WCG restart delayed...until...13May 2023

[Vester]

WCG tweeted today about the status update at worldcommunitygrid.org FOUR days ago.

 

[doneske]

By our best estimate, this issue and others related to user experience on the WCG website should be fixed by early next week.

One thing I've learned about this group is they not very good at meeting dates....

 

[Vester]

This week passed without an update. There are no more excuses. Put this thing on the road!

 

[doneske]

Has anyone seen any updates since June 3rd since they seem to use different platforms for updates? They seem to be consistently inconsistent.

 

[Jason Jung]

Nothing on Twitter or Facebook yet.

 

[Vester]

It is too hot to crunch here anyway. My computer is idle.

 

[buja]

Guys,

You can run the 'Amicable Numbers'-project, it doesn't cure cancer I think, but it works, even with the Boinc Manager

 

[Vester]

Website and Forum Restart

I have added the project to BOINC Manager but got " World Community Grid | Scheduler request failed: HTTP service unavailable." Hopefully, tasks will be available soon.

 

[supdood]

Hopefully they can get the WU side of BOINC up and running soon. There was a post in the WCG forums that I interpreted to mean that there was little to no IT support for getting the system up and running (I think it was also a polite way of saying that Krembil and IBM left the team high and dry with the short cutover contract they signed). So some of the speculation in this thread seems to be correct that it was the research team struggling through IBM's implementation that caused a lot of the delays. Still, pretty impressive if it is the case that a bunch of cancer researchers got WCG back up and running. I'm hoping that they have a good relationship with their IT department for hardware and maybe getting some support going forward if they are serious about sustaining and expanding WCG.

 

[Vester]

From the news posted on 13 September 2021 entitled World Community Grid finds a new home at Krembil Research Institute:

As always, we’re grateful for the support of each volunteer and partner, including Canada's New Digital Research Infrastructure Organization (NDRIO), Compute Ontario, Sharcnet and Cancer Computer. With their continued support we can execute the transition and ensure future growth of the program.

Dr. Igor Jurisica has been active on the forums.

 

[Vester]

"Soon" now means "sometime in the near future."

 

[Nick Name]

I saw that today.

The move of WCG from IBM to our group at Krembil Research Institute has been a significant undertaking, considering time, people and physical infrastructure. After planning, legal matters, design and setup, we had to reconfigure the system to work on our current hardware, and consider our licensing options.

I expected the move would basically be moving the system including hardware to its new home, and setup would be relatively minor. They should have blown it up and went with a standard BOINC and open source solution. I don't know how that would impact credit and workflow of current projects, and maybe that's why they're sticking with the IBM setup, but I think rebuilding from scratch would be better going forward. The "legal matters" is particularly bothersome to me.

 

[Jason Jung]

There are always legal matters when it comes to transferring ownership of an organization. Could be just basic things like making sure they have rights to use the name, web domains, what liabilities they are inheriting, debts, etc.

 

[Jason Jung]

I have work units!

 

[Nick Name]

I have work units!

That's great. Have you successfully returned any yet though.

 

[Jason Jung]

Returned them all and they are all valid or still pending. I've only gotten a single batch on two different machines. Hopefully it's a slow start and not an issue generating or distributing work units.

 

[doneske]

No changes made to the profiles are being propagated to the client. Unfortunately, I was working on ARP before shutdown and I'm now stuck with it.

 

[Vester]

My computer is working on twenty tasks with no extras in queue. I removed Rosetta from the projects list. I could not get WCG to download tasks if Rosetta were running. WCG downloaded with Rosetta suspended, but that held 21GB RAM for Rosetta.

Edit: Tasks were uploaded and more tasks were downloaded. It is running twenty and 27 are "ready to start." Sixteen are shown as "pending validation."

 

[doneske]

I'm baffled at the things that aren't working with the project. One would think that if you did a backup and restore of the configuration from IBM to the new hardware, it should look and act the same except for some IP addresses that might need to be changed. It is looking like they may have significantly altered the hardware environment which may have precluded a one-to-one restore of the configuration. The fact that the server doesn't recognize the web modified preferences says there is a broken link between BOINC and the web files. They also aren't any better at communicating than IBM was. This "black ops" mentality where everything is a secret and not open for comment is stupid. Any long term contributor could do a better job of communicating than the intern (cyclops) they hired. Too frustrating for this old guy. I think they have lost me as a contributor.

 

[Vester]

I have downloaded about thirty MCM tasks in the last hour. No problems yet, but they may have a bandwidth problem. Some of the downloads stalled and others were slow to download.

Doneske, are they utilizing any moderators? I see duplicate topics.