Actually, nobody comes around to pick up the work - your BOINC client is supposed to automatically send it in :).
I've just had a look at the results lists for your two machines. It looks like virtually all your work on one machine is ending immediately with a client error. These are being reported quite quickly, but because of the error, there will be no credit. The errors were over a week ago so what is happening with the few results you still have? If they are completed, try manually updating the project by selecting the project and clicking "update".
On the other machine you also had a client error but the most recent result looked like it was good and should have got you credit. However it says that you aborted the transfer of the completed result?? Why did you do that? It was just over the deadline but you would have got credit if you had allowed the transfer to complete.
There have been quite a few client errors recently which have been attributed to bugs in the science app that are now hopefully fixed. New apps are being beta tested at the moment so hopefully the error rate should diminish when the new apps are released.
I'm wondering why he aborted the transfer on the machine running 32 bit Windows after apparently sucessfully completing the result? This guarantees no credit.
Also, the other machine is running 64 bit Windows and there is a failure of the app to initialize correctly. So for that case going back and reviewing the threads on getting EAH to run on 64 bit hosts (which is not officially supported yet) would be the first step there.
I'm wondering why he aborted the transfer on the machine running 32 bit Windows after apparently sucessfully completing the result? This guarantees no credit.
Also, the other machine is running 64 bit Windows and there is a failure of the app to initialize correctly. So for that case going back and reviewing the threads on getting EAH to run on 64 bit hosts (which is not officially supported yet) would be the first step there.
Alinator
I aborted the transfer as it wasnt transfering it manages to transfer 5% or less a day so it would have timed out like all the others
I aborted the transfer as it wasnt transfering it manages to transfer 5% or less a day so it would have timed out like all the others
The amount of data to transfer is about 1MB or so and shouldn't take all that long to be uploaded. If it only manages to transfer 5% per day, there is something drastically wrong with your link to the internet. Are you talking about a transfer rate or the calculation rate of the science app??
Another thing that may be causing you grief. Your first post in this thread shows your account as "Jim MacKay" ID=266871 who joined on June 13 2007. Your latest post shows "Jim MacKay" ID=99679 who joined on August 4 2005. Both your accounts have computers (probably the same computer) which have received work recently. I'm not sure what would happen if you received work under one ID and then tried to submit the results under the other ID. Maybe something like this is the cause of these Validate errors??
You should decide on one account to keep and abandon the other. Make sure you always download work and upload the completed results using that one account. Maybe your problems might then be resolved.
I'm not sure what would happen if you received work under one ID and then tried to submit the results under the other ID.
That's not possible in BOINC. The computer ID that downloaded the results should also upload the results. Another computer ID can't upload the results, it'll get an error in BOINC Manager.
Actually, I should have looked more closely. There are two different machines involved, a Core 2 Duo and an AMD64 dual core. The C2D is registered under both names but the activity periods don't overlap. The C2D was first registered under the "newer" account on June 30 2007 and had its last contact on July 18 (it has no current work). It then appeared under the "older" account on July 25 and still has current work.
The AMD64 is only on the "newer" account and also has current work. I was puzzled as to how there could be two sets of current work on the same machine under two different accounts. Now that I see there are actually two different machines and there is no "results overlap" the problem is obviously not what I was imagining.
Actually, I should have looked more closely. There are two different machines involved, a Core 2 Duo and an AMD64 dual core. The C2D is registered under both names but the activity periods don't overlap. The C2D was first registered under the "newer" account on June 30 2007 and had its last contact on July 18 (it has no current work). It then appeared under the "older" account on July 25 and still has current work.
The AMD64 is only on the "newer" account and also has current work. I was puzzled as to how there could be two sets of current work on the same machine under two different accounts. Now that I see there are actually two different machines and there is no "results overlap" the problem is obviously not what I was imagining.
Just to clarify the problem I still have with Enistein I down load a work unit, I crunsh it,it moves to the "Transfers" in Boinc manager and the progress gos up at about 5% a day then after a futher day or two disapears .It doesnt happen to any other boinc projects I run only Enistein I have tried it on diffrent PC and its the same.
Just to clarify the problem I still have with Enistein I down load a work unit, I crunsh it,it moves to the "Transfers" in Boinc manager and the progress gos up at about 5% a day then after a futher day or two disapears .It doesnt happen to any other boinc projects I run only Enistein I have tried it on diffrent PC and its the same.
Hi Jim,
I'm not sure I properly understand your problem.
Are you saying that a task crunches normally and then the result uploads at the rate of about 5% per day so that it might take around 20 days to complete the upload? Perhaps you mean that the upload starts but seems to get stuck at around 5% with no further progress happening for a long time? Then at some later stage you discover that the upload has finally cleared and the operation has disappeared from the transfers tab in BOINC Manager?
The second scenario is more understandable as a temporary network difficulty could cause BOINC to backoff in attempting the transfer by as much as a day or so under certain conditions. However eventually the transfer will complete as the timeout expires and BOINC retries the operation. This is not really a problem - it is how BOINC is designed to handle potentially unreliable networks.
Also, I've just investigated further and have noticed that you still have two accounts, an older one you are using for posting messages but not using for crunching EAH and a newer one which has an active machine attached to it. So this is why it looks like you haven't been active for quite a while. Your newer account, which you weren't logged in with for your most recent message, actually has a machine that has two very recent tasks on board. I guess one of these tasks has completed and the upload of the result is stuck? If so, a timeout period on the transfers tab will be counting down before an automatic retry is attempted. How long is that timeout period? Is the machine continuing to retry on its own from time to time? What happens if you select the stuck upload and click the "retry now" button?
Just to clarify the problem I still have with Enistein I down load a work unit, I crunsh it,it moves to the "Transfers" in Boinc manager and the progress gos up at about 5% a day then after a futher day or two disapears .It doesnt happen to any other boinc projects I run only Enistein I have tried it on diffrent PC and its the same.
Hi Jim,
I'm not sure I properly understand your problem.
Are you saying that a task crunches normally and then the result uploads at the rate of about 5% per day so that it might take around 20 days to complete the upload? Perhaps you mean that the upload starts but seems to get stuck at around 5% with no further progress happening for a long time? Then at some later stage you discover that the upload has finally cleared and the operation has disappeared from the transfers tab in BOINC Manager?
The second scenario is more understandable as a temporary network difficulty could cause BOINC to backoff in attempting the transfer by as much as a day or so under certain conditions. However eventually the transfer will complete as the timeout expires and BOINC retries the operation. This is not really a problem - it is how BOINC is designed to handle potentially unreliable networks.
Also, I've just investigated further and have noticed that you still have two accounts, an older one you are using for posting messages but not using for crunching EAH and a newer one which has an active machine attached to it. So this is why it looks like you haven't been active for quite a while. Your newer account, which you weren't logged in with for your most recent message, actually has a machine that has two very recent tasks on board. I guess one of these tasks has completed and the upload of the result is stuck? If so, a timeout period on the transfers tab will be counting down before an automatic retry is attempted. How long is that timeout period? Is the machine continuing to retry on its own from time to time? What happens if you select the stuck upload and click the "retry now" button?
At the precent time I have 2 w/us completed and sitting in the "Transfers" tab one at 100.00% the other at 60.30% if I select the one at about 60% and press "retry now" 5 or 6 times I can get the % to go up by about 0.6% ie its now at 60.99% in the "message" tab I get the messages "20/11/2007 07:08:22|Einstein@Home|Started upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:26||Project communication failed: attempting access to reference site
20/11/2007 07:08:26|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__26_S5R3a_0_0: http error
20/11/2007 07:08:26|Einstein@Home|Backing off 2 hr 22 min 43 sec on upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:28||Access to reference site succeeded - project servers may be temporarily down.
It I use retry now on the other w/u I get the following messages "20/11/2007 07:18:49|Einstein@Home|Started upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:51||Project communication failed: attempting access to reference site
20/11/2007 07:18:51|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__57_S5R3a_3_0: http error
20/11/2007 07:18:51|Einstein@Home|Backing off 2 hr 22 min 14 sec on upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:52||Access to reference site succeeded - project servers may be temporarily down."
I was exspecting the w/u in transfer at 100% to vanish but it hasn't.
I going to leave these 2 w/us as they are and see what happens but it looks to me like I have a problem with my account getting permision to upload data I know sometimes w/us can be a day or two before they transfer but thats because of server down time. I have tried different PCs different OSs fresh down loads of Boinc about the only thing that it can be is a problem with the account.
Just to clarify the problem I still have with Enistein I down load a work unit, I crunsh it,it moves to the "Transfers" in Boinc manager and the progress gos up at about 5% a day then after a futher day or two disapears .It doesnt happen to any other boinc projects I run only Enistein I have tried it on diffrent PC and its the same.
Hi Jim,
I'm not sure I properly understand your problem.
Are you saying that a task crunches normally and then the result uploads at the rate of about 5% per day so that it might take around 20 days to complete the upload? Perhaps you mean that the upload starts but seems to get stuck at around 5% with no further progress happening for a long time? Then at some later stage you discover that the upload has finally cleared and the operation has disappeared from the transfers tab in BOINC Manager?
The second scenario is more understandable as a temporary network difficulty could cause BOINC to backoff in attempting the transfer by as much as a day or so under certain conditions. However eventually the transfer will complete as the timeout expires and BOINC retries the operation. This is not really a problem - it is how BOINC is designed to handle potentially unreliable networks.
Also, I've just investigated further and have noticed that you still have two accounts, an older one you are using for posting messages but not using for crunching EAH and a newer one which has an active machine attached to it. So this is why it looks like you haven't been active for quite a while. Your newer account, which you weren't logged in with for your most recent message, actually has a machine that has two very recent tasks on board. I guess one of these tasks has completed and the upload of the result is stuck? If so, a timeout period on the transfers tab will be counting down before an automatic retry is attempted. How long is that timeout period? Is the machine continuing to retry on its own from time to time? What happens if you select the stuck upload and click the "retry now" button?
At the precent time I have 2 w/us completed and sitting in the "Transfers" tab one at 100.00% the other at 60.30% if I select the one at about 60% and press "retry now" 5 or 6 times I can get the % to go up by about 0.6% ie its now at 60.99% in the "message" tab I get the messages "20/11/2007 07:08:22|Einstein@Home|Started upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:26||Project communication failed: attempting access to reference site
20/11/2007 07:08:26|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__26_S5R3a_0_0: http error
20/11/2007 07:08:26|Einstein@Home|Backing off 2 hr 22 min 43 sec on upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:28||Access to reference site succeeded - project servers may be temporarily down.
It I use retry now on the other w/u I get the following messages "20/11/2007 07:18:49|Einstein@Home|Started upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:51||Project communication failed: attempting access to reference site
20/11/2007 07:18:51|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__57_S5R3a_3_0: http error
20/11/2007 07:18:51|Einstein@Home|Backing off 2 hr 22 min 14 sec on upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:52||Access to reference site succeeded - project servers may be temporarily down."
I was exspecting the w/u in transfer at 100% to vanish but it hasn't.
I going to leave these 2 w/us as they are and see what happens but it looks to me like I have a problem with my account getting permision to upload data I know sometimes w/us can be a day or two before they transfer but thats because of server down time. I have tried different PCs different OSs fresh down loads of Boinc about the only thing that it can be is a problem with the account.
Ok I have found it "I think" checked the account"S" and out was running Boinc manager on one and Einstein on the other don't you just love AutoFill, removing the tick from the box. Sorry for being a bit of a idiot wasn't until I was reading it through that the penny dropped and I rememberd setting up the other account.
Not picking up completed work units
)
Actually, nobody comes around to pick up the work - your BOINC client is supposed to automatically send it in :).
I've just had a look at the results lists for your two machines. It looks like virtually all your work on one machine is ending immediately with a client error. These are being reported quite quickly, but because of the error, there will be no credit. The errors were over a week ago so what is happening with the few results you still have? If they are completed, try manually updating the project by selecting the project and clicking "update".
On the other machine you also had a client error but the most recent result looked like it was good and should have got you credit. However it says that you aborted the transfer of the completed result?? Why did you do that? It was just over the deadline but you would have got credit if you had allowed the transfer to complete.
There have been quite a few client errors recently which have been attributed to bugs in the science app that are now hopefully fixed. New apps are being beta tested at the moment so hopefully the error rate should diminish when the new apps are released.
Cheers,
Gary.
Hmmmm..... I'm wondering
)
Hmmmm.....
I'm wondering why he aborted the transfer on the machine running 32 bit Windows after apparently sucessfully completing the result? This guarantees no credit.
Also, the other machine is running 64 bit Windows and there is a failure of the app to initialize correctly. So for that case going back and reviewing the threads on getting EAH to run on 64 bit hosts (which is not officially supported yet) would be the first step there.
Alinator
RE: Hmmmm..... I'm
)
I aborted the transfer as it wasnt transfering it manages to transfer 5% or less a day so it would have timed out like all the others
RE: I aborted the transfer
)
The amount of data to transfer is about 1MB or so and shouldn't take all that long to be uploaded. If it only manages to transfer 5% per day, there is something drastically wrong with your link to the internet. Are you talking about a transfer rate or the calculation rate of the science app??
Another thing that may be causing you grief. Your first post in this thread shows your account as "Jim MacKay" ID=266871 who joined on June 13 2007. Your latest post shows "Jim MacKay" ID=99679 who joined on August 4 2005. Both your accounts have computers (probably the same computer) which have received work recently. I'm not sure what would happen if you received work under one ID and then tried to submit the results under the other ID. Maybe something like this is the cause of these Validate errors??
