Pages:
Author

Topic: CGWatcher 1.4.0, a GUI/monitor for CGMiner & BFGMiner to help minimize downtime - page 30. (Read 402530 times)

newbie
Activity: 29
Merit: 0
Quite odd, I don't know how I can help you, will have to wait for the dominus to reply. Yet I wonder, if you were to edit your watcher's profiles.dat file, using mine as a guide, saving and starting it, maybe it would work? Mine has two profiles, replaced first one with placeholders, second just wildmarks for sensitive. The file has 4 lines starting with profiles, 0, 1 and space. Huh

PROFILES=2
0||||||||||||
1|||[email protected] w/ failsafe|||C:\Users\***\Desktop\cgminer-3.1.1-windows\cgminer.exe||||||--scrypt -o stratum+tcp://eu.wemineltc.com:3333 -u *** -p *** -o stratum+tcp://eubackup.wemineltc.com:3333 -u *** -p *** -o stratum+tcp://stratum2.wemineltc.com:3333 -u *** -p *** --gpu-platform 1 -I 19 -w 256 --thread-concurrency 22400 --no-adl --api-listen --api-allow W:127.0.0.1
hero member
Activity: 644
Merit: 500
My bad, didn't notice (in due time). I see [5/20/2013 5:48:31 PM] No CGMiner instance found., but from what I can tell it's more than that, maybe the author can help you out, I'm just taking guesses. Tho, if you run cgminer externally, then wait a few and Refresh the Status tab, there's nothing? Or in devices?

L.E. To your edit, there isn't a default folder for cgminer to my knowledge... why it would be grayed out I don't know. Undecided If you "Manage Profiles", and create a new one, still grayed out?

As we know already I have two cards and two separate folders of cgminer while trying to use the -d dedicated miner argument with two different pools and conf files. Presently being launched by two different bat files. As to why is a long story.

However the section where you can assign the directory where to find the cgminer program is actually greyed out. I'm guessing because it does not see my cards so it does not even bother to continue with inputting that information. Very strange I made it down to the 2nd page of this thread and everyone seems to be using it with out a problem so I know its got to be with my computer or something.

Yet I have been mining for about 2 months now so I'm pretty sure I have the video cards- jokes lol

So Cgwatcher is basically a god sent imo.

 
Code:
CGWatcher information:
 CGWatcher start time............. 5/20/2013 6:32:56 PM
 CGWatcher version................ 1.1.4.1
 Active profile................... Default
 Current monitor start time....... 5/20/2013 6:32:56 PM
 Monitor duration................. 60 sec
 Checks performed................. 6
 Restarts performed............... 0
 Last restart reason..............


Last Mining Session:
 Miner............................ CGMiner
 Version..........................
 API version......................
 Arguments........................
 Config file......................
 Miner is running................. False
 Miner start time................. 12:00:00 AM
 Miner stop time.................. N/A
 Mining duration.................. 0 sec
 Current total hashrate........... 0 Kh/s
 Current average hashrate......... 0 Kh/s
 Accepted shares.................. 0
 Rejected shares.................. 0
 Stale shares..................... 0
 Discarded work................... 0
 Hash Method...................... SHA256
 Current Block Time............... 0
 Current Block Hash...............
 Current Block Difficulty......... 0
 LongPoll......................... False


Profile Information:
 Active Profile................... Default
 Config file......................
 Arguments........................ --api-listen --api-allow W:127.0.0.1
 Real arguments................... --api-listen --api-allow W:127.0.0.1


Network statistics: (all pools)
 Times Sent....................... 0
 Bytes Sent....................... 0
 Times Received................... 0
 Bytes Received................... 0
 Net Bytes Sent................... 0
 Net Bytes Received............... 0

This information was last updated: 5/20/2013 6:33:57 PM
-----------------------------------------------------------------------------------------------------------------------

Debug Information:
 Operating system................. Microsoft Windows 7 Ultimate  6.1.7601.65536
 GPUs............................. 0
 Pools............................ 0
 Miner Process ID................. 0
 Miner startup grace period....... 3 min
 Has API access................... False
 Had API access to process........ False
 Has full API access.............. False
 Had full API access to process... False
 User last stopped miner.......... False
 Last (a/r/s) share change........ 5/20/2013 6:32:56 PM (60 sec ago)
 Last accepted share change....... 5/20/2013 6:32:56 PM (60 sec ago)
 Last rejected share change....... 5/20/2013 6:32:56 PM (60 sec ago)
 Last stale share change.......... 5/20/2013 6:32:56 PM (60 sec ago)
 Last discarded share change...... 5/20/2013 6:32:56 PM (60 sec ago)
 Monitor enabled.................. True
 Monitor interval................. 10 sec
 Restart Hashrate enabled......... False
 Restart Hashrate cutoff.......... 0 Mh/s
 Restart Every X Hours enabled.... False
 Restart Every X Hours hours...... 24 hour(s)
 Restart Share Totals enabled..... False
 Restart Share Totals minutes..... 10 minute(s)
 Restart Sick/Dead GPUs enabled... False
 Restart API Access Loss enabled.. False
 Ensure miner is running enabled.. False
 Schedule Enabled................. False
 Schedule Start Time.............. 6:32:56 PM
 Schedule Stop After hours........ 8 hour(s)
 Installed UI culture............. en-US
 Culture decimal separator........ (.)
 Culture group separator.......... (,)
 Max available log entries........ 500
 Version Response.................
 Privileged Response..............
 Config Response..................
 Coin Response....................
 Summary Response.................


Last 10 Log entries:
[5/20/2013 6:33:47 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:47 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 6:33:57 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.

newbie
Activity: 29
Merit: 0
My bad, didn't notice (in due time). I see [5/20/2013 5:48:31 PM] No CGMiner instance found., but from what I can tell it's more than that, maybe the author can help you out, I'm just taking guesses. Tho, if you run cgminer externally, then wait a few and Refresh the Status tab, there's nothing? Or in devices?

L.E. To your edit, there isn't a default folder for cgminer to my knowledge... why it would be grayed out I don't know. Undecided If you "Manage Profiles", and create a new one, still grayed out?
hero member
Activity: 644
Merit: 500
Great program. I'm just having trouble having it see my cards it shows I don't have any. If the problem has been discussed about just post me a link and I'll read it through. I just need tom direction here.

Does CGMiner see your cards properly? Or in what sense do you mean it doesn't show any?

I went back and edited my post by adding the error messages however to answer your question its like I don't have any.

It basically says sorry you don't have cards that I can find on your computer.

6770
7950
Curious: Is there an appropriate folder location that cgminer should reside?
The place where you can specify the folder is greyed out. Not able to accept folder residence of the miner executable and the config file.
newbie
Activity: 29
Merit: 0
Great program. I'm just having trouble having it see my cards it shows I don't have any. If the problem has been discussed about just post me a link and I'll read it through. I just need tom direction here.

Does CGMiner see your cards properly? Or in what sense do you mean it doesn't show any?
hero member
Activity: 644
Merit: 500
Great program. I'm just having trouble having it see my cards it shows I don't have any. If the problem has been discussed. Please just post me a link and I'll read it through. I just need some direction here.

Could this help?
Code:
5/20/2013 5:48:20 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:20 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:20 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:20 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:20 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] -- Begin CGWatcher v1.1.4.1 Process -------------------------------------------------------
[5/20/2013 5:48:31 PM] Active profile is "Default"
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] No CGMiner instance found.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:31 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:32 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:32 PM] [debug] Miner.GetProcessID (B) Exception: Object reference not set to an instance of an object.
[5/20/2013 5:48:32 PM] Monitoring is turned on at 10 second intervals.
sr. member
Activity: 282
Merit: 250
I will definitely look into this and make sure it is corrected. I've also made some other improvements, so I'll be releasing 1.1.5 sooner than I expected... probably in the next few days.
ive donated 5 LTC to your way, a small token of gratitude. Also, i cant wait for CGRemote, need it badly :p

sr. member
Activity: 434
Merit: 251
CGWatcher & CGRemote
I didn't realize the latest version was still prone to it going nuts over API access but I just happened to check the log on another miner of mine and saw a similar group of entries. I will definitely look into this and make sure it is corrected. I've also made some other improvements, so I'll be releasing 1.1.5 sooner than I expected... probably in the next few days.

1.1.5 gets GPU data outside of the miner if the miner doesn't (specifically, Nvidia temperatures) and will take similar steps to prevent overheating as the miner does with AMD cards. If temps go over temp-overheat, it decreases intensity (instead of down-clocking). If temps go over temp-cutoff, it disables the GPU. Once temps return to temp-target, it will reset/re-enable. I know there aren't a lot of miners using Nvidia cards, but I thought they should at least be using the same protection against overheating as AMD miners have. I still have to do some tests and can't guarantee it works for every card, but it found most data for my GTX 285 that I've been testing it with. I don't know if anyone mines with Intel HD integrated graphics but I will probably add similar support for that just to cover all the bases. GPU data is updated continuously (about every few seconds) even if not mining.

There is also better support for non-responding miners and a restart computer option is added to the 'Restart if sick or dead GPUs are detected' check. Anyway, I will look into this and if anyone else is having a problem please let me know so I can get it fixed before the next release.
newbie
Activity: 29
Merit: 0
Thank you so much for the updates and for all your hard work. When I'll get my farm up and running, I'll be sure to donate for your time, much appreciated! Smiley

L.E. Was this a double consecutive restart? I was sleeping when it occurred so can't tell how it actually happened:

[5/18/2013 8:54:37 AM] CGMiner has been continuously mining for 8 hour(s), you've requested to restart after 8 hour(s). Attempting to restart...
[5/18/2013 8:54:38 AM] Restart command sent to CGMiner with full API access...
[5/18/2013 8:54:51 AM] CGMiner was successfully restarted.
[5/18/2013 8:55:02 AM] API access lost at 5/18/2013 8:55:02 AM!
[5/18/2013 8:55:02 AM] Full API access became read-only access at 5/18/2013 8:55:02 AM!
[5/18/2013 8:55:02 AM] I do not have API access to the currently running CGMiner process (3324)
[5/18/2013 8:55:03 AM] API access lost at 5/18/2013 8:55:03 AM!
[5/18/2013 8:55:03 AM] Full API access became read-only access at 5/18/2013 8:55:03 AM!
[5/18/2013 8:55:03 AM] I lost API access to a process in which I had it (PID: 3324). Attempting to restart CGMiner...
[5/18/2013 8:55:04 AM] API access lost at 5/18/2013 8:55:04 AM!
[5/18/2013 8:55:04 AM] Full API access became read-only access at 5/18/2013 8:55:04 AM!
[5/18/2013 8:55:05 AM] I had no API access to CGMiner so I killed the process. I will now attempt to re-open it...
[5/18/2013 8:55:10 AM] CGMiner started successfully.
[5/18/2013 8:55:11 AM] CGMiner was successfully restarted after killing its process.
sr. member
Activity: 434
Merit: 251
CGWatcher & CGRemote
18RATTT: I couldn't find anything that would cause this issue because starting the miner using the start button uses the same code as when it is started automatically. However, I did add some log entries to provide more information when auto-starting the miner, so if it happens again it may provide some help in locating the problem. It will now create debug log entries any time the miner is started (or stopped) showing the active profile, miner path, config file, and arguments that are being used to start it.

In regards to email notifications, it is something that has been requested and is on the list but no ETA on implementing.


pengoau: Thanks for the info. Did you have the 'Ensure miner stays running' option checked? When the miner is closed outside of CGWatcher, it doesn't know whether it crashed or you closed it purposely (clicking the X or using the menu in the miner) so it does not automatically start it unless this option is checked. If this option was checked and it failed to launch the miner (it may take several seconds, up to your monitor interval at most) please let me know, and include the data from the Report tab and any related log entries (make sure 'Show Debug Log Entries' is checked in the Log tab so these entries are included.)

Also, I've made some changes to the monitor that if the miner has stopped responding for 3 consecutive checks, it will be killed and restarted. This will happen as long as the monitor is enabled and does not require a specific option to be enabled. You can change the number of consecutive checks this has to happen before the miner will be restarted (default: 3) by changing the NotRespondingRestartChecks=3* to another value in the INI file in version 1.1.4.2 (and future versions.) This is something obvious that CGWatcher should have been doing already, but since I had never personally experienced the miner to stop responding I never watched for it.

Additionally, using the info you've provided, CGWatcher will now disable the WER UI ("close" window displayed when a program stops responding) upon starting and restore it to its previous setting upon closing. This may be sort of an overkill to make sure it is able to kill a non-responding miner successfully. If you should not want it to disable the WER UI (since it will also cause this "Close" window to not appear for other programs that stop responding), you can change AllowDisableWERUI=True to AllowDisableWERUI=False* in the INI file. I don't think having this window disabled is a big problem so it will disable it by default.


cyberworx: Please post in your own thread.


sydameton: I've changed the 'Restart when sick or dead GPUs are detected' option to allow you to select whether you want it to restart the miner or the computer. Currently this option is only available for when sick or dead GPUs are detected, as was requested. As I mentioned above, I also changed the monitor to check to see if the miner process is not responding and if it isn't respoding for 3 consecutive checks, it will restart the miner (killing the process if necessary.) This may be enough to correct the problem that you encountered without needing to restart the computer.

If you do set it to restart the computer when sick or dead GPUs are detected, it will give you a 30-second countdown window from which you can cancel the restart if you would like. By default, it will not force-close other programs when restarting because I didn't want it to cause anyone to lose unsaved work in other programs they may have running. If you want to change this to allow CGWatcher to force-close other programs when restarting, change ForceCloseProgramsOnRestart=False to ForceCloseProgramsOnRestart=True* in the INI file.


All of these changes have already been made and will be available in 1.1.4.2. Ultimately, I may create an option to restart the computer if restarting the miner fails... but this won't be in this update. This update should hopefully be available in the next few days.


(* - This setting may not exist in the INI file before you run 1.1.4.2 for the first time so you'll have to either start CGWatcher (which will add it to the INI file) or add the setting manually under the [Monitor] section.)
newbie
Activity: 29
Merit: 0
cgminer.exe has caused an error... lost three hours, after 48+ hours of running, no issues, no hw errors, nothing. That feature mentioned earlier would be really useful, to shutdown -r -f -t 30 or something in case of crash... Undecided Or an option to send an email in case of crash, then we can do it manually. Smiley

Also, from what I can tell it would seem when cgminer crashes (along with windows notification popping up), cgwatcher does not detect this event occuring as such, at least according to its log output.
newbie
Activity: 8
Merit: 0
I'll definitely try this one, it looks nice!
newbie
Activity: 16
Merit: 0
I've been looking for something like this, will try it out tonight  Smiley
newbie
Activity: 8
Merit: 0
Thank you for work, I'll try it for sure
full member
Activity: 208
Merit: 100
Ok to fix the problem of cgminer not closing until you hit "close" when the application crashes to desktop... add the "DontShowUI" key to your registry... Refer here: http://msdn.microsoft.com/en-us/library/bb513638.aspx.

So the program crashed and i wasn't prompted to close it, thanks to that registry setting but cgwatcher did not restart it... how come?

It knows its not running as it said so in the lower right corner..


sr. member
Activity: 282
Merit: 250
Milone,
i think i found a bug.
when you are using cgwatcher's autostart miner, the gpu fan is not set according to the parameter.

how to reproduce:
1. set cgminer.conf with the following parameter.
"gpu-fan" : "75-95,75-95,75-95",

2. turn on cgwatcher's turn on"Run when windows start" and "Run miner when started"
3. the gpu fan is not set to 75-95%, the fan doesnt run as its told, it will only run 2700ish rpm. (75-95% is around 3300rpm)
4. pause the mining.
5. start the mining from cgwatcher, the fan now runs properly.

so the conclusion is when you use the autostart miner the gpu fan didnt use the gpu-fan, however when you restart the miner it will run the gpu-fan properly.

also, could you make an email notification, when:
1. the workers is offline for 10min
2. when gpu reached a set of temp (eg 95%)

cant wait for CGremote.
sr. member
Activity: 434
Merit: 251
CGWatcher & CGRemote
Igloczek: Yes, this is possible. I will probably change the 'Restart miner when sick or dead GPUs are detected' option to allow you to choose whether you want the miner restarted or the computer restarted when this happens. Just keep in mind that you will need to have CGwatcher run at startup and automatically launch the miner if you want it to continue mining automatically after a system restart. Also, if the computer prompts to select a Windows user and/or enter a password after restarting, this will also require user input before mining resumes.


sydameton: Thanks, this was a bug as far as the log entry it made. The hashrate (0 Mh/s) was below the cutoff (1000 Mh/s), but the bug was that it reported this had been the case for at least 30 seconds, when it most likely was not. When it's inside the startup grace period, it doesn't really care how long it has been below the cutoff because it won't restart it anyway. Once outside the grace period, the timer resets and it begins to care.

In other words, ignore the 'for at least 30 seconds' part, but everything else is correct. I've modified the code so it will be fixed in the next update.
newbie
Activity: 29
Merit: 0
It's possible to add feature "Restart mining rig when dead or sick GPUs detected"? As we know, miner restart do nothing, when CCC crashes.

This feature is already implemented, along with restarting if total hash rate falls below value...

My bad, I misread the whole thing. That would indeed be an interesting function!

Potential bug report:

[5/15/2013 5:24:03 PM] Total hashrate fell below 1000 Mh/s (0 Mh/s) for at least 30 seconds, but am inside the miner's startup grace period (3 min) so did not attempt to restart yet.

I was actually looking at cgminer for something else and it was more like 10+ seconds, nowhere near 30. Is it possible it's calculating wrong? Then again, it could be me that's wrong here, but a quick code check should verify this. Smiley
newbie
Activity: 2
Merit: 0
It's possible to add feature "Restart mining rig when dead or sick GPUs detected"? As we know, miner restart do nothing, when CCC crashes.
Pages:
Jump to: