Folding Forum

PFM wrote:Any ETA on this ?
uploads failing for me too on this server....

Unless an uptime was mentioned in the Discord post that Neil-B saw, I have not seen an ETA. The server is one donated to F@h use, so the maintenance may be on that company's schedule and depend on factors not provided to F@h or possibly not to be posted publicly.

IP resolves to Beanfield.com Toronto
maybe Monday... my WU has timed out.

Both of my rigs have been trying to upload to this server since fri or sat. Came to drop notification but good to see someone else already has. Fold on....

Who is in contact with Joseph? Now my unsent WU is being actively rejected. 06:10:09:WARNING:WU02:FS00:Exception: Failed to send results to work server: Failed to connect to 206.223.170.146:80: No connection could be made because the target machine actively refused it.

tbh it makes no difference whether actively rejecting or not ... the server is down for maintenance and until it is up again an WUs won't be uploaded ... it looks as if this one may well continut to the point that the WUs expire - the client will handle the dumping of those ... it is a pain, but it happens occasionally - the researchers feel it even more than the folders do.

Hi there,

I suggest that the traffic to a down server to be redirected to a new server.

I don't know how is the infrastructure of the folding@home, but i think there are some ways to transfer the traffic to another server in case of issues.

You can use in client optional server in case the first didn't connect. Or, you can use a gateway that redirect the data to another server in case of failure.

I think that loose all the processed data in case of server issues is a lot of energy and time lost.

Best regards,

my data's only 17 MB. If the server's not back up on the 25th is there somewhere I can upload/email it?

megs.rs wrote:Hi there,

I suggest that the traffic to a down server to be redirected to a new server.

I don't know how is the infrastructure of the folding@home, but i think there are some ways to transfer the traffic to another server in case of issues.

You can use in client optional server in case the first didn't connect. Or, you can use a gateway that redirect the data to another server in case of failure.

I think that loose all the processed data in case of server issues is a lot of energy and time lost.

Best regards,

You speak from ignorance, what you suggest as possible is not doable with the current infrastructure for F@h. It would require major changes to both the client and server code to handle this.

Knish wrote:my data's only 17 MB. If the server's not back up on the 25th is there somewhere I can upload/email it?

The issue isnt you sending it by a certain time .. it is if the server doesn't receive it .. if the server isnt back up by the time wu reaches the expiration deadline then it no longer has any scientific value .. sorry but that is simply how the infrastructure and server coding works .. even if there was a CS the WU would still need to get to the WS from there by the expiration

It's pinging OK, but still not uploading.

I sent a query about this server's status. It was running out of space repeatedly, the data is being cleaned up. Tentatively it will be back in operation by next week.

Just had my WU from 11/20 upload successfully. Only 603 points but the results made it before the expiration.

If I may interject a thought for everyone to think about. In these unpressedented time we have managed to accomplish a amount of computer processing that could hanve never been thought of before. The fact that F@H has managed to get additional servers up and running in conjuction with the researchers is not as easy as everyone would think. Much less the scientist are not doubt overwhelmed with the influx of a amazing mountain of data that I'm sure they never thought possiable either. I for one tip my hat to the admins, and others with F@H for continuing their efforts and we all need to be patient and stay the course.

megs.rs wrote:I suggest that the traffic to a down server to be redirected to a new server.

I don't know how is the infrastructure of the folding@home, but ...

What your suggesting is normally true. The data is redirected to a designated server (called a Collection Server, in FAH's parlance) and that secondary server is configured so that the WU is forwarded to the original Work Server whenever service there is restored. It this particular case, no CS has been designate or it, too, has failed. Sending it to a third or fourth sever isn't practical as nobody is going to manually check for WUs that get scattered around FAH's collection of servers. A second level of redundancy is only needed occasionally. What I do not know is why that level is not assigned.

megs.rs wrote:I suggest that the traffic to a down server to be redirected to a new server.

I don't know how is the infrastructure of the folding@home, but ...

What your suggesting is normally true. The data is redirected to a designated server (called a Collection Server, in FAH's parlance) and that secondary server is configured so that the WU is forwarded to the original Work Server whenever service there is restored. It this particular case, no CS has been designate or it, too, has failed. Sending it to a third or fourth sever isn't practical as nobody is going to manually check for WUs that get scattered around FAH's collection of servers. A second level of redundancy is only needed occasionally.

What I do not know is why that level is not assigned. A wild guess is that this donated server is not fully integrated into FAH's back-end networking.