Androinc

[STMahlberg]

About AndrOINC (http://androinc.net/)

http://androinc.net/

AndrOINC is a boinc project trying to break a 1024 bit RSA key used by Motorola to sign the boot and recovery partitions on the Motorola Milestone. If this key could be broken, we would be able to sign our own boot/recovery images, which would enable us to run custom kernels and recovery images on the Milestone.

[DrPop]

Everyone who has a CPU (that should be EVERYONE, am I being redundant?) should be on this now! It looks to double even what AQUA was giving me over a 24 hour period!

Thanks for the heads up Zombie and Maxwell!

[Dandasarge]

This also is cracking a code, at these points this will not be a vary long mission. unless they are going to do more but to try a billion codes will not be that long depending on how many computers are on it. Lots of points +1 bing #1 on something noone can take it away from us un +1

[Dandasarge]

This also is cracking a code, at these points this will not be a vary long mission. unless they are going to do more but to try a billion codes will not be that long depending on how many computers are on it. Lots of points +1 bing #1 on something noone can take it away from us +1

[Maxwell]

This also is cracking a code, at these points this will not be a vary long mission. unless they are going to do more but to try a billion codes will not be that long depending on how many computers are on it. Lots of points +1 bing #1 on something noone can take it away from us un +1
It's substantially more than a billion codes. It's 1024-bit encryption, and a brute force method will take longer than the lifespan of the universe to finish, even with 10 times the computing power the project currently has...

[STMahlberg]

Everyone who has a CPU (that should be EVERYONE, am I being redundant?) should be on this now! It looks to double even what AQUA was giving me over a 24 hour period!

Thanks for the heads up Zombie and Maxwell!

I've got all 16 cores of mine on it. :)

[Dandasarge]

http://news.techworld.com/security/3214360/rsa-1024-bit-private-key-encryption-cracked/

You can do something like this narrow down about 50-75%. There are systems that have already cracked 2048bit. The nature of cracking electronics is useless if its going to run more then 6mo to a year. The item is outdated and useless by the time you figure out how to get into it.

[Maxwell]

http://news.techworld.com/security/3214360/rsa-1024-bit-private-key-encryption-cracked/

You can do something like this narrow down about 50-75%. There are systems that have already cracked 2048bit. The nature of cracking electronics is useless if its going to run more then 6mo to a year. The item is outdated and useless by the time you figure out how to get into it.
Yep. Agreed. My understanding was that the motivation was "eh. I'm curious, so I'll run a BOINC project, and not care too strongly if I ever actually solve the problem I'm testing out..."

[Teratoma]

I'm in. Always like good CPU credits, besides, I need to catch up on my MMs.

Thanks for the heads up DrPop!

[STMahlberg]

http://news.techworld.com/security/3214360/rsa-1024-bit-private-key-encryption-cracked/

You can do something like this narrow down about 50-75%. There are systems that have already cracked 2048bit. The nature of cracking electronics is useless if its going to run more then 6mo to a year. The item is outdated and useless by the time you figure out how to get into it.

This reminds me of an article I had read awhile ago. I am just paraphrasing but it basically said that the ability to crack a so called uncrackable encryption is simply a matter of equipment and money.

If you use the standard computers that most of use and have a limited budget, it would take you x amount of time to crack a 256bit key using alphanumeric and symbols. The time was, I don't remember exactly but I think the Sun will burn out before you could crack it.

On the other end of the spectrum, was the ultra elite computer which didn't exactly run any kind of store bought software, in fact they said you and I could never get one of these things... some sort of government covert computer thing and pretty much an unlimited budget to... well... purchase this thing... they could crack it in seconds.

And that's the stuff they're willing to tell you about... and people think their 4 digit ATM pin is safe. :rolleyes:

[DrPop]

I don't really think the project will end any time soon, because isn't DNETC doing a similar type of project? They're not even close to being done, and they're using HUGE amounts of GPUs...
...this project on CPU alone would take forever. When they port it to GPU, we might see it finish much sooner, depending on the credit they give out, and if we're all willing to throw thousands of high end GPUs at it.

But...by then we'll have SO MANY MM credits on this puppy that we'll actually be happy when we do crack it and the project completes successfully. :D

[zombie67]

I think androinc is being managed by the same people as dnetc.

And that means that the first GPU app will be ATI.

[NJCaNS]

"AndrOINC is a boinc project trying to break a 1024 bit RSA key used by Motorola to sign the boot and recovery partitions on the Motorola Milestone. If this key could be broken, we would be able to sign our own boot/recovery images, which would enable us to run custom kernels and recovery images on the Milestone. "

I was wondering if it's actually legal or how legal it is to so blatantly sit there and "hack" a private encryption key? Are we now accomplices in a criminal activity? ;))

[Slicker]

"AndrOINC is a boinc project trying to break a 1024 bit RSA key used by Motorola to sign the boot and recovery partitions on the Motorola Milestone. If this key could be broken, we would be able to sign our own boot/recovery images, which would enable us to run custom kernels and recovery images on the Milestone. "

I was wondering if it's actually legal or how legal it is to so blatantly sit there and "hack" a private encryption key? Are we now accomplices in a criminal activity? ;))

One could argue that it isn't illegal to pick a lock. You just can't open the door because that's when it becomes breaking and entering. The cops on TV always try the doorknob and if it is open assume they can enter and everyone knows that if you see it on TV or read it on the Internet, it must be true. ;)

[STMahlberg]

One could argue that it isn't illegal to pick a lock. You just can't open the door because that's when it becomes breaking and entering. The cops on TV always try the doorknob and if it is open assume they can enter and everyone knows that if you see it on TV or read it on the Internet, it must be true. ;)

It is my understanding that "breaking and entering" are two separate criminal acts. So if one were to pick a lock or open a door on a property that is not yours you are now chargeable with "breaking"; so if you pick a lock, open the door and enter... you can now be charged with 2 counts of breaking, 1 of entering and undoubtedly trespassing.

I don't watch COPS but my Uncle is a retired Captain in California.

@NJCaNS - But if anyone asks... you haven't seen me. :P

[Beerdrinker]

Is it just me or is all my WU´s pending...Haven´t received ANY CS since last nigth...

[STMahlberg]

Is it just me or is all my WU´s pending...Haven´t received ANY CS since last nigth...

Mine are getting validated right away.

[Beerdrinker]

Mine are getting validated right away.


Haven´t received a single CS in the last 12-15 hours....And then...Bang...When I wrote this post....CS flowing...!:cool:


Edit: Granted 10.000 CS on the last 20 minutes.....

[Crazybob]

As I mentioned to DrPop, any of my machines running Win2000 errored out. It doesn't like them.[-X

[Mr. Hankey]

Well credits are reporting on the site, but it looks like they haven't exported to the stats sites yet. I also am still seeing significant validation errors using the 64bit linux 1.02 client. On the other hand since Aqua is down it gives me something to point my stuff at so I expect I will see more errors than most. I think I have 120+ validation errors on the 1.02 app but not sure how many have been done in total.

[Maxwell]

The validator is down right now. They are aware of this. This is why you're not seeing credits...

[Mr. Hankey]

The validator is down right now. They are aware of this. This is why you're not seeing credits...

Validator is running... the site just isn't exporting... I have no tasks waiting for validation.

I have made some more observations. It looks like the 1.02 linux client is about 50% slower than the windows client. Even once they get this fixed I don't see it being able to out produce Aqua, it should be close though. I get 16.5k / day with my quads on aqua looks like I should get around 12k here once they fix the performance issues.

[Crazybob]

Keep your eyes on Aqua. They are doing server maintenance. I was going to leave my Aqua boxes alone, but some have run out of work.

[zombie67]

Looks like everything is back to normal now. Validator running, and stats exported.

My windows machines usually make around 20k/day. They made over 100K yesterday. And the day before (before they cut the credits), they made over 200k. :D

[Beerdrinker]

Looks like Sic is in for a surprise...

They think they are getting comfortable at #2.....In fact they are #3....We blew rigth past them!!

[Fire$torm]

Looks like Sic is in for a surprise...

They think they are getting comfortable at #2.....In fact they are #3....We blew rigth past them!!

