Phoenix - Efficient, fast, modular miner

[bcforum]

Shouldn't more MHash/s mean more submitted shares?

If you look closely at the bottom chart, it appears the upper portion has a slightly steeper slope than the lower section. Plot a differential chart (x[n+1] - x[n]) and you will see the change more clearly.

[1714214320]

1714214320

[1714214320]

1714214320

[1714214320]

1714214320

[1714214320]

1714214320

[TeaRex]

Shouldn't more MHash/s mean more submitted shares?

Where do you get the mhash0 number from? try using the -a parameter of phoenix with a large averaging period and see if it really goes up in the longer term. Sometimes an overclocked card will develop heat problems and stop itself or clock itself down for short periods again and again. Local overheating at some point of the chip can happen even if the overall temperature of the GPU is well within a sane range.

[c00w]

Has anyone noticed a memory leak with phoenix? I seem to be getting roughly 27% of memory usage after two days of running it.

[Peao]

Has anyone noticed a memory leak with phoenix? I seem to be getting roughly 27% of memory usage after two days of running it.

+1

[bcforum]

Has anyone noticed a memory leak with phoenix? I seem to be getting roughly 27% of memory usage after two days of running it.

I did a quick experiment last night on Ubuntu.

% memstat | grep python | head -4
   9228k: PID  1747 (/usr/bin/python2.6)
 201956k: PID  2295 (/usr/bin/python2.6)
   9248k: PID  2311 (/usr/bin/python2.6)
 219828k: PID  2344 (/usr/bin/python2.6)

8 hours later:
% memstat | grep python | head -4
   9228k: PID  1747 (/usr/bin/python2.6)
 231140k: PID  2295 (/usr/bin/python2.6)
   9248k: PID  2311 (/usr/bin/python2.6)
 249004k: PID  2344 (/usr/bin/python2.6)

So yes, there is a definite memory leak.

[c00w]

If someone could load it up in pdb and figure out what datastructure is doing it, I could probably fix it.

[simonk83]

That'd be great c00w.  Hopefully that'll fix this as well:  http://forum.bitcoin.org/index.php?topic=29977.msg376803#msg376803

[jedi95]

If someone could load it up in pdb and figure out what datastructure is doing it, I could probably fix it.

I would start with the RPC code, including the modified Twisted.web library. I run my miners through MMP, and I usually run them for weeks at a time with >100,000 accepted shares between restarts. I don't see any performance problems or memory leaks running with MMP.

[c00w]

I think its the rpc code itself not the modified twisted persistent client. bitHopper uses almost identical code from the bugfix and shows no memory gain.

[cyberlync]

Perhaps a stupid question, but where can I get the r110 version? When I go to the SVN it shows r112
I am getting a whole lot of idles and restarts on btcguild with the newest version.

[dime]

Perhaps a stupid question, but where can I get the r110 version? When I go to the SVN it shows r112
I am getting a whole lot of idles and restarts on btcguild with the newest version.

svn checkout -r 110 http://svn3.xp-dev.com/svn/phoenix-miner/trunk

[dime]

The rewrite for the RPC protocol is finished, and has been uploaded to the SVN.

It works fine in my testing, but more feedback would be great before pushing this as a new version. This SHOULD definitively fix the idle problem, since that was related to the old RPCProtocol.

i tested it and it resulted in constant disconnects from btcguild (all servers). I reverted back to r110 and the problem disappeared.

+1

I was mining for 2 days straight with constant disconnects from btcguild (all servers) and bitcoins.lc for about 10-15% rejected. There were no disconnects from slush's pool though.

Reverted to revision r110 and now all is good with btcguild and bitcoins.lc.

[cyberlync]

Perhaps a stupid question, but where can I get the r110 version? When I go to the SVN it shows r112
I am getting a whole lot of idles and restarts on btcguild with the newest version.

svn checkout -r 110 http://svn3.xp-dev.com/svn/phoenix-miner/trunk

After trying the URL, getting an error in firefox, installing chrome, retrying the url in chrome, I finally googled it, and found out it's for linux, Im a winblows nab, and can't for the love of bitcoins, figure out how I get the r110. Please help me, Im getting balder by the minute.

[TeaRex]

After trying the URL, getting an error in firefox, installing chrome, retrying the url in chrome, I finally googled it, and found out it's for linux, Im a winblows nab, and can't for the love of bitcoins, figure out how I get the r110. Please help me, Im getting balder by the minute.

Install Tortoise SVN from http://tortoisesvn.net/downloads.html and use it to perform the checkout.

Before you do that, you should know that running any SVN version of Phoenix requires you to have installed Phython and a bunch of extra Phyton packages. If you can't or won't do that, stick to the offical released version.

[cyberlync]

After trying the URL, getting an error in firefox, installing chrome, retrying the url in chrome, I finally googled it, and found out it's for linux, Im a winblows nab, and can't for the love of bitcoins, figure out how I get the r110. Please help me, Im getting balder by the minute.

Install Tortoise SVN from http://tortoisesvn.net/downloads.html and use it to perform the checkout.

Before you do that, you should know that running any SVN version of Phoenix requires you to have installed Phython and a bunch of extra Phyton packages. If you can't or won't do that, stick to the offical released version.

Thank you!

[nomnomnom]

If someone could load it up in pdb and figure out what datastructure is doing it, I could probably fix it.

I would start with the RPC code, including the modified Twisted.web library. I run my miners through MMP, and I usually run them for weeks at a time with >100,000 accepted shares between restarts. I don't see any performance problems or memory leaks running with MMP.

Hi,

I think it could be related to longpolling. For example when I mine on slush (rpc only) it works fine for days, as soon as
I mine on a pool that uses longpolling it starts to fill the memory. It happens on arsbitcoin and btcguild in my case.

[bcforum]

If someone could load it up in pdb and figure out what datastructure is doing it, I could probably fix it.

I would start with the RPC code, including the modified Twisted.web library. I run my miners through MMP, and I usually run them for weeks at a time with >100,000 accepted shares between restarts. I don't see any performance problems or memory leaks running with MMP.

Hi,

I think it could be related to longpolling. For example when I mine on slush (rpc only) it works fine for days, as soon as
I mine on a pool that uses longpolling it starts to fill the memory. It happens on arsbitcoin and btcguild in my case.

I don't get the same result.  When I mine solo against the standard client (RPC only, no long polling) memory fills up.

[Odi]

I'm another one with disconnect problems, here is my experience on Fedora Core 5 with ActivePython 2.7.2.5 in the Ars pool (phoenix with phatk kernel):

svn checkout http://svn3.xp-dev.com/svn/phoenix-miner/trunk

r111 from yesterday, 30 accepted, 4 rejected (1 real reject, 3 rejects because of disconnect), 188MH/s

Code:

