Looks like the GPU on computer 5529788 is a run-away validate error maker. Probably to do with the 7.0.24 client that the user installed, and that probably under Ubuntu and from repositories.
Any chance someone can help me troubleshoot numerous validate errors that I am receiving on one of my machines? All of the errors are "Validate error (8:00001000)" for Binary Radio Pulsar Search (Arecibo) v1.28 (opencl-ati) tasks. It looks like I'm getting ~20 results/day with validate errors, while the remainder of the tasks are all validating fine.
I'm running a 2500k @ 4.3Ghz and a 7970 @ 1100Mhz. Using BOINC client version 7.0.28 and my OS is Win7 Ultimate 64 bit. Also, I'm running 12.4 drivers.
Here are a few of the most recent validate errors:
I have no idea where to look to see what the actual error is, so if anyone can point me to where to see that output it would be greatly appreciated.
I'd hate to turn my 7970 down to stock clocks, especially since most of the work units are validating fine. If anyone else is running a 7970 and has had a similar experience I would love to hear how you solved it!
Any chance someone can help me troubleshoot numerous validate errors that I am receiving on one of my machines? All of the errors are "Validate error (8:00001000)" for Binary Radio Pulsar Search (Arecibo) v1.28 (opencl-ati) tasks. It looks like I'm getting ~20 results/day with validate errors, while the remainder of the tasks are all validating fine.
I'm running a 2500k @ 4.3Ghz and a 7970 @ 1100Mhz. Using BOINC client version 7.0.28 and my OS is Win7 Ultimate 64 bit. Also, I'm running 12.4 drivers.
Here are a few of the most recent validate errors:
I have no idea where to look to see what the actual error is, so if anyone can point me to where to see that output it would be greatly appreciated.
I'd hate to turn my 7970 down to stock clocks, especially since most of the work units are validating fine. If anyone else is running a 7970 and has had a similar experience I would love to hear how you solved it!
Thanks in advance for any thoughts/insights!
Saddly, 20 invalid WUs per day is a loss of 10K in the RAC, which Im sure is much more than what you might gain due to the OC...
OC'ing isnt as simple for crunching as its for games, if just on bit errors when rendering a frame you wont notice that there is a pixel with a may be slightly wrong color, but just one bit in a middle of a math calc is unacceptable. So the first thing you need to do is to turn off the OC and see if the you still get invalids (look at the reported time to differentiate which ones were crunched before the change). If you dont get invalids the start again with the OC but do it in small steps each day and keep checking that there are not more invalids between the results of that day... eventually you will reach a certain value that will fail again and then you will have to go back one or two steps with the OC... Too much OC on the memory clock is more likely to cause errors than the core clock.
Any chance someone can help me troubleshoot numerous validate errors that I am receiving on one of my machines? All of the errors are "Validate error (8:00001000)" for Binary Radio Pulsar Search (Arecibo) v1.28 (opencl-ati) tasks. It looks like I'm getting ~20 results/day with validate errors, while the remainder of the tasks are all validating fine.
I'm running a 2500k @ 4.3Ghz and a 7970 @ 1100Mhz. Using BOINC client version 7.0.28 and my OS is Win7 Ultimate 64 bit. Also, I'm running 12.4 drivers.
Here are a few of the most recent validate errors:
I have no idea where to look to see what the actual error is, so if anyone can point me to where to see that output it would be greatly appreciated.
I'd hate to turn my 7970 down to stock clocks, especially since most of the work units are validating fine. If anyone else is running a 7970 and has had a similar experience I would love to hear how you solved it!
Thanks in advance for any thoughts/insights!
Saddly, 20 invalid WUs per day is a loss of 10K in the RAC, which Im sure is much more than what you might gain due to the OC...
OC'ing isnt as simple for crunching as its for games, if just on bit errors when rendering a frame you wont notice that there is a pixel with a may be slightly wrong color, but just one bit in a middle of a math calc is unacceptable. So the first thing you need to do is to turn off the OC and see if the you still get invalids (look at the reported time to differentiate which ones were crunched before the change). If you dont get invalids the start again with the OC but do it in small steps each day and keep checking that there are not more invalids between the results of that day... eventually you will reach a certain value that will fail again and then you will have to go back one or two steps with the OC... Too much OC on the memory clock is more likely to cause errors than the core clock.
Thanks Horacio. I appreciate your thoughts. I'll try turning the OC down (or even completely off) and see if I still get invalids.
Until August 31 I used to crunch with my old 9600GT.
Most of the WUs get validated, some were marked as invalid.
Then at August 31 I decided to crunch with my newer GTX 560TI and I only get "Validate error" and the questions is: WHY?
This is the normal status of one task in a quorum if the validator checks the two tasks and has a problem with the other one.
Quote:
... the other user with Radeon card has a Validate error
If you look at his invalid tasks, he has lots and lots of them. I would guess that he has a problem with the way he is running that card. Perhaps it is overclocked too much or is not being cooled adequately. Perhaps he has a bad PSU or faulty RAM. When task after task fails like this, it's got to be some sort of hardware issue. I'll send him a PM and suggest that he investigates further.
Ok, My Host has not produced a single valid BRP4cuda32 task, all results are coming back validate error.
OS : Ubuntu 12.04 64 bit
32bit compatibility library/s are installed as per boinc instructions
Before I'm told it must be a hardware problem ,GPU apps for seti@home and GPUgrid work fine.
No overclocking is in use , bios is specifically set to run at stock speeds.
The computer, GPU, RAM and hard drive is all new , not a speck of dust and all fans work.
RAM has been tested for 48 hrs straight and did not produce a single error.
GPU is a GTX 460 2WIN ... 2 460's on one card. (just fyi)
Been through 6 builds of nvidia driver to find one that let seti@home work. maybe I just dont have the right one yet?
Being ubuntu 12.xx I cannot get NVclock to work anymore (it worked with older distros of ubuntu) so, I have no manual voltage or fan control of the GPU and specific manual attempts to change individual functions VIA nv-config, act like something happened but no confirmation output nor do the features change from their default settings
Any ideas?
Looks like the GPU on
)
Looks like the GPU on computer 5529788 is a run-away validate error maker. Probably to do with the 7.0.24 client that the user installed, and that probably under Ubuntu and from repositories.
Hi All, Any chance
)
Hi All,
Any chance someone can help me troubleshoot numerous validate errors that I am receiving on one of my machines? All of the errors are "Validate error (8:00001000)" for Binary Radio Pulsar Search (Arecibo) v1.28 (opencl-ati) tasks. It looks like I'm getting ~20 results/day with validate errors, while the remainder of the tasks are all validating fine.
I'm running a 2500k @ 4.3Ghz and a 7970 @ 1100Mhz. Using BOINC client version 7.0.28 and my OS is Win7 Ultimate 64 bit. Also, I'm running 12.4 drivers.
Here are a few of the most recent validate errors:
http://einsteinathome.org/task/304272771
http://einsteinathome.org/task/304258146
http://einsteinathome.org/task/304236313
I have no idea where to look to see what the actual error is, so if anyone can point me to where to see that output it would be greatly appreciated.
I'd hate to turn my 7970 down to stock clocks, especially since most of the work units are validating fine. If anyone else is running a 7970 and has had a similar experience I would love to hear how you solved it!
Thanks in advance for any thoughts/insights!
RE: Hi All, Any chance
)
Saddly, 20 invalid WUs per day is a loss of 10K in the RAC, which Im sure is much more than what you might gain due to the OC...
OC'ing isnt as simple for crunching as its for games, if just on bit errors when rendering a frame you wont notice that there is a pixel with a may be slightly wrong color, but just one bit in a middle of a math calc is unacceptable. So the first thing you need to do is to turn off the OC and see if the you still get invalids (look at the reported time to differentiate which ones were crunched before the change). If you dont get invalids the start again with the OC but do it in small steps each day and keep checking that there are not more invalids between the results of that day... eventually you will reach a certain value that will fail again and then you will have to go back one or two steps with the OC... Too much OC on the memory clock is more likely to cause errors than the core clock.
RE: RE: Hi All, Any
)
Thanks Horacio. I appreciate your thoughts. I'll try turning the OC down (or even completely off) and see if I still get invalids.
I hope this is the right
)
I hope this is the right place for my problem.
Please excuse my bad english^^
Until August 31 I used to crunch with my old 9600GT.
Most of the WUs get validated, some were marked as invalid.
Then at August 31 I decided to crunch with my newer GTX 560TI and I only get "Validate error" and the questions is: WHY?
This is my Host http://einsteinathome.org/host/4268504/tasks
The 2 GPUs are as mentioned the GTX 560TI and the 9600GT.
I'll be so happy, if someone could help me
I have this problem. May be
)
I have this problem. May be the same?
I quote (copy):
5-9-2012 19:31:13 | Einstein@Home | Scheduler request failed: Timeout was reached
5-9-2012 19:31:17 | | Project communication failed: attempting access to reference site
5-9-2012 19:31:19 | | Internet access OK - project servers may be temporarily down.
5-9-2012 19:34:04 | Einstein@Home | Sending scheduler request: To report completed tasks.
5-9-2012 19:34:04 | Einstein@Home | Reporting 1 completed tasks, not requesting new tasks
5-9-2012 19:39:12 | Einstein@Home | Scheduler request failed: Timeout was reached
5-9-2012 19:39:15 | | Project communication failed: attempting access to reference site
5-9-2012 19:39:17 | | Internet access OK - project servers may be temporarily down.
5-9-2012 19:44:12 | Einstein@Home | Sending scheduler request: To report completed tasks.
5-9-2012 19:44:12 | Einstein@Home | Reporting 1 completed tasks, not requesting new tasks
5-9-2012 19:49:19 | Einstein@Home | Scheduler request failed: Timeout was reached
5-9-2012 19:49:22 | | Project communication failed: attempting access to reference site
5-9-2012 19:49:24 | | Internet access OK - project servers may be temporarily down.
5-9-2012 19:57:54 | Einstein@Home | Sending scheduler request: To report completed tasks.
5-9-2012 19:57:54 | Einstein@Home | Reporting 1 completed tasks, not requesting new tasks
5-9-2012 20:03:02 | Einstein@Home | Scheduler request failed: Timeout was reached
5-9-2012 20:03:05 | | Project communication failed: attempting access to reference site
5-9-2012 20:03:07 | | Internet access OK - project servers may be temporarily down.
5-9-2012 20:23:32 | Einstein@Home | Sending scheduler request: To report completed tasks.
5-9-2012 20:23:32 | Einstein@Home | Reporting 1 completed tasks, not requesting new tasks
5-9-2012 20:28:39 | Einstein@Home | Scheduler request failed: Timeout was reached
5-9-2012 20:28:43 | | Project communication failed: attempting access to reference site
5-9-2012 20:28:45 | | Internet access OK - project servers may be temporarily down.
Quote ends
My other projects work correctly (cosmo(at)home, milkyway(at)home
Greetings,
Raimond
p2030.20110120.G193.61-02.25.
)
p2030.20110120.G193.61-02.25.S.b6s0g0.00000_120
Completed, validation inconclusive Task ID 308917792
the other user with Radeon card has a Validate error
RE: ... Completed,
)
This is the normal status of one task in a quorum if the validator checks the two tasks and has a problem with the other one.
If you look at his invalid tasks, he has lots and lots of them. I would guess that he has a problem with the way he is running that card. Perhaps it is overclocked too much or is not being cooled adequately. Perhaps he has a bad PSU or faulty RAM. When task after task fails like this, it's got to be some sort of hardware issue. I'll send him a PM and suggest that he investigates further.
Cheers,
Gary.
At SETI@home, Number
)
At SETI@home, Number Crunching, there is a thread "Invalid Host Messaging" where all such cases are reported.
Tullio
Ok, My Host has not produced
)
Ok, My Host has not produced a single valid BRP4cuda32 task, all results are coming back validate error.
OS : Ubuntu 12.04 64 bit
32bit compatibility library/s are installed as per boinc instructions
Before I'm told it must be a hardware problem ,GPU apps for seti@home and GPUgrid work fine.
No overclocking is in use , bios is specifically set to run at stock speeds.
The computer, GPU, RAM and hard drive is all new , not a speck of dust and all fans work.
RAM has been tested for 48 hrs straight and did not produce a single error.
GPU is a GTX 460 2WIN ... 2 460's on one card. (just fyi)
Been through 6 builds of nvidia driver to find one that let seti@home work. maybe I just dont have the right one yet?
Being ubuntu 12.xx I cannot get NVclock to work anymore (it worked with older distros of ubuntu) so, I have no manual voltage or fan control of the GPU and specific manual attempts to change individual functions VIA nv-config, act like something happened but no confirmation output nor do the features change from their default settings
Any ideas?