Bitcoin Forum
May 25, 2024, 02:38:54 AM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 2 3 4 5 6 7 [8] 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 »
141  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 02:49:33 PM
Upd2: x64 folder is empty Undecided I can give you non-root access to Centos 5.4 x64 if you wish. PM me your SSH2 public key.
Downloading the x64 DVD right now via BitTorrent, should be able to get a virtual box of this up within the day. Should get a x64 compile then for bitcoind when it's all done, I'm curious myself to see how it runs.
142  Economy / Marketplace / Re: Pizza for bitcoins? on: August 04, 2010, 02:33:01 PM
lol, I doubt the offer is still 10,000!
$600 is a great deal for pizza I thought?  Grin
143  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 02:27:57 PM
I did notice it works with CentOS 5.4, so I fired up an Amazon EC2 instance running CentOS 5.4 on 16 cores and loaded it up. After it downloaded blocks, it starting churning away on 16 cores, but only managed about 8,800 khash/s, kind of disappointing for an EC2 instance.
1) It? I couldn't run your x86 build on 5.4 x64.

2) How much per month for that instance? EC2 is a very expensive provider for a very big sites that can afford paying good money for good infrastructure. By "infrastructure" I mean their elastic computing API, load balancing, storage and content delivery services. My current record with cheap VPS is 3100 khashes/sec for $5/mo.


It should run, I was able to run it on their (cc1.x4xlarge, 23GB) CentOS Cluster even though it's not a x64 build, the x86 still worked for me. I didn't even need to install the full bitcoin, I just uploaded the bitcoind and ran it right from root  Grin

It was very expensive, $1.60/hour per "Cluster Compute Quadruple Extra Large" basically.

I only ran a 1 hour instance just to see how it performed (too expensive to be practical) and I was checking the debug.log file to see what it was recording for khash/s.

I wonder if a x64 build would run faster on that cluster than the x86 build?  Huh
144  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 02:23:58 PM
I've compiled an x86 version with the 0.3.8 source, can you try using that one? If you already have it, the only thing I can think of is have it start from the beginning by deleting the blocks and let it start fresh in case something was corrupted in the beginning.

Yep, that's the one I downloaded and am running as of last night. That has the same problems for my system as my build one so I'm guessing it's my env. Trying a full cleanup and re-downloaded right now.


Update: It still fails to download correctly. Time for a fresh re-image of this box. Oh well.
Well, you could cheat I guess and copy the full blocks from another machine if you have one (Linux box) to see if it hangs or if it can continue from that point.

That is really strange though, if anything the CentOS build downloads blocks faster than anything I've seen yet. I can download the entire block chain in a few minutes with this build.
145  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 01:59:48 PM
From what I can tell on your output, it seems to be hung at that block for some reason. When you stop the daemon and restart it, does it continue on then?

If you want, you can try a direct connect mine with ./bitcoind -server -connect=12.53.130.10 to see if it's a build thing.

If I stop and restart the daemon, it will continue for the next 500 blocks. At that point it will once again lock up, even though the program isn't crashing a thread must be.

Just connected to your node and it did the same thing. Pulls down the next 500 blocks really fast and then locks up. Thoughts?
I've compiled an x86 version with the 0.3.8 source, can you try using that one? If you already have it, the only thing I can think of is have it start from the beginning by deleting the blocks and let it start fresh in case something was corrupted in the beginning.
146  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 01:44:20 PM
You'd rather publish this link. It's much more useful as you happen to have x64 builds too. My x64 build was outdated

http://knightmb.dyndns.org/files/bitcoin/custom%20builds/

Upd: Yeah it works on my x86 box! You definitely deserve a donation once that box generates its first coins.
Upd2: x64 folder is empty Undecided I can give you non-root access to Centos 5.4 x64 if you wish. PM me your SSH2 public key.
Yeah, been moving around folders to try and organize things. I have about a dozen virtual box now of many different distro so as I get time, I'll compile other releases and label them properly. I don't have a x64 build yet because I haven't built a x64 virtual box with Centos 5.5 yet  Grin

I did notice it works with CentOS 5.4, so I fired up an Amazon EC2 instance running CentOS 5.4 on 16 cores and loaded it up. After it downloaded blocks, it starting churning away on 16 cores, but only managed about 8,800 khash/s, kind of disappointing for an EC2 instance.
147  Bitcoin / Project Development / Re: Bounty for Bitcoin Animated Movie [4550 BTC and growing] on: August 04, 2010, 05:23:23 AM
I'll pledge 500 BTC for a good animated movie that has a laymen explanation of how things work, maybe a sprinkle of technical stuff about why it's safe/secure/etc.

It can't be too technical, people's eyes will glaze over if it starts talking about SHA256 and p2p block chains.
148  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 03:15:46 AM
I've been running mine since I made the first post about where to download it.

Everything seems to be fine and the bitcoind is using 100% doing coin generation at the lowest prioirty from when I checked top
Code:
knightmb@serverfarm528 32]$ ./bitcoind getinfo
{
    "balance" : 0.000000000000000,
    "blocks" : 72169,
    "connections" : 329,
    "proxy" : "",
    "generate" : true,
    "genproclimit" : -1,
    "difficulty" : 244.2132230923753
}

From what I can tell on your output, it seems to be hung at that block for some reason. When you stop the daemon and restart it, does it continue on then?

If you want, you can try a direct connect mine with ./bitcoind -server -connect=12.53.130.10 to see if it's a build thing.
149  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 04, 2010, 12:25:42 AM
There are two versions, one built from stock code, the other modified to accept up to 1,000 nodes (hence the super node name)
I'd rather you didn't make a build of the 1000 node connecting version available.  It won't take very many people running that before we have to make another release just to limit the incoming connections.
I appreciate your concern; the current release version already limits connections to 8, this one won't establish hundreds of connections to a single client or node, it's just modified to allow hundreds of connections back to it from independent clients. I figured since I have the CPU and bandwidth horsepower, why not put it to good use for Bitcoin?
150  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 11:46:46 PM
I've posted the binary up here (32 bit only - bitcoind):
Click Here

There are two versions, one built from stock code, the other modified to accept up to 1,000 nodes (hence the super node name)

Will probably tackle the GUI client build another day, this one wore me out. Donate some BTC to my sig if you are feeling generous  Wink
151  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 11:17:01 PM
Oh, but there's more!  Shocked

Even after I did that, I couldn't get it to compile due to linking errors. I was able to get a solid compile of bitcoind though, this is the extra steps it took.

After compiling and installing the BerkeleyDB4.7, you need to add it's include path to the makefile.unix

Add -I"/usr/local/BerkeleyDB.4.7/include"

Then you need to copy all the libraries from "/usr/local/BerkeleyDB.4.7/lib" to "/usr/lib"
BUT WAIT, there's MORE!  Huh

Yes, and if that was not annoying enough, now it will fail to compile because of an OpenSSL error, Grrrrr..

Now you must copy libcrypto.a and libssl.a from "/usr/local/ssl/lib" to "/usr/lib"

Yes, those can be linked, but I got tired of chasing down compiler errors, was easier to just copy them.  Roll Eyes

After all of that, then you should be able to compile a bitcoind binary.

I'm testing it now, seems to be working just fine. Downloading blocks and such. When it finishes, I'll experiment with some coin generation and coin transfer.

I'm still caught on the GUI client compile, currently hangs at
Code:
/usr/bin/ld: cannot find -lSM
If anyone is good with Google, let me know.  Wink
152  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 10:11:05 PM
Ok, conquered the DB issue after pestering an Oracle developer all day  Tongue

When you compile DB 4.7.25, you have to use an extra config command or else the compile will be missing those vital header files (why???)

Anyway, here is what it was
Code:
../dist/configure --enable-cxx
make
When you compile the source, this fixes the DB compiling related errors.
153  Bitcoin / Project Development / Re: BitCoin Wikipedia page DELETED!!! on: August 03, 2010, 07:55:42 PM
Somebody (possibly knightmb?) should request Userfication of the deleted article, to preserve the content.
I just saw that myself, I haven't had time to dig through all the links to see what the process is since the Administrators don't have a direct "contact me" button anywhere.  Sad
154  Bitcoin / Bitcoin Discussion / Re: What happens when network is split for prolonged time and reconnected? on: August 03, 2010, 07:02:13 PM
Transactions won't disappear if they're valid.  They'll just move to the longer block chain.

Invalid transactions would be somebody trying to double-spend across the split chains (which would be tricky-- you'd have to run a modified client, or copy your wallet to a machine working on the other block chain).

Or if the split lasted long enough (more than 100 blocks), transactions that involve generated coins on the shorter chain would be invalid at the merge.

For shorter splits, immature generated coins on the shorter chain will disappear when the chains merge, but that would be about the worst consequences for honest users (unless you were unlucky enough to get an invalid coin from somebody trying to cheat).

Interesting info, so other than some double-spending issues, as long as the block chain isn't separated for more than 100 or so blocks (or 16+ hours), it should be able to merge back to together without too many issues. So coin generation would probably be the only victim (if no double-spending cheats took place), or more technically, coin generation would be assigned to someone else that was generating at the same time as those that split off.

If there a hard coded limit on split delay? Meaning if I had a small network split from the public network, spent some coin around, came back a few days later and got them sync up to the public network (other than coin generation if it happened) transactions should be fine?
155  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 06:53:37 PM
I have successfully built it with 4.8, 4.7 never would but with 4.8 bitcoind locks up whenever it dumps the initial block download to disk. Undecided
I'm probably going to load up a virtual machine of CentOS 5.5 and work with that. I've really barfed up the server I was experimenting on trying make it build the binary, LOL.
156  Bitcoin / Development & Technical Discussion / Re: Builds for Ubuntu? on: August 03, 2010, 06:52:29 PM
I've built the client and daemon for Ubuntu 9.04 (should work in later versions ?)
Not a package per say, but you can just download the current Linux release and just overwrite the binaries. Not perfect, but it will get a working version for you that way.

Yeah I've built the binaries before, I want to be able to upgrade via the update manager rather than manually downloading the binaries.
Oh you mean official binaries in package form hosted by the Ubuntu folks?
157  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 05:59:12 PM
Just for kicks, tried to compile with DB4.8 got the same error, LOL (the line number was slightly different, but at the same class)

Didn't get this on other Distros, might be a CentOS related error, but not sure why.
158  Bitcoin / Development & Technical Discussion / Re: Builds for Ubuntu? on: August 03, 2010, 05:54:46 PM
Has anyone built a package for the GUI client yet?
I've built the client and daemon for Ubuntu 9.04 (should work in later versions ?)
Not a package per say, but you can just download the current Linux release and just overwrite the binaries. Not perfect, but it will get a working version for you that way.
159  Bitcoin / Bitcoin Discussion / Re: Which Country You're From on: August 03, 2010, 05:41:07 PM
Why not break it down to continent?  Grin
160  Bitcoin / Development & Technical Discussion / Re: Bitcoind x86 binary for CentOS on: August 03, 2010, 05:32:06 PM
{snip} And its build requires more than 256 mb ram. Satoshi improved bitcoind so doesn't require wxWidgets any more, but I'm not sure whether that patch is in 0.3.7 or whatever sources you are building.
Yeah, the bitcoind build gets around any errors for those, but I still get stuck at the DB error.  Sad
Pages: « 1 2 3 4 5 6 7 [8] 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!