Bigadv 6900(5,19,10) is a bad work unit

amdgamer

Supreme [H]ardness
Joined
Oct 27, 2004
Messages
4,880
Hey guys, I want to give you guys a heads up as I think this is a bad work unit. Folding2 crashed earlier today with a FAHcore_a3.exe has stopped working message right at 0% completion. I restarted the client and it picked up right at the same spot and crashed again with the same windows message. At that point, I decided to go ahead and dump the work unit to force it to download a new one. It ended up downloading the same one again and crashed right at the same spot again. If Tobit is reading this, maybe he can pass it on as I have had a massive migraine all day and am barely functional.

I dumped it again and changed machine ID in the config and deleted the queue.dat file to force it to download something new. I ended up downloading a different 6900 which is currently running perfectly.
 

Tobit

[H]ard|DCer of the Month - March 2010/May 2011
Joined
Dec 12, 2005
Messages
5,035
send me your log file dude, I'll take care of it. jt dot n1ese AT gmale . com -- fix typos and stuff. :p
 

capreppy

[H]ard|DCer of the Month - April 2009
Joined
Jul 4, 2007
Messages
3,410
am now getting this WU.

Tobit, any way to get rid of this? I've downloaded 3 times in the last 15 minutes.

edit: tried redoing the config and that did NOT help :(

edit: posted on FF. hopefully they will just kill this thing :( :(
 
Last edited:

amdgamer

Supreme [H]ardness
Joined
Oct 27, 2004
Messages
4,880
Good, i'm glad that it isn't just me. I'm sorry that you had to get it, as I also had to dump it 3 times yesterday. I don't know what I did with my log file as I think I accidently deleted it by accident while dumping this work unit. For me, I ended up setting my SMP client to a different machine number and it finally downloaded a new 6900.
 

APOLLO

[H]ard|DCer of the Month - March 2009
Joined
Sep 17, 2000
Messages
9,089
Confirmed, I'm having problems with this RGC WU as well. It crashes immediately after it commences.
 

capreppy

[H]ard|DCer of the Month - April 2009
Joined
Jul 4, 2007
Messages
3,410
I finally got a 2686 after trying for nearly 2 hours to make it go away. I did post on FF, but haven't gotten a response to it. See my post above for the FF post.
 

APOLLO

[H]ard|DCer of the Month - March 2009
Joined
Sep 17, 2000
Messages
9,089
I finally got a 2686 after trying for nearly 2 hours to make it go away. I did post on FF, but haven't gotten a response to it. See my post above for the FF post.
Change the machine ID in the client config after deleting the work folder and queue.dat file, then restart the client. The client will download a different WU.
 

amdgamer

Supreme [H]ardness
Joined
Oct 27, 2004
Messages
4,880
Confirmed, I'm having problems with this RGC WU as well. It crashes immediately after it commences.

I always thought that Stanford never releases the same work unit more than once and sometimes twice to check for accuracy. I always thought that a particular work unit isn't re-released until after the bonus deadline expires? I wonder whether Stanford is sending out a lot more duplicates of work units now to take care of the previous bigadv shortage.............
 

Filter

[H]F Junkie
Joined
Dec 30, 2001
Messages
9,436
I always thought that Stanford never releases the same work unit more than once and sometimes twice to check for accuracy. I always thought that a particular work unit isn't re-released until after the bonus deadline expires? I wonder whether Stanford is sending out a lot more duplicates of work units now to take care of the previous bigadv shortage.............

i kinda of assumed they always sent out more of the same unit to get a backup or like a second opinion.
 

may i be worthy

[H]ard|DCer of the Month - December 2010
Joined
Aug 17, 2010
Messages
1,186
I thought each unit was unique - ie there is only one 6900 (Run 5, Clone 19, Gen 10) unit out there, and once it is completed it gets sent back out there a little further along as in this case it would be as 6900 (Run 5, Clone 19, Gen 11)

So what are the odds of the same work unit being sent in rapid succession to 3 prominent and distinguished [H] members?

CONSPIRACY! :p
 

sirmonkey1985

[H]ard|DCer of the Month - July 2010
Joined
Sep 13, 2008
Messages
22,230
I thought each unit was unique - ie there is only one 6900 (Run 5, Clone 19, Gen 10) unit out there, and once it is completed it gets sent back out there a little further along as in this case it would be as 6900 (Run 5, Clone 19, Gen 11)

So what are the odds of the same work unit being sent in rapid succession to 3 prominent and distinguished [H] members?

CONSPIRACY! :p


it could be the entire clone/run that is flawed. but with the amount of bigadv WU's available its very possible you get the same WU thats returned due to an error and redistributed. but generally its a fault in the assignment server that keeps sending out the same exact WU to different people. shit happens, we deal with it and move on.
 

APOLLO

[H]ard|DCer of the Month - March 2009
Joined
Sep 17, 2000
Messages
9,089
I always thought that Stanford never releases the same work unit more than once and sometimes twice to check for accuracy. I always thought that a particular work unit isn't re-released until after the bonus deadline expires?
No, my experience with all clients from single processor all the way to -bigadv is the same RCG will be resent many, many times over following successive EUE's. It can be hours before a different RCG will finally be received by the client unless countermeasures are taken.
 

Zero82z

Fully [H]
Joined
Jan 20, 2004
Messages
27,897
I thought each unit was unique - ie there is only one 6900 (Run 5, Clone 19, Gen 10) unit out there, and once it is completed it gets sent back out there a little further along as in this case it would be as 6900 (Run 5, Clone 19, Gen 11)
The same unit is often sent out to multiple people for the sake of redundancy. Units that EUE are also resent by the assignment server.
 

amdgamer

Supreme [H]ardness
Joined
Oct 27, 2004
Messages
4,880
I thought each unit was unique - ie there is only one 6900 (Run 5, Clone 19, Gen 10) unit out there, and once it is completed it gets sent back out there a little further along as in this case it would be as 6900 (Run 5, Clone 19, Gen 11)

So what are the odds of the same work unit being sent in rapid succession to 3 prominent and distinguished [H] members?

CONSPIRACY! :p

2 Distinguished.......

I'm more or a hanger on just barely trying to make my contribution to a team full of heavy hitters. I finally did hit 10 million the other day though.

No, my experience with all clients from single processor all the way to -bigadv is the same RCG will be resent many, many times over following successive EUE's. It can be hours before a different RCG will finally be received by the client unless countermeasures are taken.

*sigh*

The good news is that hopefully it will be removed and noone else will recieve this freak unit. We have had a really good bigadv run lately, so i'm not too upset as 3 uncompleted work units due to a bad WU isn't going to put a dent on my rather solid completion percentage these days.
 

APOLLO

[H]ard|DCer of the Month - March 2009
Joined
Sep 17, 2000
Messages
9,089
*sigh*

The good news is that hopefully it will be removed and noone else will recieve this freak unit..
Yeah, my machine that got this unit was idle for about 9 hours. The Windows error box came up right away and the client just stopped. I couldn't monitor my farm during that time so it was doing nothing for much of the day.
 

mplee

Limp Gawd
Joined
Oct 7, 2009
Messages
352
The same unit is often sent out to multiple people for the sake of redundancy. Units that EUE are also resent by the assignment server.

I had 2 machines running the same unit a few weeks ago. It probably wasn't the first time, but I noticed it since HFM.net had the 2 units highlighted.
 

may i be worthy

[H]ard|DCer of the Month - December 2010
Joined
Aug 17, 2010
Messages
1,186
I got it too! Well I'll be buggered. Grassy knoll! Black helicopters. They are watching me! :p

Luckily I was working on the shelving at the time, and noticed the temps were at idle, so I only lost maybe 10 minutes - which is just insanely lucky that I was there at that time.

EDIT: it had failed a few times now that I look at the logs. Lost an hour. Not complaining.... TOBIT SAVE US BEFORE IT KILLS US ALL! :eek:

I am heading on over to FF to report it... again!

original.jpg
 
Last edited:

zero2dash

Supreme [H]ardness
Joined
Oct 23, 2007
Messages
6,081
I had a regular a3 just twiddle its thumbs for the last 18hrs on my e8400, but unfortunately I can't post the info (already deleted the pertinent stuff....work folder, unit info txt, queue.dat etc). It was a 6702, I remember that much...sorry. :( Obviously doesn't help much but just wanted to comment to say "I feel all of your pain". :eek:
 

amdgamer

Supreme [H]ardness
Joined
Oct 27, 2004
Messages
4,880
Hey, welcome to the club! :eek:

I got it too! Well I'll be buggered. Grassy knoll! Black helicopters. They are watching me! :p

Luckily I was working on the shelving at the time, and noticed the temps were at idle, so I only lost maybe 10 minutes - which is just insanely lucky that I was there at that time.

EDIT: it had failed a few times now that I look at the logs. Lost an hour. Not complaining.... TOBIT SAVE US BEFORE IT KILLS US ALL! :eek:

I am heading on over to FF to report it... again!

original.jpg
 

zero2dash

Supreme [H]ardness
Joined
Oct 23, 2007
Messages
6,081
nevermind this one ended up being ok, had something else kooky in Windows (some DLL process) hogging cpu cycles. ignore me :)
 
Last edited:

Tobit

[H]ard|DCer of the Month - March 2010/May 2011
Joined
Dec 12, 2005
Messages
5,035
This particular unit in question, 6900 (Run 5, Clone 19, Gen 10), was finally marked as a bad unit this morning by ChelseaOilman.
 

10e

2[H]4U
Joined
Jul 20, 2006
Messages
3,383
Not fast enough I guess because I got it on the SR-2 about half an hour ago and it's already died and gone to HELL.

Lucky me I got a 2685 to replace it.
 
Top