September 25th email from Igor: juris at ai.utoronto.ca Dear Daniel - thank you for your support and patience.
I should have an update later today - and if Dylan does not run into
another challenge - boinc on wcg may be restarting today -

thank you
igor

spoiler: the good timeline did not come to be.

New update from WCG:

September 25, 2025
BOINC scheduler and uploads are projected to be up Monday, September 29th, 2025 after a final weekend sprint to test the new deployment strategy for increased performance, event driven horizontal scaling, and reduced load on the BOINC database.
-------------------------------------------------------------------------------------------------------------------------
Well, don't take it as gospel.....

Connecting the dots...

Greetings.
I am trying to understand which problem(s) the database fixes are related to.

Specifically, see this in my BOINC event log.... Is this HTTP service problem related to database discussions?
Or, perhaps, I am missing a different posting in this thread?

Fri 26 Sep 2025 08:20:41 PM EDT | World Community Grid | Requesting new tasks for CPU

Fri 26 Sep 2025 08:20:42 PM EDT | World Community Grid | Scheduler request to https://scheduler.worldcommunitygrid.org/boinc/wcg_cgi/fcgi failed: HTTP service unavailable

Or, has Kremsil just disabled the WU port until the database issues are fixed?
A explanation would help.

Thanks in advance.
Jay

Krembil/Jurisica still has the BOINC portion (database, work download and upload, etc) disabled and as posted a few messages prior in this topic they hope to get that working sometime this Monday September 29, 2025.

Monday September 29, 2025, 08:35 Toronto Time
Scheduler still down:
https://scheduler.worldcommunitygrid.org/boinc/wcg_cgi/fcgi
503 Service Unavailable
No server is available to handle this request.

Let's see in a couple of hours, if they get BOINC stuff up and running.

As I type the time in Toronto is about 10:30 on a Monday morning. So give them a few more hours to get back from Tim Hortons and get their caffeine:blood ratios corrected and we may see some progress...........

In reply to robsmith's message of 29 Sep 2025:

get their caffeine:blood ratios corrected and we may see some progress...........

Nay!
https://xkcd.com/323/

Not yet:

Mon 29 Sep 2025 12:36:44 PM EDT | World Community Grid | Sending scheduler request: Requested by project.
Mon 29 Sep 2025 12:36:44 PM EDT | World Community Grid | Not requesting tasks: don't need (job cache full)
Mon 29 Sep 2025 12:36:45 PM EDT | World Community Grid | Scheduler request failed: HTTP service unavailable

Mon 29 Sep 2025 02:18:38 PM EDT | World Community Grid | update requested by user
Mon 29 Sep 2025 02:18:42 PM EDT | World Community Grid | Sending scheduler request: Requested by user.
Mon 29 Sep 2025 02:18:42 PM EDT | World Community Grid | Requesting new tasks for CPU
Mon 29 Sep 2025 02:18:43 PM EDT | World Community Grid | Scheduler request failed: HTTP service unavailable

---

16:48 Toronto, still Nada............

Now entering week 5.... Another Monday just about run out of time and no access to the BOINC servers.
getProjectInfo: could not find info for: http://www.worldcommunitygrid.org/

Deadlines for not uploaded/reported tasks, extended again. Now 2025-10-06 and 2025-10-08.

Some progress.

https://scheduler.worldcommunitygrid.org/boinc/wcg_cgi/fcgi

Now responds with:
403 Forbidden
Request forbidden by administrative rules.

So, now there is a server there at least.

A new post on the operational status page :-

September 30, 2025
Tech team is working through issues with the WAS container configuration and the BOINC daemon builds on the science cluster.
We are also pinging data centre admis as we don't have the /science filesystem on 1/6 of the science nodes, don't have the ability to add the IP address to the NFS share
Hopefully, not a blocker but does lock that node out of some roles until we can mount the NFS share.

So I will take that message from Jurisica to mean that we can expect at least another week or more before maybe we see the BOINC servers stable enough to restart.

32 days down, going on 33 days.
Closing in on Christmas Eve 😊

Internet accessOK-project servers "may" be down.

I am now getting lots of these

Wed 01 Oct 2025 01:47:51 PM EDT | World Community Grid | Sending scheduler request: Requested by project.
Wed 01 Oct 2025 01:47:51 PM EDT | World Community Grid | Requesting new tasks for CPU
Wed 01 Oct 2025 01:47:53 PM EDT | World Community Grid | Scheduler request failed: Error 403

---

New Years Day is getting closer and another day half gone in Toronto with no sign of life for the BOINC portion of WCG.

New update from the WCG team:

October 2, 2025
Hosting resolved the issue with our NFS shares in Openstack; scheduler and download path seem to work; we are debugging the new containerized file_upload_handler FCGI worