight to the point. Also, I really like that it's a weekly magazine, no "Breaking News!!!!" type of content.


You don't need ML to do this, there are already auto landing (and auto take off) systems that work very well. The hardest part is taxiing on the airport and taking part in air control procedures doesn't work at all AFAIK.


And for taxiing hard part is likely not moving the plane, but not getting in a way of anyone else. You really don't want to be at wrong part of runway at wrong time.


Ground movements and runway clearances are still very manual processes unfortunately, and mistakes happen especially in high task-load situations or poor visibility or flight number collisions etc. The safe part about being in the airplane is actually being in the air, lol - once you push back you are at fairly elevated statistical risk until you’ve got some altitude. Takeoff/landing are high risk and there’s a lot of people and a lot of chances for mistakes on the tarmac.

There are statistically an enormous number of ground movements carried out safely but boy does the potential body count get big with runway incursion scenarios. Two 757s colliding is a lot of people, and in high-energy scenarios at least one of the aircraft isn’t walking away. You can be doing nothing wrong and someone just landings on your taxiway etc. The close calls are terrifying.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: