I have recently returned to Seti@home with a new machine ( 7862674 ). I read a number of messages about the work unit problems in early November, but understand that these should have mostly dissipated by now. However I am getting a very high percentage of invalid results.

Is there a possible problem with my machine/software/configuration that is causing this problem. If so what do I need to do to correct this situation?

Or will it eventually work itself out?

I have recently returned to Seti@home with a new machine ( 7862674 ). I read a number of messages about the work unit problems in early November, but understand that these should have mostly dissipated by now. However I am getting a very high percentage of invalid results.

Is there a possible problem with my machine/software/configuration that is causing this problem. If so what do I need to do to correct this situation?

Or will it eventually work itself out?

This should make it easier to find the machine.
https://setiathome.berkeley.edu/show_host_detail.php?hostid=7862674

Grabing the stderr for one of thes invalids before it vanishes.

Task 4637062679


Name 20fe11ab.6260.16435.10.19.232_0
Workunit 2020397652
Created 30 Dec 2015, 0:05:08 UTC
Sent 30 Dec 2015, 4:49:41 UTC
Report deadline 29 Feb 2016, 5:10:46 UTC
Received 30 Dec 2015, 11:35:32 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 7862674
Run time 4 min 19 sec
CPU time 4 min 6 sec
Validate state Invalid
Credit 0.00
Device peak FLOPS 3.24 GFLOPS
Application version SETI@home v7 v7.01
i686-pc-linux-gnu
Stderr output

<core_client_version>7.2.42</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_v7 7.00 Revision: 1782 g++ (GCC) 4.4.1 20090725 (Red Hat 4.4.1-2)
libboinc: BOINC 7.1.0

Work Unit Info:
...............
WU true angle range is : 0.307951
Optimal function choices:
--------------------------------------------------------
name timing error
--------------------------------------------------------
v_BaseLineSmooth (no other)
v_avxGetPowerSpectrum 0.000044 0.00000
avx_ChirpData_d 0.001967 0.00000
v_avxTranspose4x16ntw 0.000694 0.00000
JS AVX_a folding 0.000325 0.00000
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected equals the storage space allocated.

Flopcounter: 1479915313166.489746

Spike count: 0
Autocorr count: 30
Pulse count: 0
Triplet count: 0
Gaussian count: 0
06:33:46 (17945): called boinc_finish

</stderr_txt>
]]>

If this had been a task run on a GPU my first though would have been a thermal or overclocking issue, but this is a CPU task so less likely, but worth checking.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

If this had been a task run on a GPU my first though would have been a thermal or overclocking issue, but this is a CPU task so less likely, but worth checking.

Checking both the Asus bios and lm_sensors indicate nothing even close to hot and there have been no problems with any other applications that might indicate heat related problems.

If this had been a task run on a GPU my first though would have been a thermal or overclocking issue, but this is a CPU task so less likely, but worth checking.

Checking both the Asus bios and lm_sensors indicate nothing even close to hot and there have been no problems with any other applications that might indicate heat related problems.

I looked at the results and each and every one randomly chosen one had an issue with autocorrelations. It is a memory intensive phase in the calculations. It uses a lot of cpu too.

If the temps are ok I'd remove and reinstall the memory sticks.

If that does not help I'd check you have not set command rate T1 instead of T2 in bios memory settings by accident.

If that is not the case then I'd check the other values in bios. I have tried to overclock my CPU by notching up a bus related value (can not remember its name) from 100 to 103 or 125. It did not work. Changing multiplier worked though. I have an Asus mobo (Rampage IV black) and a push button to overclock. It does not work reliably with seti. I have to overclock it manually.

Make sure you have set the VDROOP compensation settings to mid or high. "Current capability" and "voltage compensation" or símilar: 120 - 140% both.

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

I have re-seated the memory and checked all of the bios settings. There is no overclocking and the "Current capability" and "voltage compensation" are both set to 140 now. I also changed the fan control from standard to turbo.

The increased fan speed lowered temperatures a little and for now I have limited seti to 2 cores instead of 4 which has also dropped the cpu temperature.

The 2 active cores are running around 55C and the overall cpu is at 45C. I will continue running this way to see if the invalid results situation improves.

Time will tell ...

After running overnight at much lower temperatures the results being generated still contain a very high percentage of failures. For now it is a waste of effort to run on this machine so it will be shutdown as soon as the remaining work units are completed.

I don't know what else to try ...

Just a thought, but do you have a spare PS handy? I have found this to be a problem in the past, and a new one has cleared things up.

---

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours

Just a thought, but do you have a spare PS handy? I have found this to be a problem in the past, and a new one has cleared things up.

That was the first thing I tried. No l such luck.

Just a thought, but do you have a spare PS handy? I have found this to be a problem in the past, and a new one has cleared things up.

That was the first thing I tried. No l such luck.

Well is was worth a guess. I think something is causing noise in the computer and giving you the results you are getting. Something going bad.

---

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours