No, I can't access it either.is it just me
No, I can't access it either.is it just me
still down! any news?
Use the right tool for the right job!
i have several p3's and celeries i wanted to switch to SoB, can't get client/work!
Use the right tool for the right job!
Did someone mail Dave about this? I don't think I have his mail (it's on the website ...)
I finally have a new factor from P-1 that I would very much like to submit
i hear the site is down. our ISP is having connection issues. denial of service attack of some sort. the server itself is ok but it's connection has been down since 12:30am this morning.
if someone knows of a more stable internet provider that could co-locate the server sometime in the future, we could talk about relocating. email me directly.
Cheers,
Louie
That is a bummer that the site is getting hammered down! I'm sure it's going to be difficult when it comes back online, from the deluge of clients 'waiting' for new blocks....
To bad the client could not store more tests that just the current one....
It's working but it is veeeeery slow. It took one minute to get a response from the website, and mklassons sobistrator is still working (after 5 minutes) with submitting 10 factors.
I've been reading the forums of a lot of the larger teams and it looks like most people are just taking their boxes and moving onto other DC projects.
I hope the server becomes stable soon before we bleed away most of our members.
Some of our members reported server connection problems during the last 36 hours, and they are going to leave the project, if the problem's aren't solved within an expected time period.
I hope, you're already working on a solution.
While waiting the server to be back, you could help with sieve and/or P-1 factoring. It's a great chanche to boost P-1 factoring .
And what is keeping you from coming back? If you really want to help this project (and have some free RAM), please do some P-1 factoring. If you're more interested in stats, sieving and P-1 have their own stats.
Well, for me it would help if I could reach the server a little more than once out of every 30 attempts.Originally posted by Troodon
And what is keeping you from coming back?
my boxen finally were able to connect tonight, and THEY ALL DUMPED THEIR BLOCKS AND GOT NEW ONES!!Originally posted by Troodon
And what is keeping you from coming back?
WHY DID THE SERVER RESET ALL MY BOXEN!?!?
i am taking a vacation, i'll be back to try again in a few weeks/months
Use the right tool for the right job!
2 days later and still can only connect 1 out of every 10-15 tries.
I'm about ready to give up as are a lot of my team-mates.
What can I say.. it's hosted on a connectino that's been eating a DoS attack the past few days. I've tried to alleviate the problem and not have much.. then on top of that, yesterday we had a bad storm move through the area and knocked out power for 4 hours..
What type of bandwidth do you need for the box?Originally posted by Alien88
What can I say.. it's hosted on a connectino that's been eating a DoS attack the past few days. I've tried to alleviate the problem and not have much.. then on top of that, yesterday we had a bad storm move through the area and knocked out power for 4 hours..
Do you own the box or would you also need one setup for you as well? If so, what would the specs be on a box that would handle the site and database?
Wouldn't need much bandwidth at all.. it doesnt do much data transfer right now. Sustained it'd only do at most 10k/s outgoing.
The box was basically put together out of spare parts when the old server failed, so ideally a new server would be best. Something in the 1ghz+ range along with at least 512mb of ram and a decent size hard drive (40g)..
As a side note, it seems that for the last 10 hours things have been stable.. knock on wood
--
Mike
Mike,Originally posted by Alien88
Wouldn't need much bandwidth at all.. it doesnt do much data transfer right now. Sustained it'd only do at most 10k/s outgoing.
The box was basically put together out of spare parts when the old server failed, so ideally a new server would be best. Something in the 1ghz+ range along with at least 512mb of ram and a decent size hard drive (40g)..
As a side note, it seems that for the last 10 hours things have been stable.. knock on wood
--
Mike
Yes, it has been stable for a while now. (knocking on wood myself). I am a D.B.A. for a large datacenter. I'm going to hit my boss up next week to see if I can talk him into hosting a box for you.
I'll get back to you next week.
That would be awesome.. let me know, and if I dont respond here you can hit me up off the board at mgarrison@alienz.netOriginally posted by SDC
Mike,
Yes, it has been stable for a while now. (knocking on wood myself). I am a D.B.A. for a large datacenter. I'm going to hit my boss up next week to see if I can talk him into hosting a box for you.
I'll get back to you next week.
Well, if you don't know what to say, let me say it:Originally posted by Alien88
What can I say.. it's hosted on a connectino that's been eating a DoS attack the past few days.
It's been known for many months that the networking code in the Linux client is borked. On my XP2k/Slack9 machine it'll run just fine except that it'll stop processing when it cannot connect - puts a friendly line into the logfile and then just sits there and idles. I have a little wrapper script that kills the client when the load average drops under .4 and so it'll start up, re-finish the last chunk of work, be unable to connect and fall idle again. Ad infinitum.
Now nobody has found it in their heart to FIX the stupid Linux client and nobody has found it acceptable to open the source so that some volunteer can look at it. At that rate it is patently disingenous to blame some DoS attack: it is a certainty that things like this happen on the internet - the question is only "when", not "whether". And the responsible coder prepares for them by making the code at least marginally stable.
I had mentioned this months ago: this project reminds me eerily of Seti@Home: not enough staff to handle it professionally and no willingness to allow a couple volunteers to help with the problems that come up.
Let's say a new prime is found tomorrow and the publicity attracts enough people to double the current number of CPUs -- this project would be utterly overwhelmed. Even though this would be a miniscule amount of power compared with a lot of other projects.
I can certainly see how people are losing patience with this. I'm not quite there yet myself, but I really hope someone actually starts fixing the zillion problems with this project...
Yes, it has been many months. When the bug first started to occur, louie and I worked on debugging it and trying to figure out what was causing it but we could not track the bug down.Originally posted by Lagardo
Well, if you don't know what to say, let me say it:
It's been known for many months that the networking code in the Linux client is borked. On my XP2k/Slack9 machine it'll run just fine except that it'll stop processing when it cannot connect - puts a friendly line into the logfile and then just sits there and idles. I have a little wrapper script that kills the client when the load average drops under .4 and so it'll start up, re-finish the last chunk of work, be unable to connect and fall idle again. Ad infinitum.
Now nobody has found it in their heart to FIX the stupid Linux client and nobody has found it acceptable to open the source so that some volunteer can look at it. At that rate it is patently disingenous to blame some DoS attack: it is a certainty that things like this happen on the internet - the question is only "when", not "whether". And the responsible coder prepares for them by making the code at least marginally stable.
I had mentioned this months ago: this project reminds me eerily of Seti@Home: not enough staff to handle it professionally and no willingness to allow a couple volunteers to help with the problems that come up.
Let's say a new prime is found tomorrow and the publicity attracts enough people to double the current number of CPUs -- this project would be utterly overwhelmed. Even though this would be a miniscule amount of power compared with a lot of other projects.
I can certainly see how people are losing patience with this. I'm not quite there yet myself, but I really hope someone actually starts fixing the zillion problems with this project...
To say that nobody has found it in their heart to fix the bug is a quite rude. Louie is a full time college student, I don't have the experience to find the bug, and Dave is way too busy with his job to work on it. There are plenty of threads that explain why the code is not open source and at this point will not be, feel free to go back and read those, I don't feel like reiderating the point those threads make.
If you don't believe me that I'm being DoS'd, I will gladly show you over 100mb of log files showing the attack on my line.
None of us make any money off of this project. None of us receive any donations or grants from anyone for this project. We all have our own real life things that take priority over this project. The server is built out of spare parts that came from my room.
This project has accomplished a lot in the time it's been up, it's gotten a lot bigger than any of us ever expected and attracted a lot more attention then we expected. And like I said, real life takes priority over this project.. and being rude to those of us involved with it will not get you anywhere.
Don't know 'cept this to this windows user, it sure has acted like something other than simple client problems...nothing personal but you're wrong on this one and owe somebody an apology...Originally posted by Lagardo
At that rate it is patently disingenous to blame some DoS attack: it is a certainty that things like this happen on the internet - the question is only "when", not "whether". And the responsible coder prepares for them by making the code at least marginally stable.
General observation here...Lessee...only been running this thing for a total of 45T between two accounts (not bad for a beginner) and yes, I've lost a couple of tests in the process over the last few months - which is a VERY VERY SMALL percentage of the total number of tests turned in...and a couple of the problems were because I lost power on my end...I'd call that "marginally stable"...and the last couple of days have been rough only because I couldn't get my stats fix every few hours but sweet Jesus it surely ain't nothing to get all worked up over like I've seen people getting worked up...
and yeah, I'll say it - this project may have a quirk or two and it may not be perfect but if any of you want perfect then perhaps you should go find another project to pick apart...
(this is Patrick's personal opinion and does not necessarially reflect the views of the management etc etc etc)
patrick.
I agree. This project, like everything else has a few bugs, like the linux code. But that isn't a big problem, I just don't quite get 100% production out of my linux box! If that scares other people away, great, that means I get a bigger chance of finding the prime
As for the DoS attack ... if executed well there's nothing you can do, even Microsoft had to take www.windowsupdate.com offline because of the msblast worm!
Those that Complain the Loudest, do the least --B2uc 2003
I'm currently starting a project off from scratch. When you begin something of this nature you find that no matter how smart you think you are, you can't do everything yourself. Doing everything from website, to DB, to clients, to tweaking your server, to praying you have enough money to pay for webhosting, to forums, to code...to code...to code. It all gets a little overwhelming at times.
This is a distributed nonprofit attack on a project of noncommercial gain. There isn't people sitting behind desk making $100k a year with nothing to do all day except tweak project specific code and pamper users that won't code that is 100% bug free. Even if it was funded out the butt..there would still be errors, bugs and problems nobody ever thought of.
SoB has done amazing things with such few resources and funding. A quality project that 'finally get's it' about stats and a good looking website/forum. A project that has real world merit and a project life that is resonable with todays technology.
Good work guys...you are doing a hell of a job. Don't let a few crybabies get you down. I mean what are they going to do? Go do RC5-72? O wait...their stuff has been screwed up for months.
Several people (including myself) have been posting gdb traces and core post-mortems to various forums including this one. We told you the name of the source file where you are doing something wrong, we told you the line number where you are doing it wrong and we told you what you are doing wrong (you allocate memory without ever freeing it).Originally posted by Alien88
Yes, it has been many months. When the bug first started to occur, louie and I worked on debugging it and trying to figure out what was causing it but we could not track the bug down.
Here's the first example that I can find on the quick:
http://arstechnica.infopop.net/OpenT...=22#6090913755
All that was left for you was the under-15-minute task of FIXING these things -- it is the one thing we helpful people out here cannot do because we cannot access the source.
You ignored all this. You completely ignored the vast resource at your disposal in the form of hundreds (if not thousands) of computer savvy people collectively known as "anandtech" or "arstechnica" or "freedc" etc.
It is the truth.
To say that nobody has found it in their heart to fix the bug is a quite rude.
If you find the truth "rude" then that is between you and the truth, not between you and me.
Ignoring the work that a lot of people have put into your project in the attempt to help you is at least as rude as pointing out that this is exactly what you have done.
So why do you refuse steadfastly to let one of the hundreds of knowledgable volunteers out there help you?
Louie is a full time college student, I don't have the experience to find the bug, and Dave is way too busy with his job to work on it.
If you don't have the experience, why do you insist on working on this without the help of any of the MANY people out there who do have that experience and would be happy to have a look at your code?
So why bring this up?There are plenty of threads that explain why the code is not open source and at this point will not be, feel free to go back and read those, I don't feel like reiderating the point those threads make.
"opening the source" != "open source".
You want to jealously cling to the code and deny even one or two external people to have a look at it and at the same time you tell us that you are incapable of fixing the problems that are clearly in there.
Trying to do everything yourself may just be a matter of lack of experience -- but insisting on doing everything yourself even after you have acknowledged that you can't -- that's just foolish.
So Louie is a full-time student. And? You imagine the users who are running the client are ... what? There's full-time students amongst us. And some of them are quite capable of writing stable networking code that works under noisy and tricky conditions. Which you guys demonstrably are not. So why reject the help of those who can do it for you?
Nobody is arguing this -- this is strictly a straw man.If you don't believe me that I'm being DoS'd...
Every server on the net will encounter high-traffic conditions for a number of reasons including DoS.
Many servers at least don't take all of their clients with them when they go down. That's because a client-server system can be written such as to be tolerant of a certain amoount of network trouble.
The SOB linux client is not amongst them. And this is not any DOSers fault, but yours and yours alone: you know that there are problems, we have already told you what they are and where they are and what to look for. But you decided that you will not let us help you and that you cannot fix things by yourself. At that point you are soley responsible for whatever quagmire you're in.
Neither do any of us users that you are going out of your way to ignore.
None of us make any money off of this project.
You are getting teraflops of computer time donated from hundreds and thousands of people all around the planet.
None of us receive any donations or grants from anyone for this project.
Who have been offering to donate brainpower as well. But you're too blase to accept that donation. And consequently you find yourself strapped for this exact resource: brain power.
But we others have nothing better to do than spend thousands of manhours hand-holding your client and restarting it every time there's a couple sunspots? You gallantly throw away thousands of people's time because you can't be bothered to invest some of yours? Heck, because you're too proud to let any one of use fix things FOR you?
We all have our own real life things that take priority over this project.
Now imagine what it could have accomplished if only you took advantage of the resources that you are being offered all the time and which you're snubbing.
This project has accomplished a lot in the time it's been up,
Don't have an answer yet, but I submitted a two page request. Hope to hear back soon.Originally posted by Alien88
That would be awesome.. let me know, and if I dont respond here you can hit me up off the board at mgarrison@alienz.net