AFAIK, Validate Errors are always a server problem.

This is not usually the case. Something is happening at their end to cause this rash of validate errors, but generally, validate errors are caused by the users host not performing the calculations correctly. (heat problems, bad overclock, iffy memory, etc)

I just got this one don't know why, maybe a server problem?
...

AFAIK, Validate Errors are always a server problem.

If you look at http://setiathome.berkeley.edu/explain_state.php?field=result_outcome they are explained there.

"Validate error The task was reported but could not be validated, typically because the output files were lost on the server."

BTW, AFAIK = as far as I know, BTW = by the way.

I think we need a glossary somewhere to help new users and those whose first language is not English.

Quoting my own post here as I think I should add this:

Some BOINC Client versions are "more prone" to getting VEs than others. Those which "Report Results Immediately" (RRI). Also I think some of the earlier varients of ver. 5.9 and 5.10 if the Connect Interval (CI) is set too low. Back in June, Joe Segur suggested setting a minimum CI of 0.001 days (86.4 seconds).

I am still using BOINC 5.10.7 with a CI of 0.007 days = 604.8 seconds, or just over 10 minutes.

Some of the latest BOINC versions, since 5.10.14 have updated scheduler code that I think ignores low CI settings. I could be wrong on this, but I seem to remember 5.10.14 being the one where this changed.

---

Sir Arthur C Clarke 1917-2008

AFAIK, Validate Errors are always a server problem.

This is not usually the case. Something is happening at their end to cause this rash of validate errors, but generally, validate errors are caused by the users host not performing the calculations correctly. (heat problems, bad overclock, iffy memory, etc)

Sorry Tony but I totally disagree with that.

Compute errors are caused by the things that you mentioned.

Validate errors are caused when the Validator has a problem.

[edit]
Added link.
[/edit]

AFAIK, Validate Errors are always a server problem.

This is not usually the case. Something is happening at their end to cause this rash of validate errors, but generally, validate errors are caused by the users host not performing the calculations correctly. (heat problems, bad overclock, iffy memory, etc)

Sorry Tony but I totally disagree with that.

Compute errors are caused by the things that you mentioned.

Validate errors are caused when the Validator has a problem.

[edit]
Added link.
[/edit]

And the third case is an Invalid result, which shows as Outcome: Success rather than Outcome: Validate error. Such Invalid results can be caused by the conditions that Tony described, but have nothing to do with the topic of this thread.

OTOH, Sutaru found a case where a Validate error can be caused by the host sending garbled data when it tries to upload. The Upload handler replies with a "Bad Command" hard error which makes the BOINC core client delete the failed upload, leading to the Validate error after the result is reported.

                                                                Joe

I have had two more validate errors as posted below. Do you think they will run the corrective scrip again?

652486842 176317244 6 Nov 2007 3:26:14 UTC 7 Nov 2007 8:54:46 UTC Over Validate error Done 18,969.27

653125251

---

647655998

---

Just got five validation errors - first ones I've had in ages. Running BOINC 5.10.20 and Alex's V8 on MBP Core Duo. WUs are

http://setiathome.berkeley.edu/resul...ltid=645225100
http://setiathome.berkeley.edu/resul...ltid=645085501
http://setiathome.berkeley.edu/resul...ltid=644889888
http://setiathome.berkeley.edu/resul...ltid=642986388
http://setiathome.berkeley.edu/resul...ltid=642986386

Any ideas?

C

---

Join Team MacNN

They must of ran the corrective script again, or corrected the problem. Last night I noted I had more validate errors but was too lazy to post them here. After just checking, they were all gone except for two, but with those, the computer that replaced me had sent in their results to complete the pairing.

Thanks goes out to Eric and ...

I have had at least one of these errors every day this month and so far today, (knock on wood), none.

Kevin

---

WORK HARD! PLAY HARD!

They must of ran the corrective script again, or corrected the problem.

The script was definitely run again. The other day I had...ummm...abused my BOINC buttons (joining hiamps, I suppose) by hitting update within 15 seconds of the result being uploaded and it generated a validate error. That result has since been "fixed" and I have credit for it... I had never reported in this thread that I had a validate error, so the script goes out and finds everything matching the criteria...

Brian

---

Here is an additional one - 651162153 PS - It's on 2.2B, please adjust appropriately. Thanks.

---

Here is an additional one - 651162153 PS - It's on 2.2B, please adjust appropriately. Thanks.

Good boy Philly...:)

I hope that no more validation errors you get. I have not had more since last weekend (for the moment). What BOINC version are you using? I recommend the 5.10.28. I'm using it since last week and all works fine...

[edit]Oh! you are using 5.10.13... If you updates to 5.10.28 you can improve the validation errors problems... Believe me:)[/edit]

Best Regards.

---

Logan.

BOINC FAQ Service (Ahora, también disponible en Español/Now available in Spanish)

Here is an additional one - 651162153 PS - It's on 2.2B, please adjust appropriately. Thanks.

Good boy Philly...:)

I hope that no more validation errors you get. I have not had more since last weekend (for the moment). What BOINC version are you using? I recommend the 5.10.28. I'm using it since last week and all works fine...

[edit]Oh! you are using 5.10.13... If you updates to 5.10.28 you can improve the validation errors problems... Believe me:)[/edit]

Best Regards.

When I'm able to get over to the location that computer is at, I'll do that and update the optimized version.

---

another one today http://setiathome.berkeley.edu/result.php?resultid=648220076

thanks les

[edit]Oh! you are using 5.10.13... If you updates to 5.10.28 you can improve the validation errors problems... Believe me:)[/edit]

v5.10.13 is fine - I use it on all my boxes.

But if you stick with it, you need to take the 'Computer is connected to the Internet about every' value above zero: 0.01 days seems plenty from here.

But don't let me take away your excuse to travel across town and upgrade that 2.2B......

[edit]Oh! you are using 5.10.13... If you updates to 5.10.28 you can improve the validation errors problems... Believe me:)[/edit]

v5.10.13 is fine - I use it on all my boxes.

But if you stick with it, you need to take the 'Computer is connected to the Internet about every' value above zero: 0.01 days seems plenty from here.

But don't let me take away your excuse to travel across town and upgrade that 2.2B......

You are right, Richard. 5.10.13 works fine and 5.10.20 too.

(if the 'Computer is connected to the Internet about every' is not 0)

---

Logan.

BOINC FAQ Service (Ahora, también disponible en Español/Now available in Spanish)

Snagged my very first one here. It's almost a relief - I was feeling quite left out :).

Guess it is because of the following:

11/11/2007 03:39:16|SETI@home|[file_xfer] Started upload of file 15fe07ab.3826.2526.13.6.109_1_0
11/11/2007 03:39:16|SETI@home|Sending scheduler request: To fetch work
11/11/2007 03:39:16|SETI@home|Requesting 3226 seconds of new work, and reporting 2 completed tasks
11/11/2007 03:39:21|SETI@home|[file_xfer] Finished upload of file 15fe07ab.3826.2526.13.6.109_1_0
11/11/2007 03:39:21|SETI@home|[file_xfer] Throughput 10665 bytes/sec
11/11/2007 03:39:21|SETI@home|Scheduler RPC succeeded [server version 511]
11/11/2007 03:39:21|SETI@home|Deferring communication for 11 sec
11/11/2007 03:39:21|SETI@home|Reason: requested by project


And its next report of completed tasks was only 20 seconds later at 03:39:36.

F.

---

Another validate error here 651162057

---

http://setiathome.berkeley.edu/result.php?resultid=653793219
new validate error today for me

---

E gads, I finally got a couple !
wuid=176567852
wuid=176528138

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.