Hacker Newsnew | past | comments | ask | show | jobs | submit | xd's commentslogin

I took my services offline last night, wouldn't have been able to sleep otherwise. There has been core PHP members that have outright said the mitigation's currently doing the rounds won't help and even cloudflare can only mitigate against known attack payloads.

The maintainer's comments are singularly unhelpful - "there's nothing you can do, just curl up and die" is pretty useless when you're faced with "take the entire company site offline" vs "maybe get exploited."

The comments aren't unhelpful, they're in fact very helpfully telling you that "take the entire company site offline" is the only safe choice right now.

But that's exactly why it is unhelpful. "Take your site offline" is always safe - but we don't have enough information to make the decision informed.

"X may mitigate some of the active attacks" is helpful - you can evaluate if X impacts you, and make a choice.

"Y is how we're seeing the attacks" is helpful - you can setup logging to see if you're being exploited.

"Z reduces but we can't say it eliminates the risk" is similar.

"SHUT IT ALL DOWN" doesn't do much but make good noise and increase the chance that the next one is considered to be a "boy-crying-wolf" scenario.

(Of course, taking all dev/test/unimportant sites down for a few hours is painless, but Wordpress/PHP runs some of the biggest sites in the world.)


So what if there are no known mitigations and it's such an open attack surface that logging doesn't help in containing it?

Since you wrote your comment the official blog posts describing what exactly is the in the exploit have been released, and "shut it all down" was definitely the right advice to give.


Yes, that's not very well done: when you are handling big IT services, you need more than "a person wrote a scary thing on twitter" to justify such an impacting action

Does CERT-EU telling you to shut your site down count

Honestly, not really, no, unless you've setup "CERT-EU controls our website" in your policies before.

After all this is done, CERT-EU isn't going to write the defense of your actions for you.


Bit of context. Deleting after import is unlikely to have zero'd out the data on the SD - so using photorec would have allowed them to quickly and easily recover their pictures from the SD card.


Not "easily" as the filenames and metadata that associates them with the file contents would have likely been overwritten. Also, not sure how well photorec works with Olympus' RAW files.


The new legislation likely won't apply to you: https://commonslibrary.parliament.uk/research-briefings/cdp-... - the section "Threshold conditions for categorisation of services".


A very relevant recent story: https://news.ycombinator.com/item?id=43692677, the top comment especially.

I'm not American, I'm from the UK, but it's the same in regard to putting value in to people.


The video has covers how Tool and Die wages have dropped below the median wage. We're also importing the tools/molds from China and manufacturing the parts "in the US" instead of the reverse.

I remember another Youtuber Linus Tech Tips creating their own injection molds in BC so that it would be harder for copycats. The entire Screwdriver isn't made in Canada though. The shaft and bits are made in China or somewhere else in Asia IIRC. The reason they were able to do that though is because there already existed an injection molding company nearby for them to work with though.


From a quick skim of the repo it looks to be injecting phptop_hook.php into any call to a php file with auto_prepend_file in php.ini. phptop_hook.php generates some stats on how long it took the php file to process and stores this via error_log() which can then be queried with the perl script phptop.


The thing with programming, to do it well, you need to fully understand the problem and then you implement the solution expressing it in code. AI will be used to create code based on a deficit of clear understanding and we will end up with a hell of a lot of garbage code. I foresee the industry demand for programmers sky rocketing in the future, as companies scramble to unfuck the mountains of shit code they lash up over the coming years. It's just a new age of copy paste coders.


To me, "Unknown" almost implies the possiblity of a value, whereas I've always thought of NULL as being an absence of a value.

edit: an empty string, false, 0 are all values.


I'm not sure where you've got this idea from. Even "new builds" in the UK, with there questionable building methods (which you may be referring to?) are mostly insulation and incredibly efficient to heat.


The new builds are made of brick and block, and are indeed insulated really well. It's the millions of older houses, particularly Victorian ones, that are made of brick, are not usually insulated well, and are often intentionally draughty.


What would the reason for being intentionally draughty be? The only thing I can think of is if they were burning candles/gas lamps to improve air quality?


To reduce condensation and mould.


Damp is worse than cold


I used https://wiki.osdev.org/Resources back it the day and it looks like it's still alive and well!


I've a similar way of interacting with the API of my product but instead keep everything prefix notation (lispy) so for example;

"and": [ { "property": "material", "operator": "equals", "value": "carbon" },

Would be;

["and", ["=", "material", "carbon"], ...]


Is there any standard for describing filters in json like this, or any well known library that is also able to parse this back to expressions?


In my humble opinion, it would benefit all better if we standardized something like jql (jira query language, so the user layer) first. From experience, many who find sql too complex have no problem with writing complicated jql queries. it's both easy to use and a great first step to writing actual SQL. I wish it was a standard like json.


The first thing that came to my mind is the mapbox espression language/spec, which uses a very similar format:

https://docs.mapbox.com/style-spec/reference/expressions/


The way the JSON filters work would be directly coupled to the JSON fix notation and SQL infix notation so it's directly coupled to SQL and doesn't really require a standard, imho.



If you're reinventing a language, it's probably not a bad idea to reinvent the simplest one!


Just wondering, why order =, material, and carbon that way instead of material, =, carbon?


It's "prefix" notation .. if you search something like "prefix vs infix vs postfix programming languages" it should return a decent explanation. For my case it's personal choice more than anything.. also easier for me to write a VM for.


Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

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

Search: