Well Alan, you should know that I am crunching for WCG for more than just a couple of days, so I know how things should/used to work....

As for for finding a pattern, well, that is a bit tough, at least on my side, for more than one reason. For one, any post that I try to make is delayed by well over 24h in almost all cases.
Second, WCG has hardly a track record of running consistently, which makes it hard to even establish a pattern.

I noticed the increased in over WUs even before the weekend, it became really noticeable starting Thursday.And yes, I noticed that some of the latest batch of SCC1 (not all of them!) are running maybe half the time "as usual" and that contributed to the increased number of WUs on the Results page. But I am convinced that this is not the ONLY reason.

As for those "waiting to be sent" WUs, this has been first mentioned shortly after the last "general outage" (streak counter now says 28 days, that's 4 weeks). As I mentioned before, when I checked, this seemed (for me, at least where I noticed it), macOS related WUs. Overall, those "Pending Verification" WUs has also increased to what I would call "disproportional". It used to be about a dozen WUs out of maybe 2500 WUs in Results at any given time, right now, it is 100 out of 4200.
"Pending Validation" is right now 256, which usually was maybe 150 to 200, so a not equally drastic increase.
But the most notable change for me is that with 1300 WUs "in progress", there are now 2400 WUs marked valid, which used to be in the past a 50:50 split.

The valid WUs go back now to 8/11, that's 11 days, and the oldest one is showing

Result name  OS type OS version Status Sent time Time due/ Return time Cpu time/ Elapsed time Claimed credit/ Granted credit
SCC1_0004202_MyoD1-C_37002_0	Microsoft Windows 10	Core x64 Edition, (10.00.19045.00)	Valid 	2023-08-10 20:40:24 UTC	2023-08-11 21:36:10 UTC	3 / 3	106.8 / 76.6
SCC1_0004202_MyoD1-C_37002_1	Microsoft Windows 10	Enterprise x64 Edition, (10.00.22621.00)	Valid 	2023-08-11 21:36:24 UTC	2023-08-21 13:00:57 UTC	4.11 / 4.11	46.4 / 76.6
SCC1_0004202_MyoD1-C_37002_2			Other

So, while I received and returned this WU within 25h, the wingman's WU has been sitting likely in some hoarder's cache for 10 days, and now we are stuck due to the system answering with adding an oddball "other" WU.
And then people got mad at me when I was calling out those hoarders as selfish and detrimental to the overall situation,.

Something is amiss, and it doesn't help the project in any form, if anyone (including WCG Towers and fellow crunchers) are just turning a blind eye and look the other way, just because "alles so schön bunt hier"...


Ralf

PS: it's 8:10am PST, let's see when this post is finally showing up,, and more importantly, there is an acknowledgment of an existing problem on WCG Tower's side

I would agree with you on hoarding being detrimental to good throughput. I am pretty sure a good portion of the hoarding comes from those who are afraid of outages and are trying to hedge their bets. If the supply of work units remains steady, that should lessen the incentive to hoard a disproportionate amount of work units. At the moment that is a pretty big "if."I think a cache of 2-3 days should be more than sufficient. If you run out of work units that is the purpose of a backup project. If your chosen projects run dry, take a vacation, unplug your machines, and just relax. I am certainly not going to obsess about something over which I have no control.
Cheers

Looks like the SCC queue has dried up. No new WUs and a few resend.
The MCM queue is dribbling out small numbers of WU.

Update: MCMs. WU supply is better. SCC steady new WUs. Last night a flood of resents, now back to new WUs.

I haven't really paid attention to WCG for a couple of weeks now. is there any news about when we can expect OPN1 and OPNG tasks to start flowing again?

We will know when they start flowing again when they appear in our tasks list :-)