CGWatcher 1.4.0, a GUI/monitor for CGMiner & BFGMiner to help minimize downtime

[Tigggger]

This doesn't cause me any problems but thought I would point it out, as I always end up with duplicate pools

All my settings are in a scrypt.conf file as they are always the same (no pools listed), I then use the arguments option to set the pools for each coin profile so for example for the middlecoin profile I have

Code:

-o stratum+tcp://middlecoin.com:3333 -u x -p x -o stratum+tcp://pool1.eu.multipool.in:7777 -u x -p x -o stratum+tcp://pool1.us.multipool.in:7777 -u x -p x

But when I go to the pools tab it always shows as follows with 2 entries for each.

[milone]

Dreaded_Source:

It's ok if they're blank, that just means they are using the default paths (which is inside the CGWatcher folder). The Desktop issue seems to come up frequently. I've added notifications in CGWatcher if it finds itself inside the Program Files, Program Files (x86), Windows, etc. folders to warn the user of potential permission issues. Perhaps I need to add the Desktop folder to that list.

I didn't see anything that stood out in the information you sent other than it appeared that cgminer closed immediately after starting several times, which could indicate bad config settings... though this could possibly be caused by Desktop-related permission issues where CGWatcher is unable to write to batch or config files when it needs to before starting the miner.

Let me know if moving it from the Desktop folder changes anything.


Tigggger:

Thanks, I'll look into this. You're not setting any pools in the config file, correct? It's probably just a matter of CGWatcher applying the arguments to the config file twice for some reason, which then adds the pools twice. While it shouldn't hurt anything, it isn't the expected behavior so I'll correct this in the next update.

[Tigggger]

Tigggger:

Thanks, I'll look into this. You're not setting any pools in the config file, correct? It's probably just a matter of CGWatcher applying the arguments to the config file twice for some reason, which then adds the pools twice. While it shouldn't hurt anything, it isn't the expected behavior so I'll correct this in the next update.

Correct, just from the arguments box.  Like I said everything runs fine and it's never caused an issue and it's always done this through various versions, just thought you'd like to know.

[Nordstern]

Today a donation mining pool appeard in the pool list randomly while not chosen.
The log shows:

[2013-09-20 10:56:44]  CGMiner (6748): Pool 3 (@http://ltc.kattare.com:9332) status is DEAD
[2013-09-20 10:56:47]  CGMiner (6748): Pool 3 (frog81.cgwatcher@http://ltc.kattare.com:9332) status is ALIVE
[2013-09-20 10:56:47]  CGMiner (6748): Current pool is Pool 3 (CGWatcher Donation Mining Pool)

Donation mining was NOT chosen @ this time! That's bad and misuse of my trust! So im out ...

[milone]

The code for donation mining has not changed since it was first implemented. Two nights ago ltc.kattare.com (Burnside's pool) went down (http://www.reddit.com/r/litecoinmining/comments/1mojul/burnsides_down_i_cant_connect_to_their_website/) for the first time in quite some time which means a backup donation pool is added (probably the first time this has ever been needed) and that this (ltc.kattare.com) pool was probably not removed correctly.

It is not that the pool was added while donation mining was disabled... it is that it was not removed correctly. And there are several checks in place to ensure this happens, but perhaps I need to set it to restart the miner if it fails to remove it. Or maybe there is additional code I need to write in CGWatcher. In the end CGWatcher can tell the miner to remove the pool but if it doesn't, the only last option is to restart the miner while it looks like I'll be adding.

Donation mining has been available for several months and you're the first person to ever report a problem with it, so please watch before you start making accusations. If you don't want to use the program that's fine... but it should be clear by now I'm not attempting to scam anyone. If I was going to have CGWatcher switch to a pool without your consent, don't you think I would skip writing it to the log?

Anyway, thanks for reporting this so I can look into it and correct it.

Edit: After checking burnside's pool, I do not see any significant increase to the hashrate (of my own GPUs) over the past 24 hours other than 1-2 hour bumps, which happens frequently. Nor is there anything exceptional on coinotron (the backup pool) which has only mined 1 LTC up to this point. But if you feel that you were donation mining without your consent, please send me your log and litecoin address and I'll send you a litecoin to make up for it, which is 10-20x the value you would have mined for me in that hour (or if you feel it was more than that, I'll make sure you are paid back 10x whatever you mined.)

[networkgeek24]

I'm having a problem on 1 of my 6 litecoin rigs. At some point one of my cards gets sick so cgwatcher tries to restart cgminer. Something about this process doesn't go quite right and the program waits a while and then tries to start cgminer again and again and again. By the time I see what is going on and remote into the PC there are many cgminer screens filling up the desktop. After 10 seconds or so they all go away except one, which then successfully starts mining.

I have the monitoring setup so that cgwatcher restarts cgminer in the event of a sick card, accepted shares not changing after 10 mins, hashrate below a certain threshold, and if api access can't be obtained.

What other details do I need to post to help troubleshoot this? I'm running the latest version.

[networkgeek24]

Some log exerpts for previous post:


[9/20/2013 3:21:00 PM]   [d] CGMiner running and API initialized on port 4028, process matched to profile Default.
[9/20/2013 3:21:00 PM]   [d] Finish wait for miner to start, process is 2352 (alive).
[9/20/2013 3:21:00 PM]       CGMiner started successfully.
[9/20/2013 3:51:19 PM]       The computer is now idle (30 min, 7 sec since last activity, 30 min requirement).
[9/20/2013 4:49:07 PM]       GPU2 (AMD Radeon HD 7900 Series) is SICK! Attempting to restart CGMiner...
[9/20/2013 4:49:07 PM]   [d] Miner.Restart begin.
[9/20/2013 4:49:07 PM]       Restart CGMiner was requested, preparing to stop then start the miner to make sure its configuration settings are reloaded...
[9/20/2013 4:49:07 PM]   [d] Miner.PerformStopStartRestart called. (Profile=Default;Miner=C:\Go\cgminer\cgminer.exe;HasFullAPIAccess=True;Initiator=Monitor;Reason=#NAME# was successfully restarted.)
[9/20/2013 4:49:07 PM]   [d] Miner.StopMining called: Miner=C:\Go\cgminer\cgminer.exe;ProcessID=2352 (True);BatchProcessID=0 (False);Config=C:\Go\CGWatcher\temp.conf;Arguments=--config C:\Go\cgminer\cgminer.conf;HasFullAPIAccess=True;KillOnReadOnlyAccess=True;IsBatCmdFile=False;ProcessID=2352;.
[9/20/2013 4:49:25 PM]       CGMiner failed to close completely within 15 seconds of quit command. Attempting to kill...
[9/20/2013 4:49:25 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:26 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:27 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:28 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:29 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:30 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:31 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:32 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:33 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:34 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:35 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:37 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:38 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:39 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:40 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:41 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:42 PM]       CGMiner failed to close completely after being sent the quit command, then failed to be killed. Giving up.
[9/20/2013 4:49:42 PM]       CGMiner restart unsuccessful because miner failed to stop and close completely. Ensure the miner is closed in order to resume mining.
[9/20/2013 4:49:42 PM]   [d] Miner.Restart end.
[9/20/2013 4:49:52 PM]       The miner is closed outside of CGWatcher and you've requested for me to ensure it stays running. Attempting to start...
[9/20/2013 4:49:52 PM]   [d] Miner.IsProcessAlive(2352) returns True. [First try, process has exited. First try, process has exited.] (CPID0/CMPID=0/CBPID=0)
[9/20/2013 4:49:52 PM]   [d] Miner.Kill Exception: Access is denied (KillProcessID=2352)
[9/20/2013 4:49:52 PM]   [d] Miner.StartMining called. [Profile=Default;MinerFile=C:\Go\cgminer\cgminer.exe;MinerFileUsed=C:\Go\cgminer\cgminer.exe;MinerExecutable=C:\Go\cgminer\cgminer.exe;Config=C:\Go\CGWatcher\temp.conf (C:\Go\cgminer\cgminer.conf);Arguments=--config C:\Go\CGWatcher\temp.conf;Anticipated Port=4028;Used Port=4028;IsBatCmd=False;BatCmdPath=C:\Go\cgminer\cgminer.exe (C:\Go\cgminer\cgminer.exe);LastKnownInstanceCount=0/0]
[9/20/2013 4:49:53 PM]   [d] Miner.IsProcessAlive(152) returns True. [First try, process is alive.] (CPID152/CMPID=152/CBPID=0)
[9/20/2013 4:49:53 PM]   [d] Waiting for CGMiner to initialize API. Process 152 alive=True, minerid=152.
[9/20/2013 4:49:54 PM]   [d] Watch for miner, expected process is 152 (alive).
[9/20/2013 4:49:54 PM]   [d] CGMiner process is running using ProcessID 152, expecting port 4028.
[9/20/2013 4:50:53 PM]   [d] Finish wait for miner to start, process is 0 (not alive).
[9/20/2013 4:50:53 PM]   [d] CGMiner process 0 not alive at Miner.StartMining exit.
[9/20/2013 4:50:53 PM]       Start miner failed on 1st attempt. Killing any remaining processes and trying again.
[9/20/2013 4:50:53 PM]   [d] Miner.IsProcessAlive(152) returns True. [First try, process is alive.] (CPID0/CMPID=0/CBPID=0)
[9/20/2013 4:50:54 PM]   [d] 1 miner instances found, 1 ignored.
[9/20/2013 4:50:56 PM]   [d] Process 152 ignored: Unable to match to a profile or communicate with on port 4028.
[9/20/2013 4:50:56 PM]   [d] Miner.StartMining called. [Profile=Default;MinerFile=C:\Go\cgminer\cgminer.exe;MinerFileUsed=C:\Go\cgminer\cgminer.exe;MinerExecutable=C:\Go\cgminer\cgminer.exe;Config=C:\Go\CGWatcher\temp.conf (C:\Go\cgminer\cgminer.conf);Arguments=--config C:\Go\CGWatcher\temp.conf;Anticipated Port=4028;Used Port=4028;IsBatCmd=False;BatCmdPath=C:\Go\cgminer\cgminer.exe (C:\Go\cgminer\cgminer.exe);LastKnownInstanceCount=0/1]
[9/20/2013 4:50:57 PM]   [d] Miner.IsProcessAlive(1984) returns True. [First try, process is alive.] (CPID1984/CMPID=1984/CBPID=0)
[9/20/2013 4:50:57 PM]   [d] Waiting for CGMiner to initialize API. Process 1984 alive=True, minerid=1984.
[9/20/2013 4:50:58 PM]   [d] Watch for miner, expected process is 1984 (alive).
[9/20/2013 4:50:58 PM]   [d] CGMiner process is running using ProcessID 1984, expecting port 4028.
[9/20/2013 4:51:57 PM]   [d] Finish wait for miner to start, process is 0 (not alive).
[9/20/2013 4:51:57 PM]   [d] CGMiner process 0 not alive at Miner.StartMining exit.
[9/20/2013 4:51:57 PM]       Start miner failed on 2nd attempt. Killing any remaining processes and trying again.
[9/20/2013 4:51:57 PM]   [d] Miner.IsProcessAlive(1984) returns True. [First try, process is alive.] (CPID0/CMPID=0/CBPID=0)
[9/20/2013 4:51:58 PM]   [d] 2 miner instances found, 2 ignored.
[9/20/2013 4:52:00 PM]       1 running miner(s) detected but I am unable to communicate with any of them!
[9/20/2013 4:52:00 PM]   [d] Miner.StartMining called. [Profile=Default;MinerFile=C:\Go\cgminer\cgminer.exe;MinerFileUsed=C:\Go\cgminer\cgminer.exe;MinerExecutable=C:\Go\cgminer\cgminer.exe;Config=C:\Go\CGWatcher\temp.conf (C:\Go\cgminer\cgminer.conf);Arguments=--config C:\Go\CGWatcher\temp.conf;Anticipated Port=4028;Used Port=4028;IsBatCmd=False;BatCmdPath=C:\Go\cgminer\cgminer.exe (C:\Go\cgminer\cgminer.exe);LastKnownInstanceCount=0/2]
[9/20/2013 4:52:01 PM]   [d] Miner.IsProcessAlive(2724) returns True. [First try, process is alive.] (CPID2724/CMPID=2724/CBPID=0)
[9/20/2013 4:52:01 PM]   [d] Waiting for CGMiner to initialize API. Process 2724 alive=True, minerid=2724.
[9/20/2013 4:52:02 PM]   [d] Watch for miner, expected process is 2724 (alive).
[9/20/2013 4:52:02 PM]   [d] CGMiner process is running using ProcessID 2724, expecting port 4028.
[9/20/2013 4:53:01 PM]   [d] Finish wait for miner to start, process is 0 (not alive).
[9/20/2013 4:53:01 PM]   [d] CGMiner process 0 not alive at Miner.StartMining exit.
[9/20/2013 4:53:01 PM]       Start miner failed on 3rd attempt. Killing any remaining processes.
[9/20/2013 4:53:01 PM]   [d] Miner.IsProcessAlive(2724) returns True. [First try, process is alive.] (CPID0/CMPID=0/CBPID=0)
[9/20/2013 4:53:13 PM]   [d] 3 miner instances found, 3 ignored.



Also, I'm right clicking cgwatcher and running as administrator every time.

[milone]

Can you close any open cgminer windows and CGWatcher, then re-open CGWatcher (so a clean start). Try starting the miner and when it fails to connect to it, go to the Tests tab in CGWatcher and click 'Create Debug Report'. Then email that report to my email in readme.txt. It will hopefully provide some additional information to see if anything is going wrong in CGWatcher, or if this is a Windows privilege issue (which has happened before, but should still be fixable).

Thanks,
Justin

[networkgeek24]

Can you close any open cgminer windows and CGWatcher, then re-open CGWatcher (so a clean start). Try starting the miner and when it fails to connect to it, go to the Tests tab in CGWatcher and click 'Create Debug Report'. Then email that report to my email in readme.txt. It will hopefully provide some additional information to see if anything is going wrong in CGWatcher, or if this is a Windows privilege issue (which has happened before, but should still be fixable).

Thanks,
Justin

Everytime I discover this has happened I reboot the rig and start it all fresh. I start up cgwatcher as admin and it runs happily for a day or two. I then get notification that a rig is down and when I'm able to remote in is when I see all the cgminer windows open. 10 seconds remoting in all the cgminer windows go away and things start working properly again. I'm not sure I can go to the tests tab and click a create debug report while the issue is happening.

I imaged 4 rigs with the same OS image and set them all up the same, yet I'm only having this problem on one.

[milone]

Creating a debug report doesn't have to be done while it is happening, but it only includes the last 25 log entries so the sooner, the better. You can also attach cgwatcher.log to the email so then I could see all log entries that way. The debug report also shows information about the last time the miner was started (or attempted), so creating it after a failed start/restart attempt is important to collect that data.

This doesn't seem to be a CGWatcher problem nor a CGMiner problem, but I can't say that for certain without more information. I'm assuming it's a driver crash that is causing the sick GPU, then Windows is taking longer than usual to recover. If you can provide as much info as possible I'll try to help you get it resolved. Windows Event Viewer may also show relevant errors and warnings that occurred around or after the time of the driver crash/sick GPU. It could be failing hardware in that computer (like a failing hdd or bad GPU) if you're using the same image on other computers without this problem.

