GTX 1060 3gb - CUDA error 11 - cannot write buffer for DAG

[hous26]

My 6x GPU 1060 3gb rig has been unstable recently.  I use it only to mine Ethash coins and I havent touched the Afterburner settings.  I continuously get errors now while I mine Ethash + SC and cannot mine Ethash + LBRY at all anymore.  Anyone know what the problem is?  The gpus had been running without issue for 6 months prior to this.

19:44:08:920   bf0   GPU 0, CUDA error 11 - cannot write buffer for DAG

19:44:08:921   11b0   GPU 0 failed
19:44:09:094   33c   Set global fail flag, failed GPU4
19:44:09:094   ee8   Setting DAG epoch #160 for GPU4
19:44:09:096   ee8   GPU 4, CUDA error 11 - cannot write buffer for DAG

19:44:09:097   33c   GPU 4 failed
19:44:09:120   12f8   Set global fail flag, failed GPU2
19:44:09:120   8a0   Setting DAG epoch #160 for GPU2
19:44:09:121   12f8   GPU 2 failed
19:44:09:123   8a0   GPU 2, CUDA error 11 - cannot write buffer for DAG

19:44:09:129   1014   Set global fail flag, failed GPU3
19:44:09:130   1014   GPU 3 failed
19:44:09:132   13cc   Setting DAG epoch #160 for GPU3
19:44:09:133   13cc   GPU 3, CUDA error 11 - cannot write buffer for DAG

19:44:09:131   c44   Set global fail flag, failed GPU1
19:44:09:134   c44   GPU 1 failed
19:44:09:133   f40   Set global fail flag, failed GPU5
19:44:09:136   f40   GPU 5 failed
19:44:09:137   1104   Setting DAG epoch #160 for GPU1
19:44:09:138   1104   GPU 1, CUDA error 11 - cannot write buffer for DAG

19:44:09:140   10d0   Setting DAG epoch #160 for GPU5
19:44:09:141   10d0   GPU 5, CUDA error 11 - cannot write buffer for DAG

[hous26]

Anyone?

[bigjee]

Return gpu to stock setting and see if it still fails.
If they dont your oc is unstable.

If its still failing after default then check power supply and possibly a failing riser.
Look at AB graphs to give you a clue as to why they are failing.

Also are the cards heating up in anyone what the ambient temp.
Is the rig running under new conditions for the first time in 6 months?

[hous26]

Return gpu to stock setting and see if it still fails.
If they dont your oc is unstable.

If its still failing after default then check power supply and possibly a failing riser.
Look at AB graphs to give you a clue as to why they are failing.

Also are the cards heating up in anyone what the ambient temp.
Is the rig running under new conditions for the first time in 6 months?

Thanks, I will run through this when I get home from work.  Conditions have not changed.  I noticed the heat shield (i think thats what it is) below the risers has been peeling on all of them.  The rig has a box fan on it and I haven't observed a heat problem.

[bigjee]

Return gpu to stock setting and see if it still fails.
If they dont your oc is unstable.

If its still failing after default then check power supply and possibly a failing riser.
Look at AB graphs to give you a clue as to why they are failing.

Also are the cards heating up in anyone what the ambient temp.
Is the rig running under new conditions for the first time in 6 months?

Thanks, I will run through this when I get home from work.  Conditions have not changed.  I noticed the heat shield (i think thats what it is) below the risers has been peeling on all of them.  The rig has a box fan on it and I haven't observed a heat problem.

That heatshield/insulation foam pad (prevents riser from shorting out the board components) should be okay they tend to peel off i just tape them back on.
Also look at how much dust is on the cards. Might be a good idea to remove dust every quarter to maintain airflow.
Good luck!

[hous26]

Return gpu to stock setting and see if it still fails.
If they dont your oc is unstable.

If its still failing after default then check power supply and possibly a failing riser.
Look at AB graphs to give you a clue as to why they are failing.

Also are the cards heating up in anyone what the ambient temp.
Is the rig running under new conditions for the first time in 6 months?

Thanks, I will run through this when I get home from work.  Conditions have not changed.  I noticed the heat shield (i think thats what it is) below the risers has been peeling on all of them.  The rig has a box fan on it and I haven't observed a heat problem.

That heatshield/insulation foam pad (prevents riser from shorting out the board components) should be okay they tend to peel off i just tape them back on.
Also look at how much dust is on the cards. Might be a good idea to remove dust every quarter to maintain airflow.
Good luck!

Yeah, I'll dust everything out tonight.  I'll be honest, I haven't removed any dust yet.  Best to use a can of compressed air, right?

[bigjee]

Return gpu to stock setting and see if it still fails.
If they dont your oc is unstable.

If its still failing after default then check power supply and possibly a failing riser.
Look at AB graphs to give you a clue as to why they are failing.

Also are the cards heating up in anyone what the ambient temp.
Is the rig running under new conditions for the first time in 6 months?

Thanks, I will run through this when I get home from work.  Conditions have not changed.  I noticed the heat shield (i think thats what it is) below the risers has been peeling on all of them.  The rig has a box fan on it and I haven't observed a heat problem.

That heatshield/insulation foam pad (prevents riser from shorting out the board components) should be okay they tend to peel off i just tape them back on.
Also look at how much dust is on the cards. Might be a good idea to remove dust every quarter to maintain airflow.
Good luck!

Yeah, I'll dust everything out tonight.  I'll be honest, I haven't removed any dust yet.  Best to use a can of compressed air, right?

Yup and lock the fan with your finger before you spray that way it doesnt spin too fast to damage the bearings.

[knohult]

19:44:09:141   10d0   GPU 5, CUDA error 11 - cannot write buffer for DAG

Did you find it? Same issue here with my new gtx 1060 3gb. Tried virtual memory to 16gb without results.

[hous26]

19:44:09:141   10d0   GPU 5, CUDA error 11 - cannot write buffer for DAG

Did you find it? Same issue here with my new gtx 1060 3gb. Tried virtual memory to 16gb without results.

tbh I haven't had a chance to do anything with it since I been super busy with my work and its been out of my thoughts since I've been mining other algos without any problems.

[rdluffy]

I made a post here today, I'm having issues with my nvidia 1060rig

Months before, I was able to do 25mhs each card, but now I can't and the rig crashes a lot

I don't know if is the drivers, windows updates etc, very strange

Other algos are fine