Folding Forum

Opps! hahaha I meant that the UTC timezone in my FAH settings are incorrect. The time now is 1.06pm for me.

Oops. But my advice still stands. take a nice afternoon nap and stop worrying about it for a while.

Please note that this has been reported again and will be investigated by the team. In the meantime, please leave your client running as it will automatically try again and hopefully, will be uploaded before the expiration

I had a discussion with the folks who support this server and MAYBE we figured out something else to try.

Apparently server AWS1 reaches a capacity limitation (probably bandwidth) and it simply can't handle all the uploads that the clients are throwing at it. First upload speed decays. Then it starts having a variety of errors. It's not really clear what the choke-point is but there is one.

The recent fix in the new client to limit the exponential increases in retries actually was the wrong thing to do but it's not a big factor.

Anyway, the server may be saturated many hours every day, but there still are some gaps and that gives it a chance to clean up the backlog that's trying to find its way home. The words are Be Patient.

bruce wrote:I had a discussion with the folks who support this server and MAYBE we figured out something else to try.

Apparently server AWS1 reaches a capacity limitation (probably bandwidth) and it simply can't handle all the uploads that the clients are throwing at it. First upload speed decays. Then it starts having a variety of errors. It's not really clear what the choke-point is but there is one.

The recent fix in the new client to limit the exponential increases in retries actually was the wrong thing to do but it's not a big factor.

Anyway, the server may be saturated many hours every day, but there still are some gaps and that gives it a chance to clean up the backlog that's trying to find its way home. The words are Be Patient.

Thanks Bruce, I know a lot of people are concerned about points so why not have the client write a checkpoint for points in the results so that upload issues do not affect points allocation?

Obviously a suggestion going forward

Continuing issues with this evil server. Uploaded a work unit and received a credit estimate 8 hours ago. No change on the FAH side. The worst bit is I've just finished another for this server, and you guessed it, the next is for this one as well!

I'm not the one that sets the policy, but I doubt your local checkpoint of points will pass muster.

Suppose you finish a WU and it can't upload. You wish to blame that delay on FAH. (Yes, I understand) Ok. new scenario. You finish a WU but you wait a few days before returning it. You can't blame FAH for that delay. How does the software know which scenario was applicable?

Before the COVID surge, FAH servers were on the campus of the universities. They were readily available with a backup called a Collection Server. The universal answer to the previous scenario was :it doesn't matter because one of the servers will go through quickly except in very rare scenarios and if it has to retry, it will only be a few minutes before that happens.

Just an update, the work done is wasted as it exceeded the expiration time.

SirTj wrote:Just an update, the work done is wasted as it exceeded the expiration time.

Code: Select all

01:10:21:WU01:FS01:Sending unit results: id:01 state:SEND error:NO_ERROR project:14448 run:0 clone:20 gen:19 core:0x22 unit:0x0000001e03854c135ea7b88f47788401
01:10:21:WU01:FS01:Uploading 95.13MiB to 3.133.76.19
01:10:21:WU01:FS01:Connecting to 3.133.76.19:8080
01:11:57:WU01:FS01:Upload 2.37%
01:11:57:WARNING:WU01:FS01:Exception: Failed to send results to work server: Transfer failed
02:51:12:WARNING:WU01:FS01:Past final deadline 2020-05-27T02:51:11Z, dumping
02:51:12:WU01:FS01:Cleaning up

Sucks, had that happen to a few as well