Only 2 currently on my results page. 4.48 under host OS (Xubuntu25.04) 6.89 on guest (Tiny10 a cut down version of Windows10) running under VB From Memory slow down is much less with Linux running in the VM. CPU is Ryzen 9 7950X and RAM is 64GB DDR5

I can't remember exactly what was running at the same time as those tasks I think I had 5 Development branch tasks from CPDN but not sure how many MCM were running.

You mean this one below, Mike? One of its tasks was lured into the queue of my device while I was taking the train home.

<1>   ARP1_0033793_144_0  ManjaroLinux  In Progress           2025-05-23T14:46:28  2025-05-29T14:46:28
<1> * ARP1_0033793_144_1  Linux Fedora  In Progress           2025-05-23T14:46:17  2025-05-29T14:46:17

Adri
PS If you're quick, you might get to catch a glimpse of its name on the webpage with my current tasks (updated hourly).

Mike,
If you look at the cell's known history, which I did soon after it was lured into my queue, you would certainly say: 'hey! this must be the hindmost moving unit':

ARP1_0033793_108	22-01-2025
ARP1_0033793_117	05-02-2025
ARP1_0033793_123	16-02-2025

I mean, it has taken a long journey before catching up and reaching its current generation.

Can we prove it?
The cell has to move from generation 123 on 16-02 to generation 144 on 23-05, that's 21 generations in 12(February)+31(March)+30(April)+23(May) = 96 days: about 4½ days per generation.
Let's take a look at the History of the number of workunits within each generation and follow its path:
143-144 on 23-05
142-143 on 20-05
141-142 on 15-05
140-141 on 11-05
139-140 on 06-05
138-139 on 24-04
137-138 on 20-04
136-137 on 11-04
135-136 on 06-04
134-135 on 29-03
133-134 on 22-03
132-133 on 15-03
131-132 on 10-03 (uncertain here)
130-131 on 07-03 (uncertain here)
129-130 on 05-03 (uncertain here)
You're seeing that there is uncertainty here now, but we have to go way back to generation 123 on 16-02-2025, that's a difference of (28-16=) 12 days in February and 5 days in March of this year, that's a total of 12+5 = 17 days from generation 123 to 129. That's 6 generations in 17 days, so we could conclude that this is feasible, isn't it? This is why I'm pretty certain in assuming it is your 'hindmost moving unit'.

Adri

Sunday Report

All classifications have stayed put..

All 1,943 units in generations up to and including 143 seem stuck.
There are 28 normal units in generation 144 & 18,466 in 145 which are the current generations..

We are now 63% of the way through generation 145.

There are now 15,172 units held in generation 146.

18,048 units have validated in the week, but there are 1,355,529 units to go.

Based on the last 5 weeks, we would complete ARP1 in April 2027. However, improvements seem to be coming until July, but we are currently low on work.

Mike

Regarding the second task Unixchick flagged above. It validated around 19:30 UTC today. That said, it's still a worry that so many results failed to match. I hope Unixchick's result was one of the two that validated :-)

It's at times like this that the configuration differences between WCG and most other BOINC-based projects can be a bit of a nuisance -- if this was happening at a "standard" site some keen user(s) could do some basic forensics such as "is it always the same machines having problems" or "is it specific hardware or O/S set-ups"; as it is, the "same machines" would be a guess, and the rest impossible to determine :-(

Unfortunately, I rather doubt that the WCG team will have time to do that sort of diagnosis at present (if ever).

Hopefully the earlier cases (where nothing validated) will be re-issued (if that hasn't already happened) and it's to be hoped that the reruns don't go to Darwin systems for a repeat performance!

Cheers - Al.