DF: Stats (06/20/03) - Down to 3rd :(

[Confused]

Time of Report: 11:00:00 GMT-5 (it's about 20hrs late, sorry!)


1. FreeDC....................Today: 6,646,110 Overall: 21,976,816
2. Ars Technica ........Today: 6,863,415 Overall: 17,617,813
3. Team AnandTech.Today: 3,671,105 Overall: 12,675,074

Moved down to 3rd

Sorry, can't go through all the Milestone Makers, but congrats to you I will do a "bumper edition" milestone makers tomorrow when I have more time


Top 5 Producers of the Day:
Number in parenthesis is overall position

nsmmxx: 727,118 (8)
FoBoT: 674,035 (11)
Silverdragon: 573,840 (13)
MereMortal: 216,067 (43)
ATWE Robor: 183,691 (56)

Wow, you can really tell who's hitting the higher generations already.
Congrats!
Good day! All in the top 60

Top 5 RMS
Number in parenthesis is overall position

ATWE Robor: 5.123046 (2)
indelab: 5.470141 (14)
nsmmxx: 5.841706 (52)
SilverDragon: 5.880353 (59)
TA bfour: 6.073299 (90)

Good Crunchin'



Garry

 

[ken008]

I am having problems with the client. I am on gen 246, 45 gen buffered, error message " cannot read trajectory distribution please create a new one , hit return " How do I create a new one? All three of my clients were spinning wheels this morning. I am thinking this project needs a fulltime internet connection to run right. I don`t have that. I will not put up with spinning wheels for very long and lost work really chaps my butt. I hope someone knows how to get this thing to run right.

 

[grunter]

Thanks for the stats confused!

 

[JetBlack69]

Originally posted by: ken008
I am having problems with the client. I am on gen 246, 45 gen buffered, error message " cannot read trajectory distribution please create a new one , hit return " How do I create a new one? All three of my clients were spinning wheels this morning. I am thinking this project needs a fulltime internet connection to run right. I don`t have that. I will not put up with spinning wheels for very long and lost work really chaps my butt. I hope someone knows how to get this thing to run right.

Hmmm... It doesn't need full time connection, but it is helpful. Are you just running it from a command prompt? Sorry, I'm not sure how you create a new trajectory. You might want to post on this forum Link and post your question there.

EDIT: I just noticed your post over there, I was going to point you to that thread, but you already found it.

 

[NicColt]

>I am thinking this project needs a fulltime internet connection to run right. I don`t have that.

I have 3 clients folding offline and have over 200 gens buffered no problem so far.

I just noticed that one of my offline clients has a 5.87 RMS : )

 

[Phiton]

I am having problems with the client. I am on gen 246, 45 gen buffered, error message " cannot read trajectory distribution please create a new one

I'm having the same problem.

 

[ken008]

I copied the folder renamed it then reinstalled. It is running again. I lost a lot of work. I saved the old one in case there is a way to recover the work or examine the error log to discover the problem. Losing 45 generations in the 200 range is very disappointing. Other than that my 1600MP dually seems to like this project. It was doing great for two days. I reinstalled both instances, starting over. The error log on the old install is 553 kb, that is large for only running two days. It is mostly connection errors since I am on dialup. They may need to have node numbers like D2OL to differentiate the clients on the server end for phase II. It did not matter for phase I.

 

[Falke]

The client seems very buggy to me. I had errors on almost every machine. Upload was impossible too...

ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 0 chars.
ERROR: [000.000] {foldtrajlite2.c, line 4127} Error during upload: NO RESPONSE FROM SERVER - WILL TRY AGAIN LATER
ERROR: [000.000] {foldtrajlite2.c, line 4050} Warning during upload: STATUS 910 MISSING PREVIOUS OR ILLEGAL GENERATION

I will stop working for DF until the client is fixed.

 

[ken008]

A solution to the trajectory error has been posted Here. It worked for me.

 

[NicColt]

>ERROR: [010.003] {taskapi.c, line 1217} [ReadServerResponse] Timeout waiting for response, got 0 chars.
>ERROR: [000.000] {foldtrajlite2.c, line 4127} Error during upload: NO RESPONSE FROM SERVER - WILL TRY AGAIN LATER
>ERROR: [000.000] {foldtrajlite2.c, line 4050} Warning during upload: STATUS 910 MISSING PREVIOUS OR ILLEGAL GENERATION

Apparently this is all normal. The servers are just overloaded. It will try again until it can upload it.

Missing Previous Generation message explained
I have managed to verify the cause of the 910 - Illegal or missing previous generation messages, when they occur along with 'unable to contact server'. It works like so.

Suppose you upload generation 49. Your data gets uploaded but because the server is presently overloaded and thus slow, your client times out waiting for a reply and assumes the upload failed. Meanwhile generation 49 gets added to you current movie on the server, after gen 48.

Next time around to upload, it tries to upload gen 49 again (because it thinks it failed before). This time the upload gets through, but your last generation is now 49, not 48. Hence the server issues a 'missing previous generation' error - generation 48 is no longer the last generation in your movie - and the client responds by logging the error and deleting gen 49 from disk (as intended). Then it tries to upload gen 50 which will append to the end of gen 49 without error (unless it times out again..)

Thus the 910 message really indicates that your data is getting uploaded twice. Which may be annoying but isn't as horrible as it sound. And the second time it gets deleted so its back on track. So no data is actually being lost. You can verify this by checking that your points are still increasing, and that the first filename in your filelist.txt is advancing slowly.

This is thus all inended behaviour to allow the clinet to adjust itself to a busy overloaded server.

It is of course still our problem to make the server 'unbusy' but at such time that we do, all your work shoudl still upload fine withOUT any losses (unless you've had different problems than this..)

I think one of the problems is that it's sending Gen's to fast to the server and it can't digest or crunch it at that end. One of the options that should be done on either the client or the dfGUI is to buffer gen's until you buffer the entire 250 generations and then upload it.

BTW I'm not seeing anyone with problems who have the DOS client installed, but not as a service. The problems seem to come from the Windows and Service installs. Unless I'm wrong.

 

[Insidious]

Nic,

I'm glad you found that explanation. I have one of my whiney posts over at F-DC asking about this. OK, so now I won't worry.

Thanks!

-Sid

PS: is that a quote from Howard or someone else?

 

[NicColt]

>PS: is that a quote from Howard or someone else?

Yea sorry it was a quote from Howard. But it's still a problem, what I'd like to know is, is it crashing the client when it does this ?

 

[Robor]

ATWE Robor: 5.123046 (2)

My production might not be that sweet today but that RMS looks pretty nice! Hope I put up a nice dump for tomorrow's stats.

Congrats to the MM'ers & POD's and thanks bunker!

 

[Insidious]

Originally posted by: NicColt
>PS: is that a quote from Howard or someone else?

Yea sorry it was a quote from Howard. But it's still a problem, what I'd like to know is, is it crashing the client when it does this ?

As far as I can tell, the client isn't crashing when this happens. (Been seeing it a lot here)