You should decide on one account to keep and abandon the other. Make sure you always download work and upload the completed results using that one account. Maybe your problems might then be resolved.
Cheers,
Gary.
RE: I'm not sure what would
)
That's not possible in BOINC. The computer ID that downloaded the results should also upload the results. Another computer ID can't upload the results, it'll get an error in BOINC Manager.
Actually, I should have
)
Actually, I should have looked more closely. There are two different machines involved, a Core 2 Duo and an AMD64 dual core. The C2D is registered under both names but the activity periods don't overlap. The C2D was first registered under the "newer" account on June 30 2007 and had its last contact on July 18 (it has no current work). It then appeared under the "older" account on July 25 and still has current work.
The AMD64 is only on the "newer" account and also has current work. I was puzzled as to how there could be two sets of current work on the same machine under two different accounts. Now that I see there are actually two different machines and there is no "results overlap" the problem is obviously not what I was imagining.
Cheers,
Gary.
RE: Actually, I should have
)
Just to clarify the problem I still have with Enistein I down load a work unit, I crunsh it,it moves to the "Transfers" in Boinc manager and the progress gos up at about 5% a day then after a futher day or two disapears .It doesnt happen to any other boinc projects I run only Enistein I have tried it on diffrent PC and its the same.
RE: Just to clarify the
)
Hi Jim,
I'm not sure I properly understand your problem.
Are you saying that a task crunches normally and then the result uploads at the rate of about 5% per day so that it might take around 20 days to complete the upload? Perhaps you mean that the upload starts but seems to get stuck at around 5% with no further progress happening for a long time? Then at some later stage you discover that the upload has finally cleared and the operation has disappeared from the transfers tab in BOINC Manager?
The second scenario is more understandable as a temporary network difficulty could cause BOINC to backoff in attempting the transfer by as much as a day or so under certain conditions. However eventually the transfer will complete as the timeout expires and BOINC retries the operation. This is not really a problem - it is how BOINC is designed to handle potentially unreliable networks.
Also, I've just investigated further and have noticed that you still have two accounts, an older one you are using for posting messages but not using for crunching EAH and a newer one which has an active machine attached to it. So this is why it looks like you haven't been active for quite a while. Your newer account, which you weren't logged in with for your most recent message, actually has a machine that has two very recent tasks on board. I guess one of these tasks has completed and the upload of the result is stuck? If so, a timeout period on the transfers tab will be counting down before an automatic retry is attempted. How long is that timeout period? Is the machine continuing to retry on its own from time to time? What happens if you select the stuck upload and click the "retry now" button?
Cheers,
Gary.
RE: RE: Just to clarify
)
At the precent time I have 2 w/us completed and sitting in the "Transfers" tab one at 100.00% the other at 60.30% if I select the one at about 60% and press "retry now" 5 or 6 times I can get the % to go up by about 0.6% ie its now at 60.99% in the "message" tab I get the messages "20/11/2007 07:08:22|Einstein@Home|Started upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:26||Project communication failed: attempting access to reference site
20/11/2007 07:08:26|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__26_S5R3a_0_0: http error
20/11/2007 07:08:26|Einstein@Home|Backing off 2 hr 22 min 43 sec on upload of h1_0323.20_S5R2__26_S5R3a_0_0
20/11/2007 07:08:28||Access to reference site succeeded - project servers may be temporarily down.
It I use retry now on the other w/u I get the following messages "20/11/2007 07:18:49|Einstein@Home|Started upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:51||Project communication failed: attempting access to reference site
20/11/2007 07:18:51|Einstein@Home|Temporarily failed upload of h1_0323.20_S5R2__57_S5R3a_3_0: http error
20/11/2007 07:18:51|Einstein@Home|Backing off 2 hr 22 min 14 sec on upload of h1_0323.20_S5R2__57_S5R3a_3_0
20/11/2007 07:18:52||Access to reference site succeeded - project servers may be temporarily down."
I was exspecting the w/u in transfer at 100% to vanish but it hasn't.
I going to leave these 2 w/us as they are and see what happens but it looks to me like I have a problem with my account getting permision to upload data I know sometimes w/us can be a day or two before they transfer but thats because of server down time. I have tried different PCs different OSs fresh down loads of Boinc about the only thing that it can be is a problem with the account.
RE: RE: RE: Just to
)
Ok I have found it "I think" checked the account"S" and out was running Boinc manager on one and Einstein on the other don't you just love AutoFill, removing the tick from the box. Sorry for being a bit of a idiot wasn't until I was reading it through that the penny dropped and I rememberd setting up the other account.