10 errors, #11 sent out.?.
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=3819437

That is interesting ... I think that is known in technical circles as "Oops!"

At least it has the 'too many errors' status now that the 11th copy has errored out.

---

BOINC WIKI

BOINCing since 2002/12/8

cool.
For some reason 11 crossed my mind after I had posted; 10 error + 1 quorum.
- the fun of programming :)

sorta like:
a = 0
if a > 10 goto...
else...
return

that will run 11 times.

and i remember the days when you could do the:
else...
before the:
if a = 10 goto...

which would also run 11 times.

---

-----------------------
Click to see my tag
My tag
SNAFU'ed? Turn the Page! :D

For some reason 11 crossed my mind after I had posted; 10 error + 1 quorum. - the fun of programming :)

For i=0 to 10.... I think that's what's happening here. The very first WU sent out is number 0, not number 1 (have a look at the last number of the WU name on a work unit in your cache) I'm no programmer, but javascript arrays start with element 0 as well.

---

*** Join BOINC@Australia today ***

You get the cookie! :)
The first is 0, the last is 10. Good go!

https://boinc.bakerlab.org/rosetta/result.php?resultid=4513558

https://boinc.bakerlab.org/rosetta/result.php?resultid=6336470

I love to have answers.

---

-----------------------
Click to see my tag
My tag
SNAFU'ed? Turn the Page! :D

For i=0 to 10.... I think that's what's happening here. The very first WU sent out is number 0, not number 1 (have a look at the last number of the WU name on a work unit in your cache) I'm no programmer, but javascript arrays start with element 0 as well.

The tests for error-limits is >

This example means that in SETI@Home that has "Max success-results" = 5, if also the 6th result fails validation with "no consensus yet" does the Validator mark wu as errored-out.

For too many errors and too many total results, this job is done by the Transitioner.

This example means that in SETI@Home that has "Max success-results" = 5, if also the 6th result fails validation with "no consensus yet" does the Validator mark wu as errored-out.

Assuming that this is a question. The answer is yes. The point being that if a work unit is "unstable" in the sense that it returns wildly different answers, then it is not likely we will get good science out of it. Or, that it needs a special look see.

So, rather than send it out time after time, we quit even though it is not "blowing up" because the work unit is likely "bad".

Really bad work units are likely to cause the application to client error, and we test for that at some upper limit.

And, we test for a totality of succeses and failures to make sure that we don't stay below each of the limits but can't seem to get traction on the work unit.

All of this has to be decided by the project and set to reasonable limits before they get underway ...

---