171.64.65.54

[Penfold]

I get an 'OK' when I click on that link.

I guess we just hunker down and wait,eh?

[TomJohnson]

I just tried the usual fix that I have used in the past to get a new WU by trashing everything in my "work" folder plus the "unit1/queue.dat" file then restarting F@H but this time I am not getting a new WU from another Assignment Server.

[7im]

Yes, please wait for a mod / admin to come along and ping PG about it, or a PG'r to see it...

[Joe_H]

There is only one work server available for Macs running the V6 folding client, so you will have to wait until it is fixed. Trashing the folder will not help, and discards the WU you just finished.

[Penfold]

There is only one work server available for Macs running the V6 folding client, so you will have to wait until it is fixed. Trashing the folder will not help, and discards the WU you just finished.

Just out of curiosity, can you say how many work servers are available for machines running Ubuntu? This would be interesting for me to know - I am in the mid-early stages of building my own machine running Ubuntu, dedicated entirely to F@H.

[Joe_H]

Just out of curiosity, can you say how many work servers are available for machines running Ubuntu? This would be interesting for me to know - I am in the mid-early stages of building my own machine running Ubuntu, dedicated entirely to F@H.

An exact number, no. You could count the number from the Server Status page, but that is not really necessary. There are at least several dozen work servers listed as having work for Linux machines, how many would have work for your machine would depend on its configuration.

The count of WS for OS X and version 6 is easy because there is only one WS for that combination. That is because the OS X V6 client only supports use of the A3 folding core. There are other servers with A3 work, but the projects on them require a later version of the A3 core that has not been made available for OS X. The software development effort for OS X folding has been put towards the V7 beta instead, but there are issues with it that are still being worked on. The issues are currently relatively minor if and only if the V7 beta is used with OS X 10.6, totally unusable with 10.5 and earlier, and very problematic with 10.7.

[Aardvark]

I am also unable to return a completed WU or receive a new WU. I guess I will just mosey over to the end of the queue and wait for the Pande Group magic to occur.

[bruce]

When a server is listed as DOWN or as REJECT, there's absolutely nothing you can do about it except to report the problem. Either it needs a reboot or hardware repair. Sometimes the server's owner will notice it before it's reported but it doesn't hurt to report it. (A second page of "me too" reports are no more valuable that the first notification.) Since it's the weekend, you may or not get the attention of the right people until Monday morning. There's no general rule about whether the servers are monitored by an admin on weekends or just by the project owner and no guarantee that the project owner is available on weekends.

In many cases, there are multiple servers that can supply WUs so if a single server is down, you may not be able to upload your results until it is rebooted (if that fixes it) but you may be able to get work from other Work Servers. In the case of projects for V6 on OS-X, it's possible that there's only one work server -- as has already been suggested, in which case you may not be able to get new work either. There might be another server with Work for the V7 client or perhaps not. Is anybody running V7, and are you getting work?

There's nothing directly that forum volunteers can do about it so you'll probably have to wait until morning for V6. If you want to try installing V7 it may or may not help, but you can leave V6 running. Installing both will not cause a conflict except that they might try to share the same resources if they both receive a WU and compete with each other for resources. You can complete the processing of either WU by running V6 with the -oneunit flag or by choosing "Finish" on V7.

[earthsaver]

I have two clients in different geographical locations both stuck trying to submit to 171.67.108.25. All of my clients have been unable to submit their WUs for more than 36 hours. They've also been unable to connect to 171.65.65.54.

[Joe_H]

... In the case of projects for V6 on OS-X, it's possible that there's only one work server -- as has already been suggested, in which case you may not be able to get new work either. There might be another server with Work for the V7 client or perhaps not. Is anybody running V7, and are you getting work?

Just the one server from what I found out last year when I looked into this, and no change in that status since then. Only active projects available for OS X Intel and the version 2.22 A3 core are 6026-6080 and all are from WS 171.64.65.54. From a reply by kasson back then when some other A3 projects went on advanced and were picked up by some of my V6 clients and also reported by others, the 2.27 version of the A3 core was not available for OS X. I have looked and saw no mention of this core in the topics for beta testers, so it might have been on internal test. All other WS with A3 projects are set to Windows and Linux only.

As for getting work with the V7 beta client, I currently am running that on all my Macs with OS 10.6.8. I am getting SMP core A4 WU's from projects 80nn and 764n currently. In the recent past I have also received 7600, 7610, 7611, 7808 and 7809 WU's.

As for running V7 to get work while this WS is down, I can only recommend that option to users with OS 10.6. There are major problems with the FAHControl interface under 10.7, and the V7 software does not work at all with 10.5 and earlier OS's. If persons do decide to download and run V7, just a note about V6 and the -oneunit flag. The control panel method of running V6 does not recognize that flag, you would need to run the V6 client using the command line from the Terminal app.

[Joe_H]

I have two clients in different geographical locations both stuck trying to submit to 171.67.108.25. All of my clients have been unable to submit their WUs for more than 36 hours. They've also been unable to connect to 171.65.65.54.

171.67.108.25 is a collection server and has been down for a long time. As mentioned in the postings under the "Do This First" topic, it does not need to be reported. 171.65.65.54 (though I assume you mean 171.64.65.54) is the work server and has been down since Saturday night, see this topic - viewtopic.php?f=18&t=21831.

Edit by Mod: Two posts moved to topic 21831.

[earthsaver]

According to the log, this server was down for two days and has been up but in reject mode for the last few hours.

[TomJohnson]

Server is back up -- just got a WU:

[02:32:11] Folding@Home Gromacs SMP Core
[02:32:11] Version 2.22 (May 7 2010)
[02:32:11]
[02:32:11] Preparing to commence simulation
[02:32:11] - Assembly optimizations manually forced on.
[02:32:11] - Not checking prior termination.
[02:32:11] - Expanded 1765066 -> 2254597 (decompressed 127.7 percent)
[02:32:11] Called DecompressByteArray: compressed_data_size=1765066 data_size=2254597, decompressed_data_size=2254597 diff=0
[02:32:11] - Digital signature verified
[02:32:11]
[02:32:11] Project: 6061 (Run 1, Clone 75, Gen 471)
[02:32:11]
[02:32:11] Assembly optimizations on if available.
[02:32:11] Entering M.D.
[02:32:18] Completed 0 out of 500000 steps (0%)


Index 1: folding now 481.00 pts (69.542 pt/hr, 1668.88 ppd) 20.8 X min speed; 1% complete
bonus pts: 3180.31 (4.598 pt/hr, 11034.42 ppd); bonus factor: 6.61; kfactor: 2.10
server: 171.64.65.54:8080; project: 6061
Folding: run 1, clone 75, generation 471; benchmark 0; misc: 500, 629, 12 (le)
issue: Mon Jun 18 19:32:08 2012; begin: Mon Jun 18 19:32:10 2012
expect: Tue Jun 19 02:27:10 2012; due: Sun Jun 24 19:32:10 2012 (6 days)
preferred: Thu Jun 21 19:32:10 2012 (3 days)
core URL: http://www.stanford.edu/~pande/OSX/x86/Core_a3.fah (V2.22)
core number: 0xa3; core name: GRO-A3
CPU: 1,0 x86; OS: 3,0 OSX
smp cores: 8; cores to use: 8
tag: P6061R1C75G471
memory: 16384 MB
client type: 7 BigAdv
assignment info (le): Mon Jun 18 19:30:00 2012; BC2BA661
CS: 171.67.108.25; P limit: 524286976
user: Tom_Johnson; team: 1971; ID: xxxxxxx; mach ID: 1
work/wudata_01.dat file size: 1765578; WU type: Folding@Home

[TomJohnson]

Server is back down again - 7/1/12.

[Aardvark]

It is now 07:00 GMT on 7/2/2012. I have been unable to return, or obtain a new WU, from 171.64.65.54 since very early on 7/1/2012. This would appear to be a repeat of the shutdown that occured on/about 6/17/2012. There was a time when one could feel that PG/Stanford would "go the extra mile' to ensure that the Folders would be allowed to return completed WUs as they finished. Are those days Gone? Is it now a Mon-Fri 9 to 5 operation? It does take the "fun" out of the FAH when you sense the Vitality is possibly leaving the Project.
Just Saying.