OH Yeah Baby!! U.S.A!!! U.S.A!!! U.S.A!!!

[STMahlberg]

Nice!!!

Looks like most of the bugs got shaken loose... my rigs are all running fine now. :)

[Maxwell]

Looks like the Hankey and Mumps flotillas have ironed out their issues...

[trigggl]

Looks like the Hankey and Mumps flotillas have ironed out their issues...

...and my dinghy.

[zombie67]

So, at first there were just very short tasks, each paying 29 (awesome) credits. Then longer tasks paying 115 credits. Better than most projects, but not quite as awesome. It looked like they changed the credit formula. But I notice the short/29 tasks are still being mixed in. Now, I see 58 credit tasks. So it looks like there is a third category now. I wonder if there is a way to force the best paying tasks?

[Beerdrinker]

Please also notice that our total RAC at BoincStats took a dive.....

But only until Willy adds this project...Then we can see some progress!

[John P. Myers]

As a comparison, DNETC is trying to crack 72bit RSA. Going from that to 1024 bits....ehhh...here's DNETC's current progress:

Total Blocks to Search: 1,099,511,627,776
Total Blocks Tested: 15,225,010,302
Overall Rate: 59 Blocks/sec
Total Keys to Search: 4,722,366,482,869,645,213,696
Total Keys Tested: 65,390,921,328,353,083,392
Overall Rate: 254,399,787,303 Keys/sec
Percent Complete: 1.385%
Time Working: 2,975 days

And DNETCs expected finishing time:

The odds are 1 in 32,055 that we will wrap this thing
up in the next 24 hours. (This also means that we'll
hit 100% in 32,055 days at yesterday's rate.)

So jumping from just 72bits to 1024? Yeah this project will be around awhile.

[Crazybob]

Anybody else figure out how to get it to run on slower machines? My Win2000 machines, I can see why, but I have a couple of P4's with XP and it still errors out. Could add a couple more thousand a day if I could get'm running.

[trigggl]

Anybody else figure out how to get it to run on slower machines? My Win2000 machines, I can see why, but I have a couple of P4's with XP and it still errors out. Could add a couple more thousand a day if I could get'm running.

My P4 works in Linux. Perhaps you can try the live cd Dotsch-ux.

[Fire$torm]

Anybody else figure out how to get it to run on slower machines? My Win2000 machines, I can see why, but I have a couple of P4's with XP and it still errors out. Could add a couple more thousand a day if I could get'm running.

All my boxes other than my desktop are P4s 2.8Ghz~3.2Ghz. Only my box with Windows 2000 Pro cannot crunch this project. The other boxes run XP Pro, two are 32 bit and the other is 64 bit.
Also my two laptops can crunch this project and they have Pentium M CPUs running Ubuntu.

[zombie67]

Anybody else figure out how to get it to run on slower machines? My Win2000 machines, I can see why, but I have a couple of P4's with XP and it still errors out. Could add a couple more thousand a day if I could get'm running.

I know of at least two projects that require at least xp sp2. sp1 does not work. This may be one of those.

[Fire$torm]

I know of at least two projects that require at least xp sp2. sp1 does not work. This may be one of those.

The AndrOINC website lists the Windows app as Win98 compatible so go figure.....
And anyone running XP should have had sp3 installed automatically via Windows Updates. You can also DL the XP-sp3.iso image file from M$ Downloads if they wish to do a manual install or use the image file with nLite to slipstream the update into a new/custom Windows install CD.
I gave up trying to get this project running on my Win2K box.

[GregK]

does oc'ing help significantly? i know it raises temps fast. im at 2.8 now and 70 is highest core temp if i go to 3.0 its about 90-92c.

[Fire$torm]

;5945']does oc'ing help significantly? i know it raises temps fast. im at 2.8 now and 70 is highest core temp if i go to 3.0 its about 90-92c.

90c is way too high. Did you increase the fan speed on the GPU. Some OC tools will only allow you to set the fan speed to a single setting. If yours does this then set it to at least 75% of full/high rpm.

With MSI Afterburner and Rivatuner tools you can setup multiple fan speeds based on GPU temperature.

http://img255.imageshack.us/img255/1006/fancontrol.jpg