l-albertovich

That's interesting, some time ago I saw a video about a live captioning system in Louis Rossmans (youtube id r09Hm2zd2lY) where someone asked the same question and the answer was that they only accepted local applications.

Did I misunderstand his answer or is this something that changed?

While I agree that previous engagements might have found vulnerabilities that may or may not be addressed (after all, that's up to the client) I think a core part of a decent source code audit or pentest is to find those so if that was the case then it could speak of the proficiency of those involved.

In my own experience I've noticed that a lot of times it's really hard to get the clients to address these sorts of issues, even when it's clear that they are critical and could directly translate to monetary loses which makes no sense to me but sadly doesn't surprise me at all.

Just to be clear, of course there is a scope and a web service audit doesn't usually end up with a thorough audit of all the packages used by the application because nobody will pay for that and even if they would, it's usually mind numbing enough to end up in people glancing over things and checking boxes but in most cases this is not the root of the issue.

Yeah, not only the story was really interesting from a technical point of view but the storytelling and the social aspects of his relationship with his SO and their social circle were really enjoyable.

I even chuckled when he got mad and went to the garage to sand stuff (I won't spoil the rest) because of how close to home the entire thing hit.

Oh boy! Ever since I listened to The cuckoo's egg I wanted to reach out to tell you how much I loved it, I enjoyed ever single minute out of the almost 13 hours so thank you very much for it.

I'm looking forward to reading or listening to Silicon snake oil and High tech heretic but those are not available in Storytel so I'll have to wait until I have time to read them in paper.

I'm glad you used it, even if to some your application doesn't need it, it's good for the toolkit and I personally like how it looks.

Additionally, I think that your project might be useful to people and the only reason it exists is because you invested your own time on it so if using tauri was something that motivated you or gave you any type of satisfaction then I'm happy for you.

Thank you!

Well, it runs (insert dramatic pause) over people.

According to books such as "¡Es la microbiota, idiota!" (ISBN 8413440688 but I don't know if it's available in english) it's not that simple, I think they said that not all the strains are identified and even then there are strains that aren't easy to grow in vitro (I'm NOT a scientist, don't quote me on this, I just listened to the audiobook).

What they make a point on is diet, she insists on the importance of having a diet that gives desirable bacteria strains an environment where they can thrive to achieve a balance.

IIRC she also mentions how it's not as simple as having "good" and "bad" strains and how balance is key.

It's a really interesting and funny book to listen to if you have a few hours for it.

So in the case of OP I think you're right about finding out more about the donors diet and replicating it as well as adjusting anything necessary for the newly arrived colonies to last through time.

I don't know sh!t about it nor do I intend to sound like I do so please correct me if I'm wrong.

Glanced over it, it doesn't look like a bug, just wasteful.

The way I see it, because sc_array_clear resets cap and size to zero, the next call to sc_array_add will issue a realloc call to the default minimum size (8) and it won't leak because sc_array_clear doesn't overwrite elems.

So yeah, not cool but doesn't look like a bug either.

Everything is handled in process, you can find some examples in the runtime test cases.

[1] - https://github.com/fluent/fluent-bit/blob/master/tests/runti...

I wrote the code for the OpenTelemetry metrics I/O plugins and the underlying decoder and encoder, I'd be happy to answer any questions related to it.

OMG, I just went to Eluktronics' site out of curiosity and saw LPP which sounds AMAZING. I've been thinking of something like that for a long time and seeing it readily available just blew my socks off.

I implemented something very similar in C using shared memory and a custom multi producer / consumer ring buffer for a different exchange. It was NEAT.

I loved how the code turned out and how performant it was and would love to talk to people who went through the same.

They point of having 2 drives is being able to better isolate your secure system.

If you physically unplug your secure drive then you know for sure that no fun stuff is going on behind your back.

I'm not entirely sure about secure boot when it comes to validating the boot loader of an installed system so I can't vouch for that but I do know that UEFI kits aside, unless you have a proper way to validate the secure systems boot loader it could be tampered with if the disk is accessible by a compromised system.

IMO the point is not to get rid of windows but to isolate whichever insecure system you use in a way in which it can't hurt the secure system (it could even be a secondary Linux system where you use Wine if that's what you want)

If you don't have the money or space what you can do is just have 2 hard drives, one fully encrypted running whichever OS you want and one with windows. Unplugging the secure one when booting into the insecure one would be ideal but not required. In this case you'd still be vulnerable to certain EFI based attacks but at least you'd be safe from most common crooks.

I mean, seriously, running stuff with wine is OK if your expectatives are aligned with reality which is, you're running stuff there. If you complain about wine but then are completely fine with the typical 'curl XXX | bash" well, yeah, you're SoL but other than that some degree of separation is good.

And of course a very easy to exploit bug that has been around for like 12 years hasn't gone public a few days ago because LPEs DO NOT EXIST IN LINUX.

I mean, it's quite obvious, exploitable bugs do not exist in Linux because back in the day Ken, Dennis, Brian, Doug and Joe agreed that they were not part of the Unix philosophy so they left them for windows to implement. /s

Aside from the stupid joke, what OP is saying is absolutely right, if you care about security keep your things properly isolated, keep a (whichever OS you trust but hopefully linux or a bsd) as main box and then a windows or macos computer isolated (best case scenario using a vlan capable switch, or a dedicated pi that acts as a firewall just for that box).

I think you'll need to repeat that a little bit louder beacuse it seems like they didn't get the news in the middle east.

What I found interesting about this was the bluetooth / offline communication part.

Seems like a great opportunity, I can't say I've ever been disappointed with an article I read from them and it really seems like they know their stuff so there's probably a lot of really interesting people to meet there as well as interesting stuff to work on!

Not OP but thought I could help :

Usually you have to use (or create) an unpinner frida script. In my experience it's been easier to just write down custom dumpers to avoid having to deal with that and also modifying the request (at least some time ago mitmproxy downgraded http/2 to http) which would result in the endpoint being able to detect the tampering.

As for the emulator, I haven't ever used the official one to work, I know some people who used Anbox in Linux but I don't have much to say because I use a real device most of the times (although I pleayer with ldplayer, bluestacks and genymotion all of which behaved nicely to some extent).

Not OP but thought I could help :

To some degree yes, apktool will unpack the apk and disassemble the dex files in a way in which you can modify them and then compile the project again, sign it and deploy it (as long as your device supports it).

In this case you would have to either modify the smali code itself to make those changes.

When the modifications are basic such as changing a few instructions to avoid checks or adding simplistic payloads writing them in smali is just fine but if your modifications are more substantial I'd suggest creating a placeholder java project, decompiling it and adapting that smali so it's less of a pain in the neck.

Another valid approach would be embedding a custom shared object or fridas gadget with a configuration file to run a payload on start.

As for understanding what you need to change and how, jadx is the way to go most of the times but for some nastier code you can either use dex2java and then jd on the converted file or just go straight to the smali code (honestly that's not that usual)

Also, while testing your patched version I'd suggest heavily using and abusing frida in a rooted device if possible.

Note : I don't usually use the "create a java project" option so your mileage might vary but I recon it's probably the easiest.

Well crap, there you go, giving the lunatics reasons not to get tested and make the whole thing worse than it already is...

I'm joking, I don't think you did anything wrong but I'd hate it if a ridiculous argument such as this example gained any traction :

Example : The government / aliens / whoever released the virus so we willingly gave them our DNA to sequence and match with our assigned ID so they can do XYZ in case the implant in the vaccine doesn't work or if we are smart enough not to get vaccinated.

Scary stuff.

Sorry about the late reply but this is something I've been thinking about for a long time and something for which I have been trying to design my own carrier board while waiting for the turing pi 2.

Think about Qubes OS where you define different "security contexts" for diferent workloads where each of these run in isolated virtual machines. If properly implemented it's a pretty sturdy solution but you're one hypervisor away from total compromise regardless of what you do.

Now think of the same thing except this time you have one main system that acts as a firewall, shared resource, pxe and window server and a few secondary nodes that boot up from lightweight static images they get from the main node and mount specific resources from it to achieve a sense of persistence (ie. firefox / thunderbird profiles, etc).

There you have a pretty decently marked attach surface so you know which services you need to audit and isolate in the main box.

In my mind I'd love to have the same thing but with the compute modules laying parallel to the main board to be able to fit it in an oversized laptop form factor (kind of like the mnt reform) but this would be a really cool middle ground.

I plan on getting to the software side of it as soon as I have some spare time because even if this doesn't work or isn't ever in stock the whole thing could be put together in a ghettoish way using regular PIs and it'd still be something that could be fit in a mini tower with a small switch which would be good enough for me.

To me the big plus is compartmentalizing, having my games in an isolated environment designed /optimized for that makes it easier to keep my computer safe because I no longer have to consider GPU driver availability / stability in Linux or even running Windows in my computer and that's a big plus.

That or you could go the concerned ape way which is way cooler IMO.

Huh, a few days ago I listened to the darknet diaries on kik, I guess imgur is about to become a child porn den...

Sarcasm aside, it'd be cool if they got their shit together.

I really like the concept, the only thing I'd nitpick is that I didn't quite dig the stack spread (C++, go, php and python) and that I found the grommox code very hard to follow, not due to the inherent complexity of the tasks but because of how inconsistent and messy the code is which makes it difficult to consider it an improvement.

IIRC someone mentioned something about it running apps with waydroid or anbox (can't remember) a few days ago in a related thread right? Have you tested it?

Up till now I was thinking this seemed very interesting and the price point of getting it wasn't bad at all (quickly searched amzn to see what the Xperia 10 II price was) but if what you say is true then it would be kind of worrying.

Edit: For some reason I missed a bunch of words when I posted.

Sorry about the late reply, I didn't realize there was a reply to my message. I reviewed it once again and can attest that what you say is correct, so that's cool.

Thanks for taking the time to reply!

I was using iMessage as an example of something tedious and not overly complex but not exactly low hanging fruit that's yet to be completely reversed. The S/N or certificate parts don't even matter, if people had reversed their way through it, there would be at least an option to extract the required parameters out of your hardware and plug them in a stand alone server (in fact, IIRC there a valid S/N generator some time ago that was used to deploy osx in kvm?).

So, the idea is that even though it's not an impenetrable fortress, there are still plenty of dark places to introduce subreptious changes.

As for the human-in-the-loop part, I don't know why did you get so snarky, what I was talking about was that this is the layer that should be scrutinized the most, all of those components because even without the technology those are the people that will put someone in jail with no verifiable evidence.