For some odd reason for the last few days, completed work unit results given granted credit have been disappearing immediately!!! Has this been happening for anyone else?

From the Technical News page:

March 30, 2006 - 22:30 UTC
Okay. We seem to be out of the woods for now. We vastly reduced the size of the result table in the BOINC database. How? By clearing out the results-to-delete queue, and by cranking up db_purge so that it removes all results from the database as soon as they are deleted from disk. Under normal circumstances we keep at least a days' worth of old result rows around so participants can see those recently finished in their personal lists. Over the weekend we'll keep this clamped down for observational and catch-up purposes. Next week we might relax the db_purge parameters to allow completed result rows to linger like before.

MJ

---

thanks... I hope this gets corrected soon because we're losing CPU time spent data records needed for justifying the reasons why systems are kept powered on. The ultimate reasoning is that the lesser anyone is able to prove time spent, for some people running SETI, all their systems will be shut down. That includes me with only 2 systems and other silent people with a few hundred other systems... :)

In the next week, there might be fewer users. :)

In 7 more days with 3 more missing records for me; it's farewell from me! :| (And I check constantly at least once in 12 hours.)

thanks... I hope this gets corrected soon because we're losing CPU time spent data records needed for justifying the reasons why systems are kept powered on. The ultimate reasoning is that the lesser anyone is able to prove time spent, for some people running SETI, all their systems will be shut down. That includes me with only 2 systems and other silent people with a few hundred other systems... :)

Have you thought of using one of the programs that will log your work locally? BOINCView and BOINCLogX are two I've read discussed positively. Though it's not working right this minute, there's a page of BOINC add-on software which might give you what you need.

MJ

---

No, but I'll check those progies out... as soon as that page works via that link. BETTER WORK BY NEXT MONDAY! lol :)

Hi folks...

Having the same prob as Mariah... The odd thing is, that my granted credits are deducted as well... I only had the problem once yet, but i'm with boinc for only a week now...
In addition to that, i had some odd experience with pending credits as well... maybe someone can give me an explanation for that as well... the problem is the following: i have 4 results which are pending for some time now... i know, that i have to wait for some time until they are granted, but it happens nothing... but the real strange part is, that i had credits granted already which became pending again... how does that explain?
the next thing is, that i thought the granted credits just get added to a total which will be your actual total... i might be a little nerd, but i designed an excel sheet to count up my credits and to do my own statistics... the values given by seti@home and boinc just do not fit... is there anything i might have forgotten?

Thanks for replying,

Sinner

---

BoincLogX is available at http://www.bjoernhenke.de/index_e.htm.

---

Yeah, fine... so boinc logx is there and there... still doesn't explain or solve my probs... its just a monitoring tool... but i monitored it myself already... don't need a tool for that...
but anyway, my statistics tell me that my credits, that were deducted early this morning, are back on my account again... weird :)...
so i don't have any problem left except for the unanswered question why credits get granted and then pending again... but i guess that's something noone can really explain so i just take it as a fact...
gonna test logx for some time now... maybe it is usefull... let's see...

---

thanks... I hope this gets corrected soon because we're losing CPU time spent data records needed for justifying the reasons why systems are kept powered on. The ultimate reasoning is that the lesser anyone is able to prove time spent, for some people running SETI, all their systems will be shut down. That includes me with only 2 systems and other silent people with a few hundred other systems... :)

In the next week, there might be fewer users. :)

In 7 more days with 3 more missing records for me; it's farewell from me! :| (And I check constantly at least once in 12 hours.)

If you're looking for "proof of work done" that is in credits.

If you are looking at your results page, well, those are constantly purged -- they take up space in the database, and for the most part, they just let you see how credit was granted for your recent work.

Credits are not purged, they continue to tally.

---

The splitter make a wu, then store it on the HD, then that WU is made into four results for transmission to our hosts. Hopefully, those 4 return their results and they get examined to make a "canonical result", that result is written to the master science data base for keeps. The results from our hosts are no longer useful or necessary, they are purged after a bit.

... the problem is the following: i have 4 results which are pending for some time now... i know, that i have to wait for some time until they are granted, but it happens nothing... but the real strange part is, that i had credits granted already which became pending again... how does that explain?

