Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 1086. (Read 3426947 times)

Greetings,
New prospector here and I first wanted to say my thanks for this program.  Get up to 230khash on mt GTX570.   As this is my first mining venture I have a lot to learn and I had a few questions that have been unanswered, including the aforementioned parse error, hoping you may be able to help

1.  I often see periods where I receive a high number of Stratum Timed out/Interrupted messages follow by Stratum detected new block. I did download hte mining proxy in the same folder, by my command doesnt invoke the it specifically (i really am not sure how yet), so I assume i am using the built in stratum proxy? Would it be better to set up git hubs mining proxy? Is it possible coinotron just has a shitty connection?

2. Anyone know how i can turn the intensity down? Do I just lower the kernal modifier (i.e from -l 30x8 to 20x4?)
 
3. I have also been getting that parse error where it just hangs. It has been pretty rare tho and it seems the longer it ran, the better chance of it happening.  Almost all mine were after 24hrs+ mining.  As such, I just restart my miner 1-2 times a day and its only come back once (compared to multiple times) in a 4 day period.  But I wanted to confirm this is indeed an application error and not something with the pool or my setup.

Thanks for any assistance!

Last night the program crashed on me.  This is my Event Viewer Info:
Fault bucket 3761457397, type 1
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: cudaminer.exe
P2: 0.0.0.0
P3: 5203bd81
P4: cudaminer.exe
P5: 0.0.0.0
P6: 5203bd81
P7: c0000005
P8: 000068b0
P9:
P10:

Attached files:
C:\Users\User\AppData\Local\Temp\WER97D8.tmp.WERInternalMetadata.xml

These files may be available here:
C:\Users\User\AppData\Local\Microsoft\Windows\WER\ReportArchive\AppCrash_cudaminer.exe_ca7818cba3ee8567ffd7297c0a1e4b135349e3_54fe2284

Analysis symbol:
Rechecking for solution: 0
Report Id: 37d3f330-03fc-11e3-9df9-00044b052d16
Report Status: 0

that sounds right lol
all i know is it was freezing on that line and after i changed it now at least we get a message saying there was an error (Jit debugger prompt i believe)
so getting a message that a problem occurred is better than not finding out until you wake up the next day and see you miner froze 6 hrs ago or something

honestly i haven't been looking to hard at this.. i looked quick and just tried the first thing that came to mind
not expecting to even see the error anyway since it happens so rarely and
strangely i got the same error like 3 times in about 20/30 minutes roughly and not since (maybe 8 hrs ago)
and before that it was a week or two.. bit hard to debug if it acts like that.

thanks for the input on the issue and if anyone else wants to chime in great

With the first if you are testing the address that s is pointing to, not the content.
If s is not pointing to anything then strcmp will probably cause a crash, depending on the implementation of strcmp.

just letting people know my uploaded build did not help the bug issue
i just got the error and rather than it freezing with no warning at least this time a box came up asking me if i wanna debug etc.
BUT i was not running the debug build i was running the release mode version i uploaded so it didn't tell me any useful
when i attached the debugger and stopped on the breakpoint. (code 5 memory access violation on the line i changed)
which should not have been a problem i think because all i did was modified that code to do the same thing but, in a safer way.

so instead of

i wrote

note the "s" variable is a char pointer
and we should never use math based logic on strings, it won't perform as intended sometimes.
so i explicitly changed to check if the string "null" with the strcmp run time function.
and it either is or it isn't why it triggered an error i don't know.
I suppose the error is self explanatory lol

i'll look into it for us Nvidia people though
at least with this build i uploaded though it seems we will get a warning an error occurred
which is better than no warning and having the app just freeze permanently..

so far so good, for me.. ever since i uploaded that new build i have not had a freeze up
but since i don't watch the console, i don't know if the error has come up though.

Trying it now.  My AV is hating it though. Constant warnings.

i just need to know if that error comes up AND your miner stops working and just freezes with the build i uploaded.
if it didn't make any difference i can try something else.. like i said problem for me is very rare so it's hard to track down / replicate etc.

eventually i will release a new (Windows only) miner with a GUI and a shit load of options.. i have got a lot done so far and i think its cool
It is also not another front end, it's a new program with pooler's and CB's source integrated.
It already has a console and / or log file support.. while running you can toggle either on the fly and many other options are live to..
meaning that you go to the options menu and make the change it takes effect with out restarting for example there is logging groups.
It has a Tabbed and / or TreeView interface which ever people prefer lol
I think i got a LOT of great ideas for features i wish the program had while i have been using it myself so i will make em
It'll be open source and I'm calling it "WinCudaMiner" and i will create a proper topic for it at some point so i don't keep bringing it up on this one lol

I am using WeMineLTC using their US3 server.

This is the error I get when it fails.  Been happening daily now. Almost always between 2am and 4am.

I have a GT 525M and it does 12kh/s to 15kh/s there abouts ~ oddly enough a couple of days ago I was using Skype and CudaMiner jumped to a blistering 17kh/s to 19kh/s (and has been staying at around 15kh/s to 18kh/s since.

I've found I can still use MinerD so I'm now around a mind boggling 30kh/s  (Gotta start somewhere right?  ?

works on Debian with Cuda toolkit 5.0 

geforce GTS 250 gives 33-34 kh/s without overclocking 

Ah ok, good to know. I didn't bother checking into that since I still had the older version all ready to go.

I'm guessing it's not a problem in CPU miner though. I've been using it, via ScryptMiner GUI, side-by-side with cudaMiner and it doesn't have the problem. Before anyone says, "But ScryptMiner GUI doesn't support stratum", I know, I "fixed" that.

That is basically what I did. But you don't need to go back to the older version to use getwork+stratum_proxy. The new version has it, too.

I have reported the problem to the CPU miner thread. cudaMiner  gets the stratum code from cpuminer.
https://bitcointalksearch.org/topic/m.2754826
There has been no reply. Maybe more peoeple should report to get attention.

Same here. I ended up switching back to the older version and using the stratum proxy. Every time the pool had connection problems, cudaMiner would successfully reconnect x number of times, but eventually it would lock up and sit there.

I'm getting the same issue as well, almost daily now.

I have had issues now with the Stratum.  It says something about unable to parse the new line and then it just sits there.  I have lost about 14 hours of mining due to this issue happening at 3am and I don't wake up till 10am.

Here's my command line when i used my laptop with GTX 670M.
cudaminer.exe -a scrypt -l 28x4 -i 0 -C 0 -m 1 -t 1 -q
I stopped mining with my laptop when it started randomly freeze and i had to force shutdown. Temps were about 80C. I think the overheating caused the freezing.

seems you fucked up the makefile:

Update video drivers...