+1 here 
just want to restart miner when 1 GPU fail. no reboot. windows.
I think its a bug in the program. If I use 1.3.0 watchdog works 1.3.1 & 1.3.2 watchdog doesn’t work. It shows the message but does not execute the script. Not sure about 1.4.0 as not tested that yet.
how it looks like action
The argument to “–watchdog-cmd” is the name of a script you want to run. There is a sample in the optiminer install directory called “watchdog-cmd.sh” (who would have guessed :). In the watchdog-cmd.sh script there are several examples of actions you can take when the watchdog is triggered. You have to remove the “#” in front of the commands you want to execute. If your GPU is hanging in a way that is requiring you to manually power-cycle the rig, then the last set of commands are the ones you want to use. If not, then the reboot command should suffice. Note: you must be running as “root” user for the watchdog commands to run successfully. The syntax for the command line is “–watchdog-cmd ./watchdog-cmd.sh”, which requires that you started optiminer from the install directory, i.e. you “cd /where_optiminer_is_installed” before you started the miner. The watchdog-timeout is simply the number of seconds that the GPU has gone without producing a share before running the watchdog-cmd. If you make the timeout shorter than the longest time a GPU may go without finding a share, which will vary depending on the pool (or solo/network) difficulty, then you will trigger the watchdog command when there’s nothing wrong. HTH.
so if I want to use it for my case i have to add --watchdog-cmd exit, optiminer.exe if i am using windows and want to close and run miner after gpu hangs
No, you would need a .bat file that runs “exit”, and you have to include the full path to the .bat file, or put it in the same directory you started optiminer from. I don’t use Windows, so I don’t know what “exit” does - if it just stops the current program from running, then it will just “exit” the .bat script, which accomplishes nothing.
I’m not running as root. I set a init.d script which continuously runs as root at startup to check for a file called reboot in /tmp.
My watchdog-cmd.sh then just does echo 1>/tmp/reboot.
I also echo a "rebooting… " to the screen as well.
However in v.1.3.0 everything works as it should.
In v.1.3.1 and v1.3.2 the script never gets called. The last message displayed is executing watchdog-cmd.sh
my echo doesn’t print, and it don’t reboot. If i copy the v1.3.0 optiminer executable, and the bin folder to the same directory where my mine.sh and watchdog-cmd.sh are, everything works as it should.
Perhaps optiminer was doing some debugging and commented out the call to execute the script or spawn another process…
Please share your hash rates with Optiminer 1.4, if you don’t mind:
- Card
- If custom BIOS a link to the BIOS
- RAM on Card
- Average Sols/S
- Wattage (please take wall and divide by number of cards if you can)
I’m curious to understand the relationship of total costs to Sols/S and including Sols/watt.
If I can get enough data I will share a table here with the results. So far post people are sharing one or two bits of data, but not the whole thing from which I can work out the ROIs for these cards. (This seems better than me buying one of each and testing, though I guess I can do that.)
And will it work?
–watchdog-timeout 60 --watchdog-cmd reboot.bat
reboot.bat create
We ask such parameters?
:begin
@echo off
echo restart: %date% %time% >> logs/log_restart.log
taskkill /f /im Optiminer.exe
start flypool-zec.bat
goto exit
Create a folder for logs
Thank you
Nope, just one:
$ ps auxww |grep optiminer
root 1193 6.7 0.4 17457832 99996 ? Sl 10:28 30:33 /home/zcash/optiminer-zcash/optiminer-zcash -s us1-zcash.flypool.org:3333 -c 0 -u
Hash rates with Optiminer 1.4:
Card - Saphire R9 290x
RAM on Card - 8GB
Average Sols/S - 305
Wattage - N/A
Whats the maximum intensity i could use on windows please? And where to add the line?
-i 8
Example
optiminer -s %POOL% -u %USER% -p %PASSWORD% -m 8000 -i 8
1 Like
Is 8 the max? Also is there anyway of making the command line coloured like claymore?
I tried -i 13 using 7514 MB from my MSI R9 390…Although I didn’t see a great improvement. After a while I see that the average S/s goes down a bit… More intensity equal more Memory usage. However with Claymore I never pass 3000 MB. Also I am not sure the fee percentage with Optiminer. Moreover I cannot use this miner in antpool, doesn’t connect, so I’m staying with Claymore with known fees…
Regards
[img]//cdck-file-uploads-global.s3.dualstack.us-west-2.amazonaws.com/zcash/original/2X/1/164a85c28676beaa4e99ccf748d717594bcf613d.png[/img]
2 Likes
Thank you that was very informative. Iv noticed jumps in sol/s from your screenshot? Does anyone else get this?
All that is put above -i 8
He writes error
mem object allocation failure
8 gb memory system
390 8gb
drivers 15.12
How to solve a problem?
linux rx470 and rx480 hashrates:
windows rx470 rx480 r9-295x2 hashrates:
I’m using Win10 64bits and latest drivers (Crimson Relive 16.12.2)
Regards.
From your experience do you think we are getting close to the limit in hashing for the RX 480’s or is there margin for more increases?
If yes what % do you think approx?
Thanks