Are you sure they were the same results that seemed to have become pending after validation? Do you make a note of the WU or result ID numbers, or the WU names, when you collect your statistics? The reason I ask is that the project's work database is being purged of completed results more rapidly than normal these days, as mentioned upthread, in an effort to mitigate the overloading problems they've been having. A consequence of this is that validated work may have been removed from your results page and replaced by more recent data without your noticing. If your total credit didn't go down—it never should!—then I think the most likely explanation is that the pending WUs were a different batch from the ones you saw granted credit earlier.

---

Well, i didn't watch the id's, but i knew the figures from the claimed credits column... that's how i identified th WU... and since i get 3 to 4 results a day, i can't imagine i mixed up s.th. and credits didn't really go down, it was like:
you have 13 credits, WU-A is worth 3 credits, WU-B 4 credits and WU-C is 5 credits... A is granted already, B and C are not... now you update your figure and you have 19 credits now, which is 13-3+4+5... there is no other way to get 19 credits than deducting 3 cradits again, since everything else would give 17, 18 or 22 credits... the next day you check your credits and you have 22, which is the 19+3...
the figures were just examples but maybe you understood now...
but anyhow, i keep an eye on it, so next time it happens i can tell you more...

Sinner

---

Well, i didn't watch the id's, but i knew the figures from the claimed credits column... that's how i identified th WU... and since i get 3 to 4 results a day, i can't imagine i mixed up s.th. and credits didn't really go down, it was like:
you have 13 credits, WU-A is worth 3 credits, WU-B 4 credits and WU-C is 5 credits... A is granted already, B and C are not... now you update your figure and you have 19 credits now, which is 13-3+4+5... there is no other way to get 19 credits than deducting 3 cradits again, since everything else would give 17, 18 or 22 credits... the next day you check your credits and you have 22, which is the 19+3...
the figures were just examples but maybe you understood now...
but anyhow, i keep an eye on it, so next time it happens i can tell you more...

Sinner

You're comparing apples and oranges: your results page only shows recent work, not every work unit you've ever done.

At the moment, this table is being purged a little faster than in the past, to ease the load on the database.

As I look at your stats, you have 182 credits, but only 60 of those are still shown on the results page -- you didn't lose the other 122 credits, they've been moved to the Science database.

---

Another instance... I just lost result ID: 286154817 that I had gotten about 14 hours ago, recorded initially about 12 hours ago, and now when I'm able to record everything this time of day, that result ID (one of many) is now GONE!!!! >:-(

This non-sense cannot be kept up! >:-{

This non-sense cannot be kept up! >:-{

From the Technical news page:
March 30, 2006 - 22:30 UTC
Okay. We seem to be out of the woods for now. We vastly reduced the size of the result table in the BOINC database. How? By clearing out the results-to-delete queue, and by cranking up db_purge so that it removes all results from the database as soon as they are deleted from disk. Under normal circumstances we keep at least a days' worth of old result rows around so participants can see those recently finished in their personal lists. Over the weekend we'll keep this clamped down for observational and catch-up purposes. Next week we might relax the db_purge parameters to allow completed result rows to linger like before.

They are purging quickly.

---

My movie https://vimeo.com/manage/videos/502242

Another instance... I just lost result ID: 286154817 that I had gotten about 14 hours ago, recorded initially about 12 hours ago, and now when I'm able to record everything this time of day, that result ID (one of many) is now GONE!!!! >:-(

This non-sense cannot be kept up! >:-{

You are absolutely correct, this nonsense cannot continue.

You did not lose result 286154817. You still have the granted credits for that work unit.

The work unit has transitioned to the science database, and is no longer needed in the BOINC database for SETI@HOME.

Edit: and actually, purging of completed results not only can be kept up, but must be. If not, the database will drag the project to a halt.

We can only try to convey the facts: finished work is not supposed to show forever on your results page. Expecting it to do so just does not make sense.

---

This non-sense cannot be kept up! >:-{

Reduce your connect to network option to 0.01 days so that you will be one of the first to return the results, instead of last as at the moment.

This non-sense cannot be kept up! >:-{

Reduce your connect to network option to 0.01 days so that you will be one of the first to return the results, instead of last as at the moment.

... but the real problem isn't that the results are last instead of first, the real problem is the mistaken belief that the units are "lost" because they are no longer on the results page.

Reducing the cache will keep them there longer, but they're still going to be purged (moved to the Science DB and removed from the SETI@HOME DB) sooner or later.

... and this participant is upset because they are being purged.

---

... and this participant is upset because they are being purged.

Then he needs to shut up, read all he can, and try and learn something, before opening his mouth.