Sgt. Joe ,
Ouch! At least my oldest is from 7th May...

I'm intrigued (and I suspect Adri might be as well!) -- any chance you could provide the workunit number so I can look at the cascade of retries that presumably involved?

Unixchick -- when I first posted about this on the 15th , I was hoping we'd have another one of those "lack of new work" periods that would clear things out. However, by the time of my second post on the subject (which hit the weekend, unfortunately) that hadn't happened.

I think Monday was a Canadian holiday, so I didn't expect a response to my third post at once (if at all); however, by now a little reinforcement won't go amiss, so thanks for that post! [I wonder if a search by username within post is used? :-)]

Cheers - Al.

[Edit - added more context]

TigerLily,

Thank You!

Cheers - Al.

Wakey, wakey, tech time. High time to send out all the tasks, that are still marked as "Waiting to be sent" .

They have all been sitting there way too long now.

Sgt. Joe,

Thanks for that, but unfortunately there doesn't seem to be a way to search for WUs by name (unless they happen to be in one's own result set) and the nearest WU names I have seen are MCM1_0266199_3160 (WU 510756062) and MCM1_0266200_6518 (WU 510827483)...

Note the huge difference in WU IDs between two items that one might expect to be at most a few tens of thousands apart (being adjacent batches), even allowing for the presence of OPNG WUs at that time. It appears that MCM1 WU IDs can be issued to several batches concurrently (and yes, I have observed that!) so adjacent items from a given batch can end up with significantly dispersed WU IDs :-(

As any attempt I might make to find the WU by random selection of WU IDs would be query-intensive and could end up looking like a DDOS attack, I'm not going there. Perhaps Adri has some data that might offer a better search range, but I suspect we'd still have to sample thousands of WU IDs :-(

Thanks anyway -- sometimes (such as in this case) I find WCG's perfectly reasonable (but very different!) approach to exposing user result data a bit of a nuisance, but it is what it is!

Cheers - Al.

[Edit. very slight rephrasing and typo fixes]

Thanks for that, Sgt.Joe!

Al, the closest match I got to Sgt.Joe's workunit name was 'MCM1_0216199_5':
MCM1_0216199_5868_1 and MCM1_0216199_5870_1, both sent at 2024-04-28T08:48:32. Too far away from 'the original' for me to start a short search on ID (i.e. by number). (Not that it matters, since I have a database based on the results from the API: long since purged from the website's results.)

Shall we dig up the original? Of course!

$ wcgstats -f= 510764828 -S

workunit 510764828

MCM1_0216199_5325_0  Linux Sparky  Error                 2024-04-28T08:34:36  2024-04-28T09:35:05    0.00/0.00
MCM1_0216199_5325_1                Waiting to be sent                                                0.00/0.00
MCM1_0216199_5325_2  Linuxmint     Pending Validation    2024-04-28T09:35:11  2024-04-28T17:48:45    3.12/3.12

Adri