Problems again?
5 Apr 2012 13:40:05 UTC
Topic 196271
(moderation:
Early last month my machine was failing most WU's for Einstein and Rosetta. Replaced RAM (easiest solution) with no joy. New CPU, motherboard, 16 gigs ram, harddrive and WIN7 64 bit instead of 32 bit. All was golden. Suddenly I'm getting errors. Some are "PRECISION OVERFLOW DENORMALIZED INVALID". I'm somewhat irritated and baffled as my days of being 'into' computers has drawn to a close. Any help/advice would be appreciated.
My computer ID:
Computer 4272932
Thanks in advance..john
Problems again?
)
But the same case, with the same PSU and the same fans?
I ask, as you didn't say whether or not you reused the old case etc.
Same PSU , case (big Antec) ,
)
Same PSU , case (big Antec) , fans (3). EVGA 9800GT GPU. PSU is a year old, 500 watt, Star Tech (for what that's worth). Electronics can fail at any time I know, and trusting what the manufacturer says nowadays is like believing a politician. Also running SETI on the GPU but not constantly. Time running SETI units does not match up with errors on Einstein and Rosetta.
And thanks for the reply!
Found the edit button!
Some error messages from the error file:
04-Apr-2012 03:18:27 [Einstein@Home] Computation for task h1_0370.95_S6GC1__1904_S6LV1B_0 finished
04-Apr-2012 03:18:27 [Einstein@Home] Output file h1_0370.95_S6GC1__1904_S6LV1B_0_0 for task h1_0370.95_S6GC1__1904_S6LV1B_0 absent
04-Apr-2012 03:18:09 [Einstein@Home] Computation for task h1_0370.95_S6GC1__1905_S6LV1B_1 finished
04-Apr-2012 03:18:09 [Einstein@Home] Output file h1_0370.95_S6GC1__1905_S6LV1B_1_0 for task h1_0370.95_S6GC1__1905_S6LV1B_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] [error] Signature verification failed for einsteinbinary_BRP4_1.00_graphics_windows_intelx86.exe
03-Apr-2012 21:46:10 [Einstein@Home] Computation for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 finished
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_0 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_1 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_2 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_3 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_4 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_5 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_6 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
03-Apr-2012 21:46:10 [Einstein@Home] Output file p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1_7 for task p2030.20111121.G37.01-00.01.N.b6s0g0.00000_216_1 absent
So BOINC has these files in C:ProgramData/BOINC. They were marked read only so unchecked that box. Windows promptly rechecked it. I don't know if this bears on the problem. Just irritating things fine for awhile then error messages that to my unskilled eye are not much help...Thanks again..
RE: PSU is a year old, 500
)
I'd pass by on http://extreme.outervision.com/PSUEngine and check if that's enough. I checked based on what you wrote, and came to ~400W, but you know what's exactly in your system. PSUs are about 70-75% efficient, which means that for it to output 70-75W it needs to generate 100W. So your PSU is becoming dangerously low or already too low, since 500 * (75/100) = 375W.
You have a lot of different errors, all starting since the third of April. If you don't think it's your PSU, did anything on your system change at that time? Did you add a driver, or did Windows install some update? Have you tried a reboot?
I'll check out that site for
)
I'll check out that site for power consumption. Thanks for the link. I've got plenty of old PSU's laying around from upgrades, so what's another new one?
No new software including no new windows updates. Rebooted so we'll see if that has an affect.
THanks again Ageless..john
Site recommends 358 watts which IF the PSU is operating properly my 500 watter is OK..matter pends..it's just frustrating..replacing several things and the best culprits I (we) can think of are PSU or software (Windows/BOINC/particular client) but two of the projects are throwing fits and I'm still running the same BOINC (6.2 if memory serves)..but then I've been suspicious of windows for a long time.."One ring to rule them all"..
Your BOINC is according to
)
Your BOINC is according to the stderr.txt files on tasks, 6.10.60
BOINC 7 is on the verge of being released, so you can upgrade to that when it's out.
However, it won't fix your errors.
What I saw of the error messages, exit code 99, error code 0xc00000005 etc. is pointing in the direction of software (drivers, anti-virus, firewall) or hardware. Now, it's also possible that some of the new RAM you put in is faulty, the only way to test that is with a strenuous multi-hour test run of Memtest86+.
Then there's the possibility for heat, since you changed to an FX-4100 --from what, exactly?-- did you add an after-market heat sink fan on it, or are you running with the stock cooler?
If you have the possibility, put in a stronger PSU.
Remember that when you run BOINC, that the science applications will put a severe strain on the hardware. You may get away with running a lower output PSU for daily stuff, as browsing the internet, but not when all electrons are needed, i.e. for crunching data here and at Rosetta. You don't say if you overclock either, but that needs extra power as well.
Replaced an AMD core two duo
)
Replaced an AMD core two duo (4400) with an AMD FX 4100 quad core (75 watt). CPU temps at 43C, mobo at 24C. Stock cooler. Running under full load. No overclock. Running bog standard (g). Reason I started SETI way back when was that it was supposed to be a good way to check your overclock (g). Now stock is good enough for me. Windows memory tester (for what it is worth) showed no problems. I'll d/l memtest (haven't used it in years) and give it a shot. I'll stop by the local computer shop and see what they have for PSU's.
Just went from NNT and d/l'ed 4 new WU's. Three errored out within 32 seconds.
06-Apr-2012 01:50:18 [Einstein@Home] Scheduler request completed: got 4 new tasks
06-Apr-2012 01:50:20 [Einstein@Home] Started download of skygrid_GC_Dc0.5_m0.3_0380Hz_S6Bucket.dat
06-Apr-2012 01:50:25 [Einstein@Home] Finished download of skygrid_GC_Dc0.5_m0.3_0380Hz_S6Bucket.dat
06-Apr-2012 01:50:25 [Einstein@Home] Starting h1_0371.00_S6GC1__1942_S6LV1B_0
06-Apr-2012 01:50:26 [Einstein@Home] Starting task h1_0371.00_S6GC1__1942_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:50:26 [Einstein@Home] Starting h1_0371.00_S6GC1__1941_S6LV1B_0
06-Apr-2012 01:50:27 [Einstein@Home] Starting task h1_0371.00_S6GC1__1941_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:50:27 [Einstein@Home] Starting h1_0370.95_S6GC1__1900_S6LV1B_0
06-Apr-2012 01:50:28 [Einstein@Home] Starting task h1_0370.95_S6GC1__1900_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:50:28 [Einstein@Home] Starting h1_0370.95_S6GC1__1899_S6LV1B_0
06-Apr-2012 01:50:28 [Einstein@Home] Starting task h1_0370.95_S6GC1__1899_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:50:54 [Einstein@Home] Computation for task h1_0371.00_S6GC1__1942_S6LV1B_0 finished
06-Apr-2012 01:50:54 [Einstein@Home] Output file h1_0371.00_S6GC1__1942_S6LV1B_0_0 for task h1_0371.00_S6GC1__1942_S6LV1B_0 absent
06-Apr-2012 01:50:55 [Einstein@Home] Computation for task h1_0370.95_S6GC1__1899_S6LV1B_0 finished
06-Apr-2012 01:50:55 [Einstein@Home] Output file h1_0370.95_S6GC1__1899_S6LV1B_0_0 for task h1_0370.95_S6GC1__1899_S6LV1B_0 absent
06-Apr-2012 01:50:56 [Einstein@Home] Computation for task h1_0370.95_S6GC1__1900_S6LV1B_0 finished
06-Apr-2012 01:50:56 [Einstein@Home] Output file h1_0370.95_S6GC1__1900_S6LV1B_0_0 for task h1_0370.95_S6GC1__1900_S6LV1B_0 absent
06-Apr-2012 01:51:00 [Einstein@Home] Computation for task h1_0371.00_S6GC1__1941_S6LV1B_0 finished
06-Apr-2012 01:51:00 [Einstein@Home] Output file h1_0371.00_S6GC1__1941_S6LV1B_0_0 for task h1_0371.00_S6GC1__1941_S6LV1B_0 absent
06-Apr-2012 01:52:04 [Einstein@Home] Sending scheduler request: To fetch work.
06-Apr-2012 01:52:04 [Einstein@Home] Reporting 4 completed tasks, requesting new tasks for CPU
06-Apr-2012 01:52:12 [Einstein@Home] Scheduler request completed: got 4 new tasks
06-Apr-2012 01:52:14 [Einstein@Home] Starting h1_0371.05_S6GC1__1895_S6LV1B_1
06-Apr-2012 01:52:15 [Einstein@Home] Starting task h1_0371.05_S6GC1__1895_S6LV1B_1 using einstein_S6LV1 version 110
06-Apr-2012 01:52:15 [Einstein@Home] Starting h1_0371.00_S6GC1__1940_S6LV1B_0
06-Apr-2012 01:52:16 [Einstein@Home] Starting task h1_0371.00_S6GC1__1940_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:52:16 [Einstein@Home] Starting h1_0370.95_S6GC1__1898_S6LV1B_0
06-Apr-2012 01:52:16 [Einstein@Home] Starting task h1_0370.95_S6GC1__1898_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:52:16 [Einstein@Home] Starting h1_0370.95_S6GC1__1897_S6LV1B_0
06-Apr-2012 01:52:17 [Einstein@Home] Starting task h1_0370.95_S6GC1__1897_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:52:42 [Einstein@Home] work fetch suspended by user
06-Apr-2012 01:52:43 [Einstein@Home] Task h1_0370.95_S6GC1__1898_S6LV1B_0 exited with zero status but no 'finished' file
06-Apr-2012 01:52:43 [Einstein@Home] If this happens repeatedly you may need to reset the project.
06-Apr-2012 01:52:43 [Einstein@Home] Computation for task h1_0371.05_S6GC1__1895_S6LV1B_1 finished
06-Apr-2012 01:52:43 [Einstein@Home] Output file h1_0371.05_S6GC1__1895_S6LV1B_1_0 for task h1_0371.05_S6GC1__1895_S6LV1B_1 absent
06-Apr-2012 01:52:43 [Einstein@Home] Computation for task h1_0371.00_S6GC1__1940_S6LV1B_0 finished
06-Apr-2012 01:52:43 [Einstein@Home] Output file h1_0371.00_S6GC1__1940_S6LV1B_0_0 for task h1_0371.00_S6GC1__1940_S6LV1B_0 absent
06-Apr-2012 01:52:44 [Einstein@Home] Restarting task h1_0370.95_S6GC1__1898_S6LV1B_0 using einstein_S6LV1 version 110
06-Apr-2012 01:52:48 [Einstein@Home] Computation for task h1_0370.95_S6GC1__1897_S6LV1B_0 finished
06-Apr-2012 01:52:48 [Einstein@Home] Output file h1_0370.95_S6GC1__1897_S6LV1B_0_0 for task h1_0370.95_S6GC1__1897_S6LV1B_0 absent
Just added C:/ProgramData/BOINC to my antivirus (Avast) exclusion list. Had already added E:/Internet/BOINC.
My bafflement continues. My continuing thanks..john
RE: an AMD FX 4100 quad
)
According to AMD, the TDP is 95Watt.
Now, you went from one CPU to another. Same Windows, or did you reinstall Windows? If same Windows (*), did you uninstall the previous motherboard chipset drivers? Did you install the new chipset drivers? These are normally available on a disc with the motherboard and otherwise from the motherboard manufacturer's web site.
(Sorry, I may be explaining things you already know. Just explaining for those who don't. ;-))
(*) Edit: I saw you went from 32bit to 64bit Windows. But even then, how did you do that? Completely clean install, or upgrade?
RE: RE: an AMD FX 4100
)
Hello Jord. Don't worry about over explaining, it is too easy to make assumptions about some ones experience. Appreciate you taking the extra time! Brand new install, installed new mobo drivers from their install disk. I got the power use number from the newwegg site when I bought the new hardware..my meat memory could be malfunctioning too!
And speaking of memory..d/l'ed the memtest prg and put it on a thumb drive..one 4 gig stick total garbage. Running 12 gigs now, just finished four Rosetta units and working on my final Einstein WU. See how they work out and if my machine is not making garbage I'll allow new work units and all will be well in the world (g). Thanks again for all of your assistance and hand-holding (g)..john
RE: And speaking of
)
You're welcome. Hope you have warranty on the memory, most good memory makers have lifetime warranty of memory sticks, but they then need to have factory or shipment errors. Check the warranty requirements on the manufacturer's web site.
I think you're clean now.
New stick $25.00 minus $10.00
)
New stick $25.00 minus $10.00 gift certificate + free shipping. Definately not worth returning the 'kit' of two sticks for replacement. I remember when ram finally went under $50.00 for one meg (g). A two meg video card was the cats pajamas! But way back then a gallon of gas was around $1.25...now $25.00 won't fill up your tank...thanks again for your help, I'll keep my fingers crossed..john