OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[kano]

... or just install 11.04 or 11.10 ...

Well, he mentioned he didn't want to do that to be able to carry over changes in settings...

But he only installed it a few days ago.
Start again from scratch sounds like a way better idea ...

I'd also mention that since this is all to do with messing with installing and upgrading linux - there are WAY better forums to ask about that ... like the forum for the linux he's using as an example ...

[1713429214]

1713429214

[1713429214]

1713429214

[Proofer]

I'd also mention that since this is all to do with messing with installing and upgrading linux - there are WAY better forums to ask about that ... like the forum for the linux he's using as an example ...

Yeah, I was also thinking along those lines.  My problems stem from ATI drivers which I need for cgminer, but cgminer's involvement is pretty indirect at this point.  So I'll go over there (I'm "brec" on ubuntuforums.org) and come back here if/when I have cgminer issues.

Thanks, though, to all and I expect I'll be back after not too long...

[Red Emerald]

You guys have scared me away from ever updating my graphics drivers

[-ck]

You guys have scared me away from ever updating my graphics drivers

I have one major mining box that's been mining for 6 months solid at 1750+MH/s without ever changing the drivers from the reliable 11.6 drivers... oh and it's never once crashed.

[kano]

Same here
With the USB script in my sig (which I also had help directly from ckolivas and his README when I wrote it) to install 11.04 - which I did back in July - that machine has also never crashed while mining nor has cgminer ever crashed for me except when I was working on the code of course

Of course it's had the rare USB boot problem, but that's nothing to do with the linux version or cgminer - that's to do with USB and BIOS issues (and the simple 10Watt solution is to install an HDD instead if you can't have that USB problem ever happen due to not having direct access to the machine)
I reboot it about 5 times a week coz I shut it down weekdays at 2pm and start it up again at 8pm and of those around 100 reboots I've only had a few failures when I don't power it off and just reboot.
It always works when I do power it off before restarting it.

Again the reason why I said just install 11.04 or 11.10: people use it and it works.

I've been using Fedora for many years (setting up many offices and servers for clients) and in all that time I always have done a wipe and install on upgrade (of course I have auto backups and such that save all the data and system configuration)
It is starting to get better on some of the distros to do upgrades, but I've avoided it in order to avoid those random problems you find just when you don't want them.
In the next few days I'm going to update this desktop (fc12) to fc14 or fc15 and even though it's got almost 2 years of configuration I've played with, I'm still thinking that I'll wipe and reinstall rather than upgrade ... not 100% decided yet but I'm still wary of upgrading.
If I do change my mind and upgrade, it will be my first time ever to use that path.

[tnkflx]

Con,

Do you have an idea about the screenshot I posted?

[simonk83]

Ok, so recently one of my GPU keeps being set to disable.  I assume it's hitting the temp cutoff (it must be running hotter lately)?   Am I correct in thinking that cgminer ought to be attempting to restart this GPU once it's disabled?   Perhaps it's trying and failing (or perhaps it's not trying at all), I'm not sure.. just thought I'd best clarify that here first

Here's my startup script if that helps:



start /B /affinity 8 cgminer -o http://127.0.0.1:8337 -u simonk83 -p x -I 9,9,d,9 --auto-fan --auto-gpu --gpu-engine 930,910,900,930 --gpu-memclock 305 --temp-target 75,80,80,75 2>"%date:~4,2%-%date:~7,2%-%date:~12,2%%time:~0,2%-%time:~3,2%-%time:~6,2%.log"

pause

EDIT:  And here's the relevant section of the log:

[2011-12-18 18:03:56] Overheat detected on GPU 1, increasing fan to 100%
[2011-12-18 18:03:57] Hit thermal cutoff limit on GPU 1, disabling!
[2011-12-18 18:03:57] Thread 5 being disabled
[2011-12-18 18:03:57] Thread 1 being disabled

So it is overheating.   I was hoping I could get cgminer to re-enable after a certain period?

[-ck]

Con,

Do you have an idea about the screenshot I posted?

If the ADL library is returning unsupported as the value from a device, cgminer won't try and show temp/fanspeeds for that device. If it's failing sometimes only for you, then the library is doing something funky :\ It's a miracle we ever got the library to control and monitor those values but it's an even bigger miracle that it works mostly as advertised. Unfortunately it's not always very reliable...

