Phoenix - Efficient, fast, modular miner

[check]

Just to follow up, "1.48 debug" seemed to improve things at first for a 6950 under Win7, but I'm still not having much luck with it.  It doesn't seem to actually get stuck, but it spends most of its time idling.  It seems to eventually recover, but then it will idle again after a few minutes.

phoenix-1.48-debug>phoenix -u http://user:pass@bitcoinpool.com:8334/ -v -q 2 -k poclbm DEVICE=0 AGGRESSION=10 BFI_INT PLATFORM=1 WORKSIZE=128
...
[24/05/2011 10:42:46] Warning: work queue empty, miner is idle
[24/05/2011 11:00:13] LP: New work pushed
[24/05/2011 11:00:13] Server gave new work; passing to WorkQueue
[24/05/2011 11:00:13] New block (WorkQueue)
[24/05/2011 11:00:16] Result 00000000f651082d... accepted
[24/05/2011 11:00:18] Result 000000002db4051c... accepted
[24/05/2011 11:00:25] Warning: work queue empty, miner is idle
[24/05/2011 11:03:00] LP: New work pushed
[24/05/2011 11:03:00] Server gave new work; passing to WorkQueue
[24/05/2011 11:03:00] New block (WorkQueue)
[24/05/2011 11:03:12] Warning: work queue empty, miner is idle
[0 Khash/sec] [360 Accepted] [6 Rejected] [RPC (+LP)]

Killing it and starting fresh always seems to do well for a while, which makes me think it's not solely a pool problem.

[1714196110]

1714196110

[1714196110]

1714196110

[1714196110]

1714196110

[1714196110]

1714196110

[1714196110]

1714196110

[jedi95]

I'm working on an extensive debug build that will dump out a highly detailed logfile every time the miner goes idle. Hopefully this will enable me to find the cause of this bug.

[bitcoinington]

Hoping someone can help me here because i'm seeing some strange behavior in phoenix miner

when i'm mining on a pool i get a good hash rate and my accepted/rejected ratio is around 2000/40.

however, when i'm mining solo (using bitcoind as server and http://user:pass@localhost:8332 as the server) it shows that it is connected to the server and the hash rate is 430mh/s but it consistently stays at 0 Accepted and 0 Rejected.

anyone have an idea of what's going on? here's the commandline i'm using

-u http://user:pass@localhost:8332 -k poclbm device=0 vectors bfi_int aggression=7

any help would be greatly appreciated!

[error]

Hoping someone can help me here because i'm seeing some strange behavior in phoenix miner

when i'm mining on a pool i get a good hash rate and my accepted/rejected ratio is around 2000/40.

however, when i'm mining solo (using bitcoind as server and http://user:pass@localhost:8332 as the server) it shows that it is connected to the server and the hash rate is 430mh/s but it consistently stays at 0 Accepted and 0 Rejected.

anyone have an idea of what's going on? here's the commandline i'm using

-u http://user:pass@localhost:8332 -k poclbm device=0 vectors bfi_int aggression=7

any help would be greatly appreciated!

If you mine solo, you won't see anything accepted until you find a full 50BTC block.

[twmz]

however, when i'm mining solo (using bitcoind as server and http://user:pass@localhost:8332 as the server) it shows that it is connected to the server and the hash rate is 430mh/s but it consistently stays at 0 Accepted and 0 Rejected.

That's normal.  When you are in a pool, the pool "lies" to your mining client and tells it to calculate work at an artificially low difficulty.  As a result, you find what you think are valid blocks pretty frequently.  Pools call these shares.  The pool does this so that it can more effectively tell that you are doing real work on a regular basis.  Every once in a while these one of these shares will end up being a valid block for the "real" difficultly and that is when the pool ends up finding a block.

When you mine solo, you are now mining with the "real" difficulty directly.  You'll only get an "accepted" message when you find a real block (and earn 50 BTC).  This will be much much less often than you are used to as compared to finding accepted shares on a pool (days or weeks between blocks depending on your hash rate).

[dishwara]

however, when i'm mining solo (using bitcoind as server and http://user:pass@localhost:8332 as the server) it shows that it is connected to the server and the hash rate is 430mh/s but it consistently stays at 0 Accepted and 0 Rejected.

That's normal.  When you are in a pool, the pool "lies" to your mining client and tells it to calculate work at an artificially low difficulty.  As a result, you find what you think are valid blocks pretty frequently.  Pools call these shares.  The pool does this so that it can more effectively tell that you are doing real work on a regular basis.  Every once in a while these one of these shares will end up being a valid block for the "real" difficultly and that is when the pool ends up finding a block.

When you mine solo, you are now mining with the "real" difficulty directly.  You'll only get an "accepted" message when you find a real block (and earn 50 BTC).  This will be much much less often than you are used to as compared to finding accepted shares on a pool (days or weeks between blocks depending on your hash rate).

Thanks, this solved my doubt, why i got zero only even though my gpu usage is 100%.

[trebo]

I'm working on an extensive debug build that will dump out a highly detailed logfile every time the miner goes idle. Hopefully this will enable me to find the cause of this bug.

jedi95, I have a patch that logs to syslog on linux machines, can I send you the patches?

[gentakin]

Thanks for this miner, it works pretty good for me with phatk (at least if I can trust the hash rate shown in the phoenix client)!

However, there is one problem (that I already posted in deepbits thread, knowing it's probably a phoenix problem):
Sometimes long polling stops working until I restart phoenix. Example:

[25/05/2011 10:49:38] Result: cea614c1 accepted           
[25/05/2011 10:49:42] Result: cc38d5ab accepted           
[25/05/2011 10:49:55] Result: 26e10d54 accepted           
[25/05/2011 10:50:08] Result: 5dd94d57 rejected           
[25/05/2011 10:50:18] Result: 3a240e80 rejected           
[25/05/2011 10:51:05] Result: 029664f3 accepted           
[25/05/2011 10:51:31] Result: 58578139 accepted           
[25/05/2011 10:51:58] Result: f7aa2b93 accepted           
[25/05/2011 10:52:12] Result: 8cffaddf accepted           
[25/05/2011 10:52:37] Result: 1f244382 accepted           
[25/05/2011 10:53:13] Result: d8f362a7 accepted           
[25/05/2011 10:53:22] Result: 9ea37ad4 rejected           
[25/05/2011 10:53:29] Result: af350f97 rejected           
[25/05/2011 10:53:44] Result: 3ebf2662 accepted           
[25/05/2011 10:54:13] Result: 2ef96a00 accepted           
[25/05/2011 10:54:14] Result: e40a052a accepted           
[25/05/2011 10:54:14] Result: 4c64a417 accepted           
[25/05/2011 10:54:15] Result: 6af612e9 accepted

And so on: No LP messages for a (very) long time, instead lots of (compared to usual amounts) rejected shares.

Restarting phoenix will fix it:

[25/05/2011 11:23:05] Phoenix r86 starting...
[25/05/2011 11:23:05] Connected to server
[25/05/2011 11:23:22] Result: da42db08 accepted         
[25/05/2011 11:23:43] LP: New work pushed               
[25/05/2011 11:23:53] Result: 08a0d468 accepted         
[25/05/2011 11:24:00] Result: ceca5dec accepted         
[25/05/2011 11:24:07] Result: b7c71390 accepted         
[25/05/2011 11:24:22] Result: 8c7814d1 accepted         
[25/05/2011 11:24:28] Result: 160c836f accepted         
[25/05/2011 11:24:54] Result: 39784e84 accepted         
[25/05/2011 11:25:18] Result: d4b22132 accepted         
[25/05/2011 11:25:24] Result: adf76db7 accepted         
[25/05/2011 11:25:30] Result: 4df9dc3c accepted         
[25/05/2011 11:25:43] Result: 00d196f2 accepted         
[25/05/2011 11:25:49] Result: 0aa99500 accepted         
[25/05/2011 11:26:06] Result: 29d79d72 accepted         
[25/05/2011 11:26:41] Result: 3a370dc4 accepted         
[25/05/2011 11:26:45] Result: 90e4bceb accepted         
[25/05/2011 11:27:14] Result: 9aa7a90a accepted         
[25/05/2011 11:27:17] Result: 15d32279 accepted         
[25/05/2011 11:27:20] LP: New work pushed               
[25/05/2011 11:27:52] Result: 140f2597 accepted         
[25/05/2011 11:27:55] Result: 1736843d accepted         
[25/05/2011 11:28:22] LP: New work pushed

It might be related to my rather unstable network connection (wireless lan) which might cause the LP connection to get lost without phoenix re-connecting when my network is back up again.
I'm using the latest revision from SVN (r99 if I remember correctly, although phoenix says it's r86 when starting..)

[error]

That seems to be it. If the network connection is interrupted, phoenix doesn't seem to reestablish the long polling connection. That would explain everything (except for it stopping completely, which is still an outstanding issue).

[nelisky]

That seems to be it. If the network connection is interrupted, phoenix doesn't seem to reestablish the long polling connection. That would explain everything (except for it stopping completely, which is still an outstanding issue).

Unlike what I mentioned previously, even without LP the hang still happens. Looking a bit closer at the code it seems to me that the getWork request that WorkQueue performs when the miner is found to be idle fails silently and is not retried, so phoenix sits waiting for something to happen but with no action deferred.

[drgr33n]

I don't seem to have a problem ?

LinuxCoin 0.2a
AMD-SDK 2.4
fglrx 11.4

1 x 5870 @ 998/300
1 x 5830 @ 998/300

Phoenix has been ticking along for days now. I do get the odd queue is idle but its due to my sketchy mobile broadband. With the phatk kernel I'm getting around 425MHs on my 5870 and 300MHs from my 5830.

Great work !!

[commlinx]

That seems to be it. If the network connection is interrupted, phoenix doesn't seem to reestablish the long polling connection. That would explain everything (except for it stopping completely, which is still an outstanding issue).

Unlike what I mentioned previously, even without LP the hang still happens. Looking a bit closer at the code it seems to me that the getWork request that WorkQueue performs when the miner is found to be idle fails silently and is not retried, so phoenix sits waiting for something to happen but with no action deferred.

I've only had it happen once, but wanted to +1 that because it happened to me on slush's pool that doesn't support LP a few days ago when he had some problems. One instance was going fine, but other was idle. I run under Win7 64-bit with SDK 2.4. My Internet is normally pretty reliable which might explain why I only had a problem after the pool being down a while.

[perryh]

I got phoenix to work for a short while, but now it no longer shows my video card in the device list.

aticonfig and flgrxinfo both display my card

dsasda@dsay-home:~/phoenix/phoenix-1.48$ ./phoenix.py -u http://dsadssad:blah@btcguild.com:8332/ DEVICE=1 BFI_INT VECTORS AGGRESSION=12 -k phatk WORKSIZE=12
No device specified or device not found, use DEVICE=ID to specify one of the following

    

• Intel(R) Core(TM)2 Extreme CPU Q6800  @ 2.93GHz

[0 Khash/sec] [0 Accepted] [0 Rejected]perryh@perry-home:~/phoenix/phoenix-1.48$

any ideas?

thanks

edit: fixed. just defined $DEVICE

[disq]

BTW, increasing queue size cant help those ppl having "miner is idle"? i think it worth a try.

I think this should do the hash/s log(export) thing.

I gona try it once i figured out how to compile something in Python hahaha.

Here is shivansps's logtotext modification, working diff (against phoenix 1.47, should not change anything with .48 tho)

Edit: If running on a flash drive, you should use /tmp for destination. Like so: "./phoenix.py PARAMS -t /tmp/my-awesome-ati-9990.txt"

Code:

diff -ur phoenix-1.47/ConsoleLogger.py phoenix-1.47-modified/ConsoleLogger.py
--- phoenix-1.47/ConsoleLogger.py	2011-05-13 06:43:53.000000000 +0300
+++ phoenix-1.47-modified/ConsoleLogger.py	2011-05-26 20:27:40.000000000 +0300
@@ -47,9 +47,10 @@
     
     UPDATE_TIME = 1.0
     
-    def __init__(self, verbose=False):
+    def __init__(self, verbose=False, logtotext=None):
         self.verbose = verbose
         self.lastUpdate = time() - 1
+        self.logtotext = logtotext
         self.rate = 0
         self.accepted = 0
         self.invalid = 0
@@ -119,6 +120,10 @@
                 "[" + str(self.accepted) + " Accepted] "
                 "[" + str(self.invalid) + " Rejected]" + type)
             self.say(status)
+            if(self.logtotext != None):
+                fileHandle = open (self.logtotext, 'w')
+                fileHandle.write(datetime.now().strftime(self.TIME_FORMAT) + ' ' + status)
+                fileHandle.close()
             self.lastUpdate = time()
         
     def say(self, message, newLine=False, hideTimestamp=False):
Binary files phoenix-1.47/ConsoleLogger.pyc and phoenix-1.47-modified/ConsoleLogger.pyc differ
Binary files phoenix-1.47/kernels/poclbm/167c418e01275326142dd6e832cab098.elf and phoenix-1.47-modified/kernels/poclbm/167c418e01275326142dd6e832cab098.elf differ
diff -ur phoenix-1.47/phoenix.py phoenix-1.47-modified/phoenix.py
--- phoenix-1.47/phoenix.py	2011-05-13 06:43:53.000000000 +0300
+++ phoenix-1.47-modified/phoenix.py	2011-05-26 20:26:39.000000000 +0300
@@ -44,6 +44,7 @@
         self.connection = None
         self.kernel = None
         self.queue = None
+        self.logtotext = None
         
         self.kernelOptions = {}
         
@@ -62,6 +63,7 @@
         parser.add_option("-a", "--avgsamples", dest="avgsamples", type="int",
             default=10,
             help="how many samples to use for hashrate average")
+        parser.add_option("-t", "--logtotext", dest="logtotext", default=None, help="Log to text")
         
         self.parsedSettings, args = parser.parse_args()
         
@@ -88,7 +90,7 @@
     
     def makeLogger(self, requester):
         if not self.logger:
-            self.logger = ConsoleLogger(self.parsedSettings.verbose)
+            self.logger = ConsoleLogger(self.parsedSettings.verbose, self.parsedSettings.logtotext)
         return self.logger
     
     def makeConnection(self, requester):

[jimraynor]

whenever i try to run phoenix miner it just closes its self a split second after i run it, what am i doing wrong?

[bobR]

whenever i try to run phoenix miner it just closes its self a split second after i run it, what am i doing wrong?

With out some details I'd say mining

[jimraynor]

definatly not mining

[bobR]

definatly not mining

or learning ...
if you cant say exactly what you did that did not work
how can anyone help you

smart ass reply's  just make you look like an ass
and you just lost my help
have a good life

[jimraynor]

i did not make a smart ass reply

i didnt exactly know what you ment but i thought you ment i might of been mining with other software so i said i definately wasnt

i just extracted the zip and ran phoenix.exe and a cmd window pops up then instantly closes

[twmz]

i just extracted the zip and ran phoenix.exe and a cmd window pops up then instantly closes

phoenix.exe can't be double clicked on.  It has required command line parameters.

Open a cmd window manually, change the the folder with phoenix.exe and run it from there with the required command line parameters.