[03/08/2011 15:19:53] Phoenix r111 starting...
[03/08/2011 15:19:53] Connected to server
[03/08/2011 15:19:53] Server gave new work; passing to WorkQueue
[03/08/2011 15:19:53] New block (WorkQueue)
[03/08/2011 15:19:53] Server gave new work; passing to WorkQueue
[03/08/2011 15:19:54] Using new LP URL /LP
[03/08/2011 15:19:54] LP connected to arsbitcoin.com:XXXX
[03/08/2011 15:20:16] Server gave new work; passing to WorkQueue
[03/08/2011 15:20:28] Result 000000001c389660... accepted
[03/08/2011 15:20:29] Result 0000000076722dbe... accepted
[03/08/2011 15:20:32] Result 00000000fc3029f4... accepted
[03/08/2011 15:20:39] Server gave new work; passing to WorkQueue
[03/08/2011 15:20:49] Result 0000000009f0a1e5... accepted
[03/08/2011 15:20:53] Result 00000000f1fbf3eb... accepted
[03/08/2011 15:21:01] Server gave new work; passing to WorkQueue
[03/08/2011 15:21:02] Result 000000004169e08d... accepted
[03/08/2011 15:21:24] Server gave new work; passing to WorkQueue
[03/08/2011 15:21:41] Result 00000000c3d105dc... accepted
[03/08/2011 15:21:47] Server gave new work; passing to WorkQueue
[03/08/2011 15:22:05] Result 000000005f316441... accepted
[03/08/2011 15:22:10] Result 0000000019c401ca... accepted
[03/08/2011 15:22:10] Server gave new work; passing to WorkQueue
[03/08/2011 15:22:23] Result 00000000bedb83ed... accepted
[03/08/2011 15:22:33] Server gave new work; passing to WorkQueue
[03/08/2011 15:22:45] Result 00000000819e2c9d... accepted
[03/08/2011 15:22:53] Result 00000000814424af... accepted
[03/08/2011 15:22:55] Server gave new work; passing to WorkQueue
[03/08/2011 15:23:18] Server gave new work; passing to WorkQueue
[03/08/2011 15:23:22] Server gave new work; passing to WorkQueue
[03/08/2011 15:23:22] New block (WorkQueue)
[03/08/2011 15:23:22] LP: New work pushed
[03/08/2011 15:23:22] Result 0000000095bb2fe8... rejected
[03/08/2011 15:23:22] Server gave new work; passing to WorkQueue
[03/08/2011 15:23:36] Result 00000000127cc49d... accepted
[03/08/2011 15:23:42] Result 00000000621b950d... accepted
[03/08/2011 15:23:45] Server gave new work; passing to WorkQueue
[03/08/2011 15:23:51] Result 000000004af6c2ac... accepted
[03/08/2011 15:24:08] Server gave new work; passing to WorkQueue
[03/08/2011 15:24:14] Server gave new work; passing to WorkQueue
[03/08/2011 15:24:14] New block (WorkQueue)
[03/08/2011 15:24:14] LP: New work pushed
[03/08/2011 15:24:14] Server gave new work; passing to WorkQueue
[03/08/2011 15:24:36] Server gave new work; passing to WorkQueue
[03/08/2011 15:24:38] Result 000000008f0f15dc... accepted
[03/08/2011 15:24:57] Result 00000000aa324577... accepted
[03/08/2011 15:24:59] Server gave new work; passing to WorkQueue
[03/08/2011 15:25:03] Result 00000000a21f563c... accepted
[03/08/2011 15:25:22] Server gave new work; passing to WorkQueue
[03/08/2011 15:25:26] Result 000000008d60ee63... accepted
[03/08/2011 15:25:31] Result 000000004db5858f... accepted
[03/08/2011 15:25:36] Result 0000000009cd448f... accepted
[03/08/2011 15:25:45] Server gave new work; passing to WorkQueue
[03/08/2011 15:25:58] Result 0000000087439926... accepted
[03/08/2011 15:26:08] Server gave new work; passing to WorkQueue
[03/08/2011 15:26:31] Server gave new work; passing to WorkQueue
[03/08/2011 15:26:35] Server gave new work; passing to WorkQueue
[03/08/2011 15:26:35] New block (WorkQueue)
[03/08/2011 15:26:35] LP: New work pushed
[03/08/2011 15:26:35] Server gave new work; passing to WorkQueue
[03/08/2011 15:26:58] Server gave new work; passing to WorkQueue
[03/08/2011 15:27:11] Disconnected from server
[03/08/2011 15:27:11] Result 00000000c4abbd72... rejected
[03/08/2011 15:27:21] Connected to server
[03/08/2011 15:27:21] Server gave new work; passing to WorkQueue
[03/08/2011 15:27:43] Server gave new work; passing to WorkQueue
[03/08/2011 15:28:06] Server gave new work; passing to WorkQueue
[03/08/2011 15:28:10] Result 0000000034ca9854... accepted
[03/08/2011 15:28:15] Result 000000006989ee7d... accepted
[03/08/2011 15:28:17] Result 00000000bc931382... accepted
[03/08/2011 15:28:29] Server gave new work; passing to WorkQueue
[03/08/2011 15:28:33] Result 00000000a7a0c3ad... accepted
[03/08/2011 15:28:38] Result 000000001922e028... accepted
[03/08/2011 15:28:52] Server gave new work; passing to WorkQueue
[03/08/2011 15:29:15] Server gave new work; passing to WorkQueue
[03/08/2011 15:29:25] Result 000000003e1f7bee... accepted
[03/08/2011 15:29:37] Server gave new work; passing to WorkQueue
[03/08/2011 15:30:00] Server gave new work; passing to WorkQueue
[03/08/2011 15:30:23] Server gave new work; passing to WorkQueue
[03/08/2011 15:30:43] Disconnected from server
[03/08/2011 15:30:43] Result 000000001770ff49... rejected
[03/08/2011 15:30:46] Connected to server
[03/08/2011 15:30:46] Server gave new work; passing to WorkQueue
[03/08/2011 15:30:55] Result 00000000babf0a08... accepted
[03/08/2011 15:31:09] Server gave new work; passing to WorkQueue
[03/08/2011 15:31:31] Server gave new work; passing to WorkQueue
[03/08/2011 15:31:54] Server gave new work; passing to WorkQueue
[03/08/2011 15:32:08] Disconnected from server
[03/08/2011 15:32:08] Result 0000000086bde4b3... rejected
[03/08/2011 15:32:16] Connected to server
[03/08/2011 15:32:16] Result 000000009beed6f1... accepted
[03/08/2011 15:32:17] Server gave new work; passing to WorkQueue

svn checkout http://svn3.xp-dev.com/svn/phoenix-miner/tags/release-1.50

The 1.50 version that I've had good success with, and I modified the #define Ma, 31 accepted, no rejected or disconnects, 186MH/s

Code:

[03/08/2011 15:44:30] Phoenix 1.50 starting...
[03/08/2011 15:44:30] Connected to server
[03/08/2011 15:44:30] Server gave new work; passing to WorkQueue
[03/08/2011 15:44:30] New block (WorkQueue)
[03/08/2011 15:44:32] Server gave new work; passing to WorkQueue
[03/08/2011 15:44:54] Server gave new work; passing to WorkQueue
[03/08/2011 15:45:17] Server gave new work; passing to WorkQueue
[03/08/2011 15:45:17] Result 00000000fc9d82a9... accepted
[03/08/2011 15:45:18] Result 0000000063e2ff80... accepted
[03/08/2011 15:45:40] Server gave new work; passing to WorkQueue
[03/08/2011 15:45:49] LP: New work pushed
[03/08/2011 15:45:49] Server gave new work; passing to WorkQueue
[03/08/2011 15:45:49] New block (WorkQueue)
[03/08/2011 15:45:49] Server gave new work; passing to WorkQueue
[03/08/2011 15:46:12] Server gave new work; passing to WorkQueue
[03/08/2011 15:46:35] Server gave new work; passing to WorkQueue
[03/08/2011 15:46:59] Server gave new work; passing to WorkQueue
[03/08/2011 15:47:10] Result 00000000dfd9c5ce... accepted
[03/08/2011 15:47:22] Server gave new work; passing to WorkQueue
[03/08/2011 15:47:45] Server gave new work; passing to WorkQueue
[03/08/2011 15:48:08] Server gave new work; passing to WorkQueue
[03/08/2011 15:48:16] Result 00000000f66b0d25... accepted
[03/08/2011 15:48:25] Result 00000000a8dc29b2... accepted
[03/08/2011 15:48:31] Server gave new work; passing to WorkQueue
[03/08/2011 15:48:54] Server gave new work; passing to WorkQueue
[03/08/2011 15:48:59] Result 0000000095e655b0... accepted
[03/08/2011 15:49:08] Result 000000001559c7a1... accepted
[03/08/2011 15:49:17] Server gave new work; passing to WorkQueue
[03/08/2011 15:49:40] Server gave new work; passing to WorkQueue
[03/08/2011 15:50:03] Server gave new work; passing to WorkQueue
[03/08/2011 15:50:17] Result 0000000060ef8b17... accepted
[03/08/2011 15:50:26] Server gave new work; passing to WorkQueue
[03/08/2011 15:50:49] Server gave new work; passing to WorkQueue
[03/08/2011 15:50:50] Result 0000000031dae576... accepted
[03/08/2011 15:50:50] Result 00000000b3c72d2c... accepted
[03/08/2011 15:51:00] Result 000000001bc7ceab... accepted
[03/08/2011 15:51:12] Result 00000000557c9e6f... accepted
[03/08/2011 15:51:13] Server gave new work; passing to WorkQueue
[03/08/2011 15:51:35] Server gave new work; passing to WorkQueue
[03/08/2011 15:51:58] Server gave new work; passing to WorkQueue
[03/08/2011 15:52:22] Server gave new work; passing to WorkQueue
[03/08/2011 15:52:29] Result 000000008fdb22f3... accepted
[03/08/2011 15:52:39] Result 000000001205cc4a... accepted
[03/08/2011 15:52:45] Server gave new work; passing to WorkQueue
[03/08/2011 15:52:50] Result 00000000b1d01806... accepted
[03/08/2011 15:52:53] Result 000000005d550742... accepted
[03/08/2011 15:53:08] Server gave new work; passing to WorkQueue
[03/08/2011 15:53:23] Result 000000002ca8dcdc... accepted
[03/08/2011 15:53:25] Result 00000000d22cc6d8... accepted
[03/08/2011 15:53:31] Server gave new work; passing to WorkQueue
[03/08/2011 15:53:54] Server gave new work; passing to WorkQueue
[03/08/2011 15:54:17] Server gave new work; passing to WorkQueue
[03/08/2011 15:54:40] Server gave new work; passing to WorkQueue
[03/08/2011 15:54:49] Result 00000000e90515ef... accepted
[03/08/2011 15:54:52] Result 00000000429bbab9... accepted
[03/08/2011 15:55:04] Result 00000000efdb7449... accepted
[03/08/2011 15:55:04] Server gave new work; passing to WorkQueue
[03/08/2011 15:55:26] Server gave new work; passing to WorkQueue
[03/08/2011 15:55:33] Result 00000000d6231842... accepted
[03/08/2011 15:55:41] Result 0000000045d609a3... accepted
[03/08/2011 15:55:45] Result 000000006d1961d6... accepted
[03/08/2011 15:55:49] Server gave new work; passing to WorkQueue
[03/08/2011 15:55:49] Result 000000001de475e4... accepted
[03/08/2011 15:55:50] Result 00000000453ceef2... accepted
[03/08/2011 15:55:55] Result 00000000d98eebe8... accepted
[03/08/2011 15:56:06] LP: New work pushed
[03/08/2011 15:56:06] Server gave new work; passing to WorkQueue
[03/08/2011 15:56:06] New block (WorkQueue)
[03/08/2011 15:56:07] Server gave new work; passing to WorkQueue
[03/08/2011 15:56:30] Server gave new work; passing to WorkQueue
[03/08/2011 15:56:53] Server gave new work; passing to WorkQueue
[03/08/2011 15:56:53] Result 000000005d89ad5e... accepted
[03/08/2011 15:57:16] Server gave new work; passing to WorkQueue
[03/08/2011 15:57:39] Server gave new work; passing to WorkQueue
[03/08/2011 15:57:43] Result 00000000de4c7f98... accepted
[03/08/2011 15:57:44] Result 000000005706b190... accepted
[03/08/2011 15:57:55] Result 00000000e2b690fd... accepted

[nomnomnom]

I don't get the same result.  When I mine solo against the standard client (RPC only, no long polling) memory fills up.

Hmm, thats weird, look at my memory graph, most of it is slush and at the end is arsbitcoin


Maybe there is some other difference between those pools that triggers this, oh well...

[bcforum]

Very curious. I switched to:

Code:

#!/bin/sh
export DISPLAY=:0
while true; do
    timeout -k 60 24h python phoenix.py -v \
		-u http://user:pass@localhost:8332 \
		-k phatk_dia DEVICE=0 AGGRESSION=13 BFI_INT WORKSIZE=128 \
		VECTORS FASTLOOP=false
done

to force a restart every 24 hours, so I can't collect the data anymore.