Justin

[Lauda]

Interesting

[milone]

New in version 1.3.2

• Desktop folder added to protected directories list to notify user that CGWatcher, miner, and config files should not be stored there.
• Advanced Profile options to set CPU affinity and priority for miner process. Additional options to come in future versions.
• Profile matching relaxed during miner launch to reduce risk of miner being ignored (rare).
• Fixed ads only reloading after restarting CGWatcher.
• Button added to Tests tab to list all installed .NET updates for troubleshooting purposes.
• Option to restart computer if CGWatcher is unable to close or kill miner processes. Although rare, this indicates a more serious problem (Task Manager is also unable to kill the processes) and usually leads to a BSOD (crash). The only way to resolve this is to restart the computer.
• Pools truncated from miner's API response (~Pool 35+) will get info from config to avoid chinese-looking characters being shown for url/username/password. No stats are available for these pools though because the data is cutoff from the miner's response, which has a fixed maximum length.
• Fixed pools set in arguments being added twice to temp config file before miner starts.
• Added scheduled action to send specified API command(s) to miner.
• Settings tab added to Coin Manager, 'Remove coins no longer on CoinChoose' option moved to Settings tab of Coin Manager.
• Coin profitability data refresh interval setting now in Settings tab of Coin Manager.
• Setting created to base coin profitability on bitcoin or litecoin, located in Settings tab of Coin Manager.
• New coin notification option moved to Settings tab of Coin Manager.
• Scheduled action frequency added - 'When event occurs...' allowing you to select from a list of events that will trigger the action being run. Events currently consist of miner events, profile events, and coin profitability events. More events will be added in future updates.
• Create your own profitability formula in a custom coin field that can be used when creating scheduled actions that switch profile based on profitability. Instead of using an existing field (profitability, difficulty, etc) you can create your own mathematical expression using all existing fields and mathematical functions.
• Fixed CGWatcher not trying to start the miner indefinitely when the 'Keep trying indefinitely' failure option was selected. (It will try up to 2147843647 times.)
• Miner process not added to checked process list until it has been running for 60+ seconds to prevent incorrectly ignoring it.
• Added 'Send email' scheduled action. You can specify an email address for each action, the last used will be filled in automatically. Emails are currently limited to 25 per computer per day, but this may increase or decrease over time depending on usage. Counter is reset at midnight EST/EDT (U.S. Eastern). Emails will be coming from @cgwatcher.com, and you will need to ensure you can access http://minerremote.com for email to work correctly.
• More scheduled action events will be added, with additional options like 'when hashrate drops below/%', 'when miner restart fails X consecutive times, etc. I figured I'd add them in groups instead of trying to do it all at once.
• If CGWatcher is set to try starting the miner indefinitely (and it keeps failing), it will wait one second per 10 attempts in between attempts over 10, up to 60 seconds. So after 600 attempts it will wait one minute between each attempt.
• CGWatcher restarting GPUs that cgminer has disabled due to overheat no longer requires CGWatcher's overheat protection to be enabled. It will do this for all GPUs since the miner usually fails at re-enabling them.
• Before restarting computer, CGWatcher will temporarily set itself to start with Windows and launch miner at startup if these options are not enabled. It will reset the options back to their original settings the next time CGWatcher is started.
• Restart computer prompt changed to always use CGWatcher's prompt only, which provides a cancel option, rather than the Windows notification that the computer is restarting. This was already how scheduled computer restarts were handled, but is now done for all computer restarts.
• Profitability-based scheduled actions (switch profile based on profitability) will update coin data before selecting a profile unless it had been updated within the past minute instead of within the past 5 minutes.
• Added average time per share to Monitor tab to help in setting appropriate number of minutes without share increase for this monitoring option.
• Added "% of Avg" to Hashrate Cutoff monitoring option, which will restart the miner if the current hashrate drops below the specified percentage of the current average hashrate. This is in addition to being able to set actual hashrate values.
• Elapsed mining time added to Stats tab.
• In Pools tab, pool drop-down will default to current pool and revert to current pool if user has not selected a different pool to view within the last 5 minutes.
• Added support for pool quota option in CGMiner 3.4.3+
• Added config file and argument options up through CGMiner 3.5.0 and BFGMiner 3.2.1.
• Select a different value to display on the Status tab in place of Efficiency, or create your own value using existing values and mathematical functions to create a custom expression.

[superresistant]

Hello,

How to make CGWatcher work with a KNC miner ?

Thx

[milone]

I'm still awaiting more info from KNC owners as they get their machines to get information regarding the best way to manage them with CGWatcher. CGRemote will work with anything running cgminer or bfgminer (I believe KNC uses cgminer). CGWatcher, as it is written now, is very dependent on the miner process to be running on the same computer as this gives it the most control over monitoring and resolving problems. While bfgminer may be able to act as a proxy for self-contained mining devices allowing CGWatcher to work, CGWatcher would be monitoring that bfgminer process and not the mining process running on the miner, which is not ideal. CGRemote will ultimately provide almost all of the functionality of CGWatcher, but can do it for multiple miners on any OS and in any location, all in one program.

If you're interested in trying CGWatcher with a KNC miner, you would setup bfgminer and CGWatcher on the same computer similar to that of a Block Erupter Blade (assuming they work in the same manner). See bfgminer's README.ASIC.txt (more info online, including this thread). You're basically pointing the mining device at the computer bfgminer (and CGWatcher) is running on. Set bfgminer's http-port to the same port the mining device is set to. The linked thread mentions it is only available for Win32 and Linux, so I would try it with the 32-bit version of bfgminer, not the 64-bit. Please report any problems/results so I can get them fixed. Once running, CGWatcher would then be watching the bfgminer process which would be displaying the device and stats it gets via http. I'm curious as to how API commands work when sent to bfgminer in this case, particularly privileged commands. Make sure to start bfgminer with the mining device's IP address in the api-allow W: group.

I'll also be making changes to CGWatcher and CGRemote that will allow changing profiles without restarting the miner. This feature has been requested even before ASICs as some people prefer to restart the miner as least often as possible. It will work any time the profile is switched to another profile using the same algorithm (sha256->sha256 or scrypt->scrypt). If any problems occur during the switch, it will restart the miner as a failsafe (what it currently does when switching profiles). Aside from changing pools, it will change any other config settings that can be set via API. If too many config settings are changed that cannot be set via API, then the miner can be restarted instead.

Right now I'm back to working on CGRemote's next update, so if anything is needed to add support or functionality in these miners I'll be adding it as quickly as possible. Any information or feedback regarding KNC miners is appreciated.

[superresistant]

Interesting. Thanks for the explanation.

[milone]

New in version 1.3.3 (most significant changes)

• CoinWarz added as optional coin data provider (API key required, see http://coinwarz.com/v1/api for more info)
• Do not show CGRemote ads if user is using CGRemote.
• Drop-down list of coins in Coin Manager -> Custom Field -> Test sorted.
• Coin algorithm added to custom field value expression variables. Returns 1 for sha256, 2 for scrypt.
• On Status tab, Efficiency replaced with custom stat. You can leave it at Efficiency, select a different statistic, or create your own statistic.
• Fixed "when highest/lowest custom field value changes" event description not displaying correctly in scheduled actions list.
• Added "Device elapsed" to GPU and USB devices as added in miner API 1.32.
• Unrecognized remote commands will be tested by being sent to miner (if running) rather than just reporting that the command is unrecognized. This will allow new miner API commands to be used manually without necessarily requiring a CGWatcher update.
• Fixed "Manage Pools" button opening temporary config file instead of actual profile config file.
• Fixed causing "Switch Profile" in CGRemote to not start miner using new profile.
• Add "log" and "report" commands for CGRemote to get log and report data.
• Add "New Block average (seconds)" and "Found Block average (seconds)" values for custom profitability formula field and custom value expression.
• Add "Utility" and "Work Utility" values for custom profitability formula field.
• Current and average hashrates added to window title bar.
• Fixed icarus-timing option incorrectly mirroring usb option in Config File Editor.
• Tests tab reorganized, additional debug reports and utilities moved to drop-down list.
• 'Clear all Windows autorun entries' in Tests tab (drop-down) will remove all CGWatcher entries from Windows startup as a way to get a clean slate. It also disables the "Run when Windows starts" setting, which can be re-enabled afterward.
• Chart on Status tab now shows mining devices only, including FPGA/ASIC devices that report temperature.
• Setting created to ensure CGWatcher & miner are relaunched automatically after CGWatcher-initiated computer restart.
• "Restart CGWatcher" scheduled action created, which may help in situations where CGWatcher stops responding or crashes after running for X consecutive days on low memory computers. This has only been reported once, and I often check for memory leaks, but this is a simple alternative (temporary) solution.
• "Restart CGWatcher" command added for CGRemote.
• Miner auto-launch delay setting added to UI (Monitor -> Other -> Continuous Mining)
• Displayed hashrate formatting improved for consistency.
• CGRemote command to have CGWatcher automatically check and update itself added.
• Other minor fixes and improvements.

[carly200]

Hi,

first of all: GREAT TOOL !!

I have a question regarding the settings/restart after the display dirver crashes

unfortunateley when my displaydriver crasshes, cgminer shows low hashrate in 5s average, but not in the "current hashrate", that is stuck at the last value.
so it seems that cgwatcher does not detect that.

i need cgwatcher to detect if clockspeed is below a certain value for a certain amount of time and then only a complete reboot helps.
any way to set this up?

example, driver crashed, clocks are at minumum, but cgminer displays/reports false hashrate...

[milone]

The "Current Hashrate" in CGWatcher is the "5s" hashrate in cgminer. The fact that they're showing something different means either CGWatcher hasn't refreshed the data yet to reflect recent changes or cgminer is reporting the wrong data via API (which shouldn't happen, although it will sometimes still report a normal hashrate for a sick or dead GPU). It's odd that none of the info in cgminer matches CGWatcher... how long after the crash was this screenshot taken? From my experience, shortly after a display driver crash at least one GPU will become sick or dead, which is why there is a monitor option to watch for this. If this isn't the case all of the time, then I'll create an additional option to watch for a sudden drop in clock speed.

[carly200]

the problem is, that if the driver crashes, cgminer will report a wrong hashrate  (the last one from before the crash!)

i also noticed, that running cgwatcher decreases my hashrate by 15-20 khashes. can somebody verify?

[milone]

I realize the hashrate may be wrong after a driver crash, but as I mentioned from all of my experiences, the driver crash causes at least one GPU to get a 'SICK' or 'DEAD' status, which CGWatcher checks for if this monitor option is enabled. It may take a minute after the crash for cgminer to change the status of a GPU to sick or dead. If that is not the case and there are times where this does not happen (particularly with Windows XP because of changes to display driver architecture (WDDM) introduced in Windows Vista), I can add an option to watch the clock speeds.

CGWatcher should not negatively affect hashing performance, and the couple times this has come up it was always a setting that was being set differently. The most common being where you're mining a scrypt coin and you see a difference in hashrates with/without CGWatcher, it is likely due to an issue with cgminer where it uses one gpu thread even though the readme states the default is two.

So if you run cgminer without CGWatcher and you don't explicitly set gpu-threads, it will run with one gpu thread.

If you start the miner with CGWatcher, it uses two as the default since that is what cgminer's readme states. In this case you have to set gpu-threads to one to get the same results as you would get from running cgminer alone.

You can see the exact arguments/config settings CGWatcher used the last time it launched the miner by going to the Tests tab, go to 'Debug Reports & Utilities' section and select 'Show last miner start configuration' from the drop-down list and click the 'Run' button.