[simonk83]

Ok, so recently one of my GPU keeps being set to disable.  I assume it's hitting the temp cutoff (it must be running hotter lately)?   Am I correct in thinking that cgminer ought to be attempting to restart this GPU once it's disabled?   Perhaps it's trying and failing (or perhaps it's not trying at all), I'm not sure.. just thought I'd best clarify that here first

Here's my startup script if that helps:



start /B /affinity 8 cgminer -o http://127.0.0.1:8337 -u simonk83 -p x -I 9,9,d,9 --auto-fan --auto-gpu --gpu-engine 930,910,900,930 --gpu-memclock 305 --temp-target 75,80,80,75 2>"%date:~4,2%-%date:~7,2%-%date:~12,2%%time:~0,2%-%time:~3,2%-%time:~6,2%.log"

pause

EDIT:  And here's the relevant section of the log:

[2011-12-18 18:03:56] Overheat detected on GPU 1, increasing fan to 100%
[2011-12-18 18:03:57] Hit thermal cutoff limit on GPU 1, disabling!
[2011-12-18 18:03:57] Thread 5 being disabled
[2011-12-18 18:03:57] Thread 1 being disabled

So it is overheating.   I was hoping I could get cgminer to re-enable after a certain period?

Right OK, my mistake, just noticed the following in the readme:

If the temperature goes over the
cutoff limit (95 degrees by default), cgminer will completely disable the GPU
from mining and it will not be re-enabled unless manually done so.

Is there any chance of getting a watchdog or something added so that it automatically re-enables the GPU after 5 mins or so?   In my case, the GPU is generally fine, but only occasionally overheats depending on the room.   If I manually restart a disabled GPU it tends to be fine for quite a while going forward.

[Proofer]

I regret to say.... I'm back.

For those who were following my troubles through yesterday:  I'm having essentially the same problem after having upgraded to Ubuntu 11.10.  It doesn't matter whether I run 2.0.8 that I rebuilt under 11.10 or the distributed binary (although I notice that the distributed one is about 36K smaller than the one I built).

I can't connect to the pool I joined, Eclipse, because it's returning a 403 (Forbidden) -- don't know why yet.  I know 403 is the return due do previous runs with -D (debug output) in cgminer.  Because of that I'm failing over to solo mining with a connection to bitcoind on my Mac on my LAN.

In sum, after some initial activity for a few seconds there is no further output (even with -D) and there is no response to keypresses.  Here is the record of a run without -D in which I pressed ^C after six minutes -- it WAS responsive to the ^C and output a summary, but I couldn't get control of the terminal back.

Code:

[2011-12-18 14:59:45] Started cgminer 2.0.8
[2011-12-18 14:59:51] Pool down, URL or credentials invalid
[2011-12-18 14:59:51] Unable to get work from pool 0 http://us.eclipsemc.com:8337

htop from another terminal:
  PID USER     PRI  NI  VIRT   RES   SHR S CPU% MEM%   TIME+  Command 
 2259 brec      20   0  450M  154M 21916 S 98.0  3.9  4:22.78  |               |   `- cgminer -c cgminer.conf
 2279 brec      20   0  450M  154M 21916 D  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2278 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2276 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.35  |               |       `- cgminer -c cgminer.conf
 2275 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2273 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.38  |               |       `- cgminer -c cgminer.conf
 2272 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2270 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.39  |               |       `- cgminer -c cgminer.conf
 2269 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2267 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.39  |               |       `- cgminer -c cgminer.conf
 2266 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2264 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.42  |               |       `- cgminer -c cgminer.conf
 2263 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2262 brec      20   0  450M  154M 21916 R 16.0  3.9  0:43.45  |               |       `- cgminer -c cgminer.conf
 2261 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf
 2260 brec      20   0  450M  154M 21916 S  0.0  3.9  0:00.00  |               |       `- cgminer -c cgminer.conf

on the cgminer terminal:
###################Ctrl-C entered here after six minutes######################
^C[2011-12-18 15:06:41]
Summary of runtime statistics:

[2011-12-18 15:06:41] Started at [2011-12-18 14:59:51]
[2011-12-18 15:06:41] Runtime: 0 hrs : 0 mins : 0 secs
[2011-12-18 15:06:41] Average hashrate: 0.0 Megahash/s
[2011-12-18 15:06:41] Solved blocks: 0
[2011-12-18 15:06:41] Queued work requests: 7
[2011-12-18 15:06:41] Share submissions: 0
[2011-12-18 15:06:41] Accepted shares: 0
[2011-12-18 15:06:41] Rejected shares: 0
[2011-12-18 15:06:41] Hardware errors: 0
[2011-12-18 15:06:41] Efficiency (accepted / queued): 0%
[2011-12-18 15:06:41] Utility (accepted shares / min): 0.00/min

[2011-12-18 15:06:41] Discarded work due to new blocks: 0
[2011-12-18 15:06:41] Stale submissions discarded due to new blocks: 0
[2011-12-18 15:06:41] Unable to get work from server occasions: 0
[2011-12-18 15:06:41] Work items generated locally: 0
[2011-12-18 15:06:41] Submitting work remotely delay occasions: 0
[2011-12-18 15:06:41] New blocks detected on network: 1

[2011-12-18 15:06:41] Pool: http://us.eclipsemc.com:8337
[2011-12-18 15:06:41]  Queued work requests: 0
[2011-12-18 15:06:41]  Share submissions: 0
[2011-12-18 15:06:41]  Accepted shares: 0
[2011-12-18 15:06:41]  Rejected shares: 0
[2011-12-18 15:06:41]  Efficiency (accepted / queued): 0%
[2011-12-18 15:06:41]  Discarded work due to new blocks: 0
[2011-12-18 15:06:41]  Stale submissions discarded due to new blocks: 0
[2011-12-18 15:06:41]  Unable to get work from server occasions: 0
[2011-12-18 15:06:41]  Submitting work remotely delay occasions: 0

[2011-12-18 15:06:41] Pool: http://192.168.168.103:8332
[2011-12-18 15:06:41]  Queued work requests: 7
[2011-12-18 15:06:41]  Share submissions: 0
[2011-12-18 15:06:41]  Accepted shares: 0
[2011-12-18 15:06:41]  Rejected shares: 0
[2011-12-18 15:06:41]  Efficiency (accepted / queued): 0%
[2011-12-18 15:06:41]  Discarded work due to new blocks: 0
[2011-12-18 15:06:41]  Stale submissions discarded due to new blocks: 0
[2011-12-18 15:06:41]  Unable to get work from server occasions: 0
[2011-12-18 15:06:41]  Submitting work remotely delay occasions: 0

[2011-12-18 15:06:41] Summary of per device statistics:

[2011-12-18 15:06:41] GPU0 59.0C 4524RPM | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:4
[2011-12-18 15:06:41] GPU1 62.5C | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:4
[2011-12-18 15:06:41] GPU2 60.5C 4530RPM | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:8
[2011-12-18 15:06:41] GPU3 55.5C | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:8
[2011-12-18 15:06:56] GPU4 58.5C | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:8
[2011-12-18 15:06:56] GPU5 61.5C | (5s):0.0 (avg):0.0 Mh/s | A:0 R:0 HW:0 U:0.00/m I:8
#######unable to get control of terminal back from zombie'd process########

after the ^C, from another terminal...
htop:
 2262 brec      20   0  450M  154M 22128 R 17.0  3.9  1:48.65 cgminer -c cgminer.conf                                   
 2264 brec      20   0  450M  154M 22128 R 16.0  3.9  1:48.62 cgminer -c cgminer.conf
 2267 brec      20   0  450M  154M 22128 R 16.0  3.9  1:48.62 cgminer -c cgminer.conf
 2270 brec      20   0  450M  154M 22128 R 16.0  3.9  1:48.60 cgminer -c cgminer.conf
 2273 brec      20   0  450M  154M 22128 R 16.0  3.9  1:48.59 cgminer -c cgminer.conf
 2276 brec      20   0  450M  154M 22128 R 16.0  3.9  1:48.57 cgminer -c cgminer.conf
 2279 brec      20   0  450M  154M 22128 D  0.0  3.9  0:00.00 cgminer -c cgminer.conf

ps -C cgminer
  PID TTY          TIME CMD
 2259 pts/0    00:10:27 cgminer <defunct>

[-ck]

Right OK, my mistake, just noticed the following in the readme:

If the temperature goes over the
cutoff limit (95 degrees by default), cgminer will completely disable the GPU
from mining and it will not be re-enabled unless manually done so.

Is there any chance of getting a watchdog or something added so that it automatically re-enables the GPU after 5 mins or so?   In my case, the GPU is generally fine, but only occasionally overheats depending on the room.   If I manually restart a disabled GPU it tends to be fine for quite a while going forward.

It's possible, but currently cgminer can't differentiate a GPU being disabled from the menu or at startup from being disabled due to overheating, so it was chosen as being intrinsically safe to be that way. I'll think about it.

[simonk83]

Right OK, my mistake, just noticed the following in the readme:

If the temperature goes over the
cutoff limit (95 degrees by default), cgminer will completely disable the GPU
from mining and it will not be re-enabled unless manually done so.

Is there any chance of getting a watchdog or something added so that it automatically re-enables the GPU after 5 mins or so?   In my case, the GPU is generally fine, but only occasionally overheats depending on the room.   If I manually restart a disabled GPU it tends to be fine for quite a while going forward.

It's possible, but currently cgminer can't differentiate a GPU being disabled from the menu or at startup from being disabled due to overheating, so it was chosen as being intrinsically safe to be that way. I'll think about it.

Ah, I see.

This is probably simpler than the actual solution will be, but are you able to do something like:

If the GPU was previously active in the current session (eg it wasn't disabled at startup), and is now inactive without manual intervention (eg, disabled from the menu), then re-enable?

I guess you'd need some sort of flag to check for (a "disabled from the menu flag").    

I should probably shut up anyway as I have no idea how you've coded this and what is and isn't possible so I probably sound like an idiot

[-ck]

Heh, it's okay. I can implement anything I like if I feel like it, but I was just trying hard not to make any significant changes to the code for fear of breaking the heard-earned stability it now has.

[kano]

I'd suggest that any auto re-enabling of a GPU that has over-heated is a bad idea.

If you do that I'm certain someone somewhere will have their GPU kill itself and then ...

On the other hand, if someone chooses to re-enable a GPU - well that's their problem isn't it.

If you do go down that path - make it an option that is OFF by default and someone has to turn on at their choice to risk killing their hardware by their choice.

[Red Emerald]

Same here
With the USB script in my sig (which I also had help directly from ckolivas and his README when I wrote it) to install 11.04 - which I did back in July - that machine has also never crashed while mining nor has cgminer ever crashed for me except when I was working on the code of course

Of course it's had the rare USB boot problem, but that's nothing to do with the linux version or cgminer - that's to do with USB and BIOS issues (and the simple 10Watt solution is to install an HDD instead if you can't have that USB problem ever happen due to not having direct access to the machine)
I reboot it about 5 times a week coz I shut it down weekdays at 2pm and start it up again at 8pm and of those around 100 reboots I've only had a few failures when I don't power it off and just reboot.
It always works when I do power it off before restarting it.

Again the reason why I said just install 11.04 or 11.10: people use it and it works.

I've been using Fedora for many years (setting up many offices and servers for clients) and in all that time I always have done a wipe and install on upgrade (of course I have auto backups and such that save all the data and system configuration)
It is starting to get better on some of the distros to do upgrades, but I've avoided it in order to avoid those random problems you find just when you don't want them.
In the next few days I'm going to update this desktop (fc12) to fc14 or fc15 and even though it's got almost 2 years of configuration I've played with, I'm still thinking that I'll wipe and reinstall rather than upgrade ... not 100% decided yet but I'm still wary of upgrading.
If I do change my mind and upgrade, it will be my first time ever to use that path.

If one of my GPUs dies, cgminer hasn't ever successfully restarted it.  If I hit "q" to quit, it hangs.  I've just been rebooting the server and that works easy enough.  5970 and a 5830 on BAMT.  I'm planning on following your ubuntu guide next I redo my rig.  It would be nice to have a simple ubuntu setup with live-build kind of like how I have for bitsafe but with cgminer instead of bitcoind.

Side note: When you do a fedora upgrade, you can only go up 1 or 2 versions.  So fc12 -> fc15 will give you an error.  They also changed the boot partition to being 500MB instead of 200 so that can be an annoyance. Also, 16 is out, but I miss chkconfig and haven't learned systemctl fully yet

[kano]

The fc upgrade would be 12->13 then 13->14 if I did it ... if ... hmm yep I still don't think I will.
My wipe/reinstall is very simple to do: boot a live USB/CD then on / and /boot create a folder called saveYYYYMMDD and move all directories into it (except lost+found) then boot and do an install and tell it to NOT touch/change the partitions.
That makes it easy to find old setting in /etc and anywhere else by just looking in /saveYYYYMMDD/etc/... of course I also have the last night backup on my main file server also
I don't use the default disk layout (and never have in the last 10 years) - it's silly.
I have a boot partition (partition 1) usually about 1G, a swap partition (partition 3) between 1.5 and 2 times RAM a system partition (partition 2) of usually around 100GB and the rest partition 4 I call 'extra' which is temporary backup area used during the nightly backup and to store any data that can be deleted without caring about it like disk images and such
Then of course there is always 2 more disks in RAID 1 that have all important data on it including: /home, /var/log, /var/www, /var/spool, /root and anything else needed that the computer uses like /var/lib/pgsql and /var/named all of these are links to a directory on the RAID1 disk.
Hmm yep off topic, but hopefully of use to anyone considering actually doing anything with linux other than just mining
Edit: N.B. there is a little trick required to actually move those directories to the RAID disk of: making it fail boot and go to the special recovery login option where you can then remount / and mount the RAID1 disk and copy the data and make the link and when you exit it reboots
(you do this by creating a specially designed faulty mount in /etc/fstab that you remove to make it boot properly)

[bronan]

You guys have scared me away from ever updating my graphics drivers

Well really honest if you do not game like me, and have no issues which are directly related to screen drivers DO NOT UPDATE
I game a lot so yea i have to plunge into the dark deepness of new drivers which often have a fix for the games i play but most of the time the result is that something else gets messed.
But its not allways a true gpu driver problem, i am not going into this further since i becomes really complex if i do so.
Just saying the graphics driver is a very important one in your system and many things try to "talk" to it direct or indirect.
When i get into trouble i either install 11.6 when using cgminer or if possible even much older catalyst versions in the 10.x range.
It seems every version has its problems and the guy who made the wonderfull omega drivers never returned so we are stuck on what nvidia and amd crappy developers put out into the world

[Proofer]

I have done a clean install of Ubuntu 11.10 as part of a conservative strategy to eliminate as many variables as possible in order to get cgminer 2.0.8 to run successfully.  Anyone unfamiliar with my problems thus far who's interested can see my previous posts here and here.

I have 5970 GPUs. How shall I install the ATI driver/utility/dev, or in other words, which version?  The packages version offered by Ubuntu 11.10 (not to be confused with AMD Catalyst 11.10) is 8.881, or with the post-release updates, 8.901.  The version in the current Catalyst 11.12 download from AMD is 8.920.  Or if I should be seeking some even earlier version, where?

[-ck]

If one of my GPUs dies, cgminer hasn't ever successfully restarted it.  If I hit "q" to quit, it hangs.  I've just been rebooting the server and that works easy enough.  5970 and a 5830 on BAMT. 

There are numerous potential modes of failure for a GPU, some soft hangs that cgminer successfully restarts, some soft hangs that cgminer cannot restart, and some hard hangs like the one you describe. I spent literally a month trying to find the right balance with the restarts such that more than anything else, cgminer would not immediately crash the machine or stop mining with whatever was remaining.

The restart we're talking about above, however, is for when cgminer voluntarily stops the GPU due to overheat, not because the GPU has stoppped responding to the code. This should be safe to restart at any time, PROVIDED I modify the code to know the GPU has not died, but has been voluntarily shut down. It will take some more code to do this of course, but I'm waiting for the RPC mods by Kano to be completed before doing anything more to cgminer. It makes it much easier for him to hack on if the code target does not move.

Also, Merry non-denominational festive season, everyone.

[AniceInovation]

Hi,

Just installed CGMINER 2.08 on a windows rig with a 5870 overclocked to 990 MHZ, i was getting 400 Mhash/s.
Quite low i thought, so i tried 2.05.
Now i'm getting 440 MHash/s.

Did some code changed from 2.05 to 2.08 that can get this difference in performance?