I have more than a few validate errors on WUs that were returned during SETI's recent 'unable to upload' episode on 28/29 Dec. I'm sure these WUs, mine and the others, were good WUs that the validator screwed up. Are they lost forever? Or can they be re-validated. I just hate to loose credit for what I'm sure is good work.

106359076
106471080
106534823
106539786
106540972
106569867

I'm not so sure it was the validator which messed up. There doesn't seem to be a claimed credit amount, which is telling me there is no result return data file on record to validate with. So there wouldn't be a way to just tell the validator to try again.

My guess is the project may make a manual correction for this depending on how widespread it turns out to be.

Alinator

I had two validate errors after last nights upload problem. Both WUs had two others get the same error.

106458463
106374182


EDIT: there is claimed credit within the result IDs.

---

Calm Chaos Forum...Join Calm Chaos Now

Everything I returned on 29/12 has come up with validate error. Where the other results in those work units were returned on another date, they have validated ok. Where the other results were returned on 29/12 they have also failed to validate.

Neil

---

There doesn't seem to be a claimed credit amount

Look at any of the result IDs, rather than the WU ID, there is a returned credit claim. I'm sure there are a lot of returned results during that time period (just before midnight Pacific Time till about 6 AM) that people haven't found yet in their results because it's a holiday week and they haven't looked.

Ahhh, OK. I was doing something else at the same time and didn't look through listings thoroughly.

Still, IIRC the last time I had a straight validate error it still showed the claimed credit on the results page for that host. What I was thinking was the actual data file was lost or the DB link to it got broken. I'm hypothesizing the validator couldn't verify the claimed figure in the report with what was in the upload file, which was why it didn't show it on the results screen but did on the RID detail page.

In any case, events like this usually require Admin intervention to make a correction for the lost credit.

Alinator

I had only one WU with validation error. As I looked, there were three others with validation errors. a fouth one got through on dec 30 with success and there are three 'in progress'.
I hope these validatio errors will be corrected and the 61.23 creds ackowledged.

---

Veni,vidi,Foetsie
(I came, I saw, I was gone)

Indeed, odd failures.

Check out this workunit:

105979872

---

Doubtless related to the current server problems, I’m seeing validate errors on the WUs reported today.

Something for the BOINC wish-list: split the Suspend button in the Projects tab into two—as in the Activity menu—one to suspend processing and one to suspend network activity. That way one could prevent a project app from uploading when the servers are misbehaving, without affecting other projects, but still continue crunching any cached WUs.

---

Doubtless related to the current server problems, I’m seeing validate errors on the WUs reported today.

Something for the BOINC wish-list: split the Suspend button in the Projects tab into two—as in the Activity menu—one to suspend processing and one to suspend network activity. That way one could prevent a project app from uploading when the servers are misbehaving, without affecting other projects, but still continue crunching any cached WUs.

I think I would like to refine that into suspending network activity by project rather than a blanket suspend.
So that you could keep crunching WU's in your cache but not upload or report until the server probs are over but you can download etc from other projects.

---

I'm not so sure it was the validator which messed up. There doesn't seem to be a claimed credit amount, which is telling me there is no result return data file on record to validate with. So there wouldn't be a way to just tell the validator to try again.

My guess is the project may make a manual correction for this depending on how widespread it turns out to be.
Alinator

Here is another guy that has the same problem. Within each unit are the requested credit numbers but they don't seem to be coming out in to the chart and therefore getting no credits.

---

Something for the BOINC wish-list: split the Suspend button in the Projects tab into two—as in the Activity menu—one to suspend processing and one to suspend network activity. That way one could prevent a project app from uploading when the servers are misbehaving, without affecting other projects, but still continue crunching any cached WUs.

I think I would like to refine that into suspending network activity by project rather than a blanket suspend.

That’s exactly what I meant: you can already suspend all network activity, from the menu, but at present there’s no way to do that for individual projects short of suspending them, processing and all.

---

I saw two on my machines today. It's a Berkeley problem. My hopes are not high that they will be reevaluated and time recovered. They're wasted forever - 12,000 secondsworth (on one core). But, with a club there are no guarantees. It's not like an Apple or a Wal-Mart. Sometimes misfortunes happen with Seti and Boinc.

---

I can only add that I see 11 validate errors on my two machines since 8 Jan. These look like good files. i suppose they were somehow lost on the server.

---