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

Was it on btrfs by any chance?

I've burned my fingers once on it and recommend xfs ever since, no issues so far.


Interesting..! I've had something very similar happen on Fedora running btrfs, too.. No hardware failures (drive was great, still run it to this day a year later), no kernel panics, the partition just stopped booting out of the blue, with the rescue shell also being completely broken. I think I almost managed to get some of my files back, but after an error on my part it corrupted itself to the degree where the only answer you could find online was something like "just give up, no one can help you now :(". Lost a bunch of important data and still a bit mad about it, but at least now I'm back on NixOS with ext4 with no issues so far!

It was ext3 if I remember correctly. I usually stick to the default file system unless I have a really good reason to do otherwise.

btrfs has been the default on Fedora for a while. Ext4 was the default before that. So if it was ext3 days that's been a long time ago.

Can relate fedora borked my btrfs twice. I do not use either anymore. Debian+ext4 or Freebsd+zfs depending on purpose.

Why xfs over the more popular ext4?

Not who you are replying to, but I used XFS extensively in production workloads because at the time it had a few features I needed that ext4 didn't have. (I want to say one of them was support for 64-bit inodes? I don't really recall now.) And most importantly no waiting for the system to fsck on a large or slow filesystem that was unmounted uncleanly by a power or system failure. At the time, XFS was also a bit faster than ext4 but I'm not sure if that's still true. The differences between the two are fewer these days but it's still a great general-purpose workhorse filesystem.

reflinks, cheap/easy dedupe in a sense.

xfs is high performance and also rock-solid, I picked it for servers and like to be consistent in my experience, I never shrink volumes and to be frank I am a bit of a hipster/eternal contrarian.

I remember it in the second one at least:

https://youtu.be/r99hB4FzhXw?&t=153


> Java is the most backwards compatible language and it is not even a close competition.

In competitions consisting of Java, PHP and Python, I presume?


In competitions of real world code in a language that has been in use for close to 30 years, and I can find a java 1.1 program that will both compile as source on the latest version, AND the original compiled version itself will run on a modern JDK as is.

And I assume this mythical 1.1 program does of course do a lot more than System.out.println and the reason that it took until recently, in part thanks to the Log4J fiasco, for 8 (almost 11 years old) to no longer be the most widely used version, was just superstition?

Tiobes's index is quite literally worthless, especially with regards to its stated purpose, let alone as a general point of orientation.

I'd wish that purple would stop lending it any credibility.


Did you mean reflog?

Either way, even simpler, imho, than any log that one has to comb through after the fact is to create a named backup

  branch=$(git branch --show-current) && git switch -c backup-${branch} && git switch -
Carry on as planned and if you bork it all, switch to the backup branch which retains the original commits and all, delete the borked one and have another go

  git switch backup-somebranch && git branch -D somebranch && git branch -m somebranch


You don’t have to comb through the reflog for the pre-rebase branch state. Use `@{1}` from the reflog of the branch (not `HEAD`).[1]

Note: First I thought that `ORIG_HEAD` was the thing. But that won’t work if you did `git reset` during the rebase.

(`ORIG_HEAD` is probably “original head”, not “origin head” (like the remote) that I first thought…)

[1] You just have to comb through documentation!


I do this as well but `git reset --hard backup-somebranch` and try again if I mess it up.


I have a custom bash function named "backup_branch" that does exactly that, along with "restore_backup" and "delete_backups". It's made my life 10x simpler.


GP might have meant that the researchers are veiling their actual goal, in the vein of "detecting humans in drone feeds and autonomously dropping bandages on them for search and rescue"


This is how I interpreted the comment.

Spy satellites surveying Iran in order to detect underground water infrastructure seems like, to me, the ulterior motive is to identify infrastructure so that it can be compromised in some way, likely with drones and bombs.


There's a repo with categories that gets updated every now and then

https://github.com/danluu/post-mortems


People do not protest when the government is already doing what they think is right.

What should people do, meet up just to chant "thanks for doing the thing, keep doing it" ?


That would be nice, yes. Like in Civilization where if the citizens are really happy they hold a "we love the king" event. Then your population expands a lot immediately after.


As others have mentioned in this thread, russia is a terrorist state. This is but one example.

How about at least pushing officials to give an honest answer as to why they are not on the list of terrorist states?


The honest answer is they have a lot of ICBMs. And besides, what would it do? We’ve got no more sanctions left. Short of invading (which nobody is going to do because of those ICBMs) there’s nothing left to do. We shot our wad on Ukraine sanctions.

At this point unless we somehow convince the big nations that still trade with them (like China) to stop, we’ve got no levers to pull, and we don’t care enough to give China whatever massive ask they’d have for that.


> The honest answer is they have a lot of ICBMs.

That is not the honest answer. Also, the response to the threat of ICBMs is well known, it's called mutually assured destruction.

The honest answer is that it would lead to sanctions on countries who continue to do business with russia. These are countries who we could like to consider as allies, such as India.


Not really. Why would it? Those same allies trade with other nations we’ve designated as terrorists.

And we’ve already pulled basically every sanction lever against Russia anyway. At this point we’ve probably sanctioned them more than anyone we call a terrorist state.

The difference is nukes.


I respect your opinion, even though I disagree.

FYI, the website in your bio is down.


Ha I think I joined here in 2007. It worked back then.


The only reason that I found this is that I recently made a Firefox extension that shows a tooltip with the user's bio, when hovering over a username on this site.

I am not sure what to name it, HN Inappropriate Snoop?


What to do? Instead of only giving them scraps, send some serious weapons their way. Stop the trickle support, step up.


No, the honest answer is much more mundane - natural gas and Europe’s continued reliance on it to get through the winter and keep overall energy costs a bit more controlled. Who knew that denuclearizing the energy grid would have such benefit for Russia /sarcasm.


Color sequences are not portable and known to break things (e.g. Jupyter sessions crashing due to colored pytest reports) and nothing but a liability if you're just piping the output. I think it's more about having the option.


Sounds like those pieces of software should also work on being more resilient to color sequences.


Major advantage of Fedora is being closer to the upstream sources, both in terms of freshness and in terms of not meddling with libs or similar. Debian patches lead to several possible exploits over the last few years.


Fedora removes elliptic curve algorithms from the source code level [1] and disables hardware acceleration for H.264 / H.265 [2].

[1] https://bugzilla.redhat.com/show_bug.cgi?id=615372

[2] https://www.phoronix.com/news/Fedora-Disable-Bad-VA-API


Yes, distributing non-free, patented code that requires a license, requires a license. The same goes for Debian actually[1], including blocking requests and removing packages that were included before by mistake.

I would even dare say that this is another point for Fedora, enabling https://rpmfusion.org/ is a one-liner and feels entirely native, never a broken package.

[1] https://www.debian.org/legal/patent


RPM Fusion does not give me uncrippled crypto libraries. It’s caused by their paranoia about export restrictions, not patents.


> Debian patches lead to several possible exploits over the last few years.

Which ones? There was the OpenSSL entropy bug, of course, but that was 1. in 2006, and 2. run by upstream so feels a bit unfair.


I have to admit that I never compiled a list of this type and it seems exceedingly difficult to find useful search results. I couldn't dig up the examples I had in mind from the last 2 years, but stumbled upon others I didn't know of yet in turn, e.g. RCE via Redis, no special config required:

> This post describes how I broke the Redis sandbox, but only for Debian and Debian-derived Linux distributions. Upstream Redis is not affected. That makes it a Debian vulnerability, not a Redis one. The culprit, if you will, is dynamic linking

https://www.ubercomp.com/posts/2022-01-20_redis_on_debian_rc...


Consider applying for YC's W25 batch! Applications are open till Nov 12.

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

Search: