171.64.65.56 not responding

[314159]

@pi, I'm not quite as hospitable as you I guess because I think it's deplorable that situations such as this should be allowed not only to persist but recur. I'm not looking for a debate, just stating my opinion. If this were a business it'd have gone under months ago. It's not a business, it's a charity relying on goodwill so double the emphasis. Imagine a Goodwill with no doors for donors to enter.

I can only tell you that I know that Dr. Kasson is doing his best to get this server up and running properly and that over the course of the past few months he has done the same thing, often at times when he should have been sound asleep. (read this entire thread for examples)

I have many criticisms concerning this project developed over my 6-7 (maybe 8?) years of activity, but this is not one of them.

That said, I have come to absolutely DREAD weekends!!!

All 12 of my C2D's have eventually connected to either this server or the CS (mainly via their clients' autosend feature).
ALL have received new work. ALL were a2's from this server. The worst delay was about 4 hours as I recall (I was catching some Zzzzz's at the time).
So the situation is really not that bad. Is it?

Ever had a server down for two weeks? I have - long ago. Did I lose credit for the WU? Nope. (it was worth 60 points or so with a looong deadline, typical of many of my 50k+ completed (and now worthless other than from a scientific perspective) WUs.

Our good friend subscribes to this thread and I am sure that whatever can be done, will be done.
I am not sure that a simple binary restart will correct the "Netload situation" (that I do not think actually exists).
We'll see.... be patient - please!

Beta=Beta and this stuff happens.
Also keep in mind that Dr. Pande has what I would consider relatively limited (but excellent) personnel resources.

Edit: Maybe we can all donate a few dollars to the Project with the stipulation that the funds be utilized for a new (and powerful) server devoted to SMP a2 WUs for dual core and C2D machines ONLY.

[ArVee]

Absolutely, John, and the frustration and criticism wasn't directed at Kasson personally and certainly wasn't directed at any way shape or from at you, it was directed at project admin in general. I fear for donors tacitly dropping out due to what they might view as incompetence. They might deduce that they spend more effort on this than the recipients. I'm going to guess that the bulk of project recipients don't inhabit the hallowed halls of academia.

This type of stuff NEEDS to stop happening so regularly, as evidenced for instance by your comment re hating weekends. Vijay or somebody needs to wake up and realize that the bulk of the population won't put up with this, they'll wonder why they should bother if the recipients don't bother. They'll also be aware that the people USING this work they've completed are on salary (I can't think of what the remuneration that research profs receive is called at this moment, ) and that it's their job.
Man, hire a kid to sit there and kick the servers every ten minutes if that's what it takes, just do something that yields results.

This has been put more eloquently before, but perhaps the different trajectory taken by plain English will strike the target differently or in a different spot.

EDIT: to fix a particularly glaring typo

[bollix47]

171.67.108.24 is "reserved" for Quad processors (which most of mine are).
I may be wrong on this if the client arguments (or overclocking issues) make a difference.

This would indeed explain why it's stuck on .56 as it is a C2D.

On a happier note it was finally able to get work from .56.

See you tomorrow around the same time.

[314159]

Hey BW:

Did you notice the time of day that those machines picked up the a1's?

My next C2D comes due in about 6 hours (730PM PDT) and it just may end up having a -oneunit argument added (ditto for 8 other DC or C2D machines).

I have 10 Quads completing prior to then. I will add a here for good luck on them.

If xx.108.24 goes down, we are all screwed. <--sad, not mad
(it's up as I write this)

[314159]

@ ArVee

Absolutely, John, and the frustration and criticism wasn't directed at Kasson personally and certainly wasn't directed at any way shape or from at you, it was directed at project admin in general. I fear for donors tacitly dropping out due to what they might view as incompetence. They might deduce that they spend more effort on this than the recipients. I'm going to guess that the bulk of project recipients don't inhabit the hallowed halls of academia.

In no way did I or will Dr. Kasson interpret your observations negatively!

I too fear attrition and have observed it first-hand (and "big time"!). That's a long and complex story that I need not relate here.

Unlike you, the VAST majority of volunteers do not even read this Forum or their Team Forums when situations such as this occur.
I applaud you for taking the time to do so, and your opinions WILL be read by the P.G. folks.

If you had the total perspective on what Dr. Pande's staff has to contend with, I firmly believe that your opinion might be somewhat different.
(For example, there just "may" be forums on this board that your present privileges do not permit you to read. I won't even mention the time often required during the peer review process that preceeds publication of results - Oops! I just did!)

You may have experience with servers and server code problems. At times they are a true p.i.t.a. to fix.
Overall, and although this thread is quite long, the server problems have not been what I would consider excessive and I expect them to be less frequent in the not too distant future.

Hang in there, cool off, and relax.

Most importantly, Fold on.......

Geesh! I feel like a #!#% Moderator!

[BrokenWolf]

John,

At the time of this writing they were between 7 hours and 6 hours and 21 minutes ago. (when they downloaded that is)

Noah and I have 14 SMP clients finishing up in the next 10 hours as well. Some are dual and some quad. I hope things can stay running at optimal efficiency for a while.

BW

[314159]

Thanks Tim,

My batch completed at 637AM PST (all got a2's) with one other completing at 1046AM PST (also got an a2).

It looks as if I escaped the "problem". I have no idea why you did not given what I am seeing.....
Are you holding your mouth in the correct position to ensure receipt of an a2?

I hope that my next batch this evening has the same good luck - but they DO now have -oneunit on them.

[ArVee]

Thanks, John, and that would be the best sign of all in future, very short threads on problems. I have no direct experience with servers but I can just about imagine, the sentiment is that ya gotta do what it takes, but I won't start up on that again. Ok, let's agree, three terrible jobs to have: 1) server guy 2) the good professor when 1) is having problems 3) mod at any time
Enjoy the weekend.

[314159]

You too ArVee - It's been a pleasure meeting you!

I plan on introducing a Bill in Congress that eliminates all future weekends.
Since our "beloved" Congressmen apparently don't read proposed legislation prior to voting, it will probably pass.

Now: Where are the Forum Mods and other valued staff?
Someone on the "inside" must have some DETAILED insight into this situation.

Edit: THIS SERVER IS UP RIGHT NOW - 730PM EDT.

[BrokenWolf]

All is good from my end. All the systems that finished recently got a2 WUs. Now I just have to wait 3 days for all the a1's to complete. Not that long for all of them but one of my C2D systems got an a1 and it has 3 days left (for a total of 3.9 days total). Thank goodness for the long deadline on these.

It usually takes about ~1 to 1.5 days for an a2 depending on the P-R/C/G.

BW

[bollix47]

FYI

Wed Aug 12 04:40:10 PDT 2009 171.64.65.56 SMP vspg4 kasson full Reject 0.15 148 0 5 17883 932 1 2.27 1584 1584 6874

[314159]

Still rejecting.

Please HELP!!

Thanks!

[kolinsb]

i need help to solve this problem:

Code: Select all

kolins@kolins-desktop:~/fah$ ./fah6 -smp -verbosity 9 -send all

Note: Please read the license agreement (fah6 -license). Further
use of this software requires that you have read and accepted this agreement.

2 cores detected


--- Opening Log file [August 20 15:34:47]


# SMP Client ##################################################################
###############################################################################

Folding@Home Client Version 6.02

http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: /home/kolins/fah
Executable: ./fah6
Arguments: -smp -verbosity 9 -send all

[15:34:47] - Ask before connecting: No
[15:34:47] - User name: kolinsb (Team 51246)
[15:34:47] - User ID: 675FCBE500573B45
[15:34:47] - Machine ID: 1
[15:34:47]
[15:34:47] Loaded queue successfully.
[15:34:47] Attempting to return result(s) to server...
[15:34:47] Trying to send all finished work units


[15:34:47] + Attempting to send results
[15:34:47] - Reading file work/wuresults_03.dat from core
[15:34:47] (Read 48667661 bytes from disk)
[15:34:47] Connecting to http://171.64.65.56:8080/
[15:54:49] Posted data.
[16:08:05] Initial: 7254; - Uploaded at ~23 kB/s
[16:08:05] - Averaged speed for that direction ~30 kB/s
[16:08:05] - Unknown packet returned from server, expected ACK for results
[16:08:05] - Error: Could not transmit unit 03 (completed August 19) to work server.
[16:08:05] - 7 failed uploads of this unit.


[16:08:05] + Attempting to send results
[16:08:05] - Reading file work/wuresults_03.dat from core
[16:08:05] (Read 48667661 bytes from disk)
[16:08:05] Connecting to http://171.67.108.25:8080/

i running ubuntu 64-bit on VMWare. never before i didn't had any problem like this. what's wrong? i can download unit, finish it, and i can't send results and every wu is 1920 points.

[4ndr3w]

If nobodies replies on this thread make a new one under this board. with the ip address of the server (171.64.65.56)

[314159]

Server is in REJECT status as of 1:15 PST, 20 August.

Please help!

Thank you.