171.64.65.56 ???

[314159]

DL on this server=0 and completed WUs are being queued.
However - each of my Quads that have completed over the past hour ARE being assigned work from this server (normally after a short delay).
Am anxiously awaiting to see if the one that was sent received proper credit.
I assume that someone is working on this? Peter?

What the heck happened to the alternative server (xx.54).
Have not see a WU from it all day.

[dutchmm]

The server had some problems over the weekend.

OK, no problems, so long as the results actually arrived OK ...

[toTOW]

I've just returned a WU to this server and got a new one ... it was slow, but it went fine :

Code: Select all

[21:06:45] + Attempting to send results [November 15 21:06:45 UTC]
[21:06:45] - Reading file work/wuresults_06.dat from core
[21:06:45]   (Read 43646612 bytes from disk)
[21:06:45] Connecting to http://171.64.65.56:8080/
[21:11:46] Posted data.
[21:11:46] Initial: 0000; - Uploaded at ~141 kB/s
[21:11:47] - Averaged speed for that direction ~1444 kB/s
[21:11:47] + Results successfully sent
[21:11:47] Thank you for your contribution to Folding@Home.
[21:11:47] + Number of Units Completed: 532

[21:11:50] Trying to send all finished work units
[21:11:50] + No unsent completed units remaining.
[21:11:50] - Preparing to get new work unit...
[21:11:50] Cleaning up work directory
[21:11:50] + Attempting to get work packet
[21:11:50] Passkey found
[21:11:50] - Will indicate memory of 12036 MB
[21:11:50] - Connecting to assignment server
[21:11:50] Connecting to http://assign.stanford.edu:8080/
[21:11:51] Posted data.
[21:11:51] Initial: 40AB; - Successful: assigned to (171.64.65.56).
[21:11:51] + News From Folding@Home: Welcome to Folding@Home
[21:11:52] Loaded queue successfully.
[21:11:52] Connecting to http://171.64.65.56:8080/
[21:11:53] Posted data.
[21:11:53] Initial: 0000; - Receiving payload (expected size: 764408)
[21:11:55] - Downloaded at ~373 kB/s
[21:11:55] - Averaged speed for that direction ~464 kB/s
[21:11:55] + Received work.

[314159]

Good for you. (probably a special deal for nice Super Moderators)

DL still = 0 and I have 10 of these suckers scheduled to complete over the next several hours.

The server DOES require maintenance, at least IMHO.

Do you know what the deal is on the other SMP server that typically also assigns to my Quads? (xx.54)
Any interesting posts related to this situation in your Mods Forum?

[toTOW]

Do you know what the deal is on the other SMP server that typically also assigns to my Quads? (xx.54)

This one has been quite low on work for a while ... which explains why most of the client are assigned to 171.64.65.56

[314159]

Umm, you might want to check serverstats.
It is loaded with WUs and has been.

What am I missing?

Addendum: WEIGHT also has not been changed. I went from a nice mix of WUs from xx.56 and xx.54 to xx.56 only,

[RAH]

PG has locked the server, so that most WU will come from the other (56). They can do this with "WEIGHT".
Doing the same with fermi WUs right now. They may have a project that needs to be completed in a hurry.

[314159]

Hmm, I believe - perhaps in error - that higher WEIGHT tells the AS to take that server as the preference.

Who's right?

[PantherX]

The higher the "weight" the more priority it gets.

[RAH]

I just put everything to "one unit". I'll just shut down until they decided to get back to normal.
All the talk about blocking IP, and such to regulate WU's, and they do the same thing. Then
they wonder why some do it.

Its like I don't own my machine.

[kasson]

We haven't changed anything (to my knowledge) regarding those two servers recently. So hopefully you'll go back to getting a mix.

[314159]

Have there been changes to the AS logic?
Previously, a bad WU made three (or whatever) attempts and then was assigned a new WU if it was defective.(Linux)
The one machine was the first one that I have had put into "sleep" mode since my 7-8 years in the project.

Keep in mind that I am 90% SMP and rely on consistentcy from the PG. (assuming that things appear well, i.e. no widescale problems - we all understand those)

"to my knowledge"? With ALL of these SMP machines here, I sense that something is amiss.

Edit: Oops! Forgot to say THANK YOU.

[Machobear]

I'm having problem connecting to this server. My rigs can get new work units from it but cannot upload finished results to it. I have 2 units on 1 machine and 1 on another just stuck there. I checked the firewall log and it shows that connection is being made, giving the fact that I can download new work units I don't think my link to the server is bad.

Is there any alternative way of sending result manually?

Thanks.

[bruce]

There is no alternative way of delivering results manually.

The server has had problems on and off and generally recovers from whatever is wrong in a few hours though it often takes longer than that to catch up with the backlog that was created during the outage.

[VijayPande]

Short term (days): I don't see any problem at the moment, but there is a known issue with overloading of this server due to the nature of the very large WUs. We have implemented a short term hack which has helped solve most of the issue, but have also been working on long term fixes.

Longer term (weeks): We're about to order 4 new big servers which should help the load here. It usually takes some time to get and install the new servers, but it's good to know help is on the way here. Also, it looks like we have the CS working with the new WS code, which would help this sort of issue in the future.