Project: 18258 (Run 93, Clone 5, Gen 14)

Moderators: Site Moderators, FAHC Science Team

Post Reply
Mactin
Posts: 253
Joined: Sun Dec 02, 2007 1:08 pm
Location: Outremont, Montréal, Québec

Project: 18258 (Run 93, Clone 5, Gen 14)

Post by Mactin »

Overnight (2025-10-24), I processed p18258-r93-c5-g14 on my GPU. I got 324406 points for it. No apparent problems.

However, the same WU was processed successfully by another donor last month (2025-09-30).
Checking the same string they where all processed in double down to Gen 0

p18258-r93-c5-g13 in double.
p18258-r93-c5-g12 in double.
p18258-r93-c5-g11 in double.
...
p18258-r93-c5-g2 in double.
p18258-r93-c5-g1 in double.
p18258-r93-c5-g0 in double.

looking past my results
p18258-r93-c5-g15 in double.
p18258-r93-c5-g16 as allready been folded on 2025-09-30,

I dont know if this anomaly repeats elswhere or if this normal behavior.
Image
muziqaz
Posts: 2129
Joined: Sun Dec 16, 2007 6:22 pm
Hardware configuration: 9950x, 9950x3D, 5950x, 5800x3D
7900xtx, RX9070, Radeon 7, 5700xt, 6900xt, Intel B580
Location: London
Contact:

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by muziqaz »

It probably is display bug on the client or server
FAH Omega tester
Image
Mactin
Posts: 253
Joined: Sun Dec 02, 2007 1:08 pm
Location: Outremont, Montréal, Québec

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by Mactin »

muziqaz wrote: Fri Oct 24, 2025 3:37 pm It probably is display bug on the client or server
What is a « display bug » ? I dont understand !
Image
muziqaz
Posts: 2129
Joined: Sun Dec 16, 2007 6:22 pm
Hardware configuration: 9950x, 9950x3D, 5950x, 5800x3D
7900xtx, RX9070, Radeon 7, 5700xt, 6900xt, Intel B580
Location: London
Contact:

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by muziqaz »

It means things are broken server/client side ;)
FAH Omega tester
Image
Joe_H
Site Admin
Posts: 8226
Joined: Tue Apr 21, 2009 4:41 pm
Hardware configuration: Mac Studio M1 Max 32 GB smp6
Mac Hack i7-7700K 48 GB smp4
Location: W. MA

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by Joe_H »

In this case there is no display bug. There are two possibilities here. One is that the researcher started a project and had some configuration problem and restarted from the beginning. That is discouraged, but some do it instead of restarting with a new project number.

The other possibility is a server problem causing reissue of WUs that have been completed in the past.

So far Gen 17 and 18 have only been completed once. For the older generations the first completion was several weeks ago.
Image
muziqaz
Posts: 2129
Joined: Sun Dec 16, 2007 6:22 pm
Hardware configuration: 9950x, 9950x3D, 5950x, 5800x3D
7900xtx, RX9070, Radeon 7, 5700xt, 6900xt, Intel B580
Location: London
Contact:

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by muziqaz »

Joe_H wrote: Fri Oct 24, 2025 6:45 pm In this case there is no display bug. There are two possibilities here. One is that the researcher started a project and had some configuration problem and restarted from the beginning. That is discouraged, but some do it instead of restarting with a new project number.

The other possibility is a server problem causing reissue of WUs that have been completed in the past.

So far Gen 17 and 18 have only been completed once. For the older generations the first completion was several weeks ago.
3rd possibility: client imagines that there is duplicate work done, while in reality it isn't.
This one is Justin's project. It is safe to say he did not mess up the settings or released another project under same project number. He himself told us, that there is no duplicate WUs on the server, even though client reports that duplicates have been sent out
FAH Omega tester
Image
Joe_H
Site Admin
Posts: 8226
Joined: Tue Apr 21, 2009 4:41 pm
Hardware configuration: Mac Studio M1 Max 32 GB smp6
Mac Hack i7-7700K 48 GB smp4
Location: W. MA

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by Joe_H »

The app for checking WUs is reporting multiple returns, that would be a problem on the stats server. That is not a client issue.
Image
muziqaz
Posts: 2129
Joined: Sun Dec 16, 2007 6:22 pm
Hardware configuration: 9950x, 9950x3D, 5950x, 5800x3D
7900xtx, RX9070, Radeon 7, 5700xt, 6900xt, Intel B580
Location: London
Contact:

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by muziqaz »

Joe_H wrote: Fri Oct 24, 2025 7:13 pm The app for checking WUs is reporting multiple returns, that would be a problem on the stats server. That is not a client issue.
You are correct. Though users usually see these kind of things within the client's WU History section.
Either way, I sent this to Justin, will see what he digs out
FAH Omega tester
Image
jjmiller
Scientist
Posts: 183
Joined: Fri Apr 09, 2021 4:43 pm

Re: Project: 18258 (Run 93, Clone 5, Gen 14)

Post by jjmiller »

Hi all,

Thanks Mactin for reporting this. At the moment we're thinking this is tied to a bunch of server havoc that happened on/around Oct 1. Several servers went down unexpectedly, causing work to pile up on collection servers and then as they came back online all the servers became overloaded trying to transfer all the work around. I've gotten a few reports directly on other projects that this happened to at the same time, namely 18242.

At the moment we're not sure what is causing this, but are suspicious that there's a bug somewhere in the server code that's causing specific RUNxCLONEs of a project to be reset to GEN0. It seems like this is an isolated incident at this time as it only happened immediately after the server chaos.

Unfortunately though this is "duplicate" work (as in the WU were folded in the past), it is actually necessary work as the server deleted those prior WU.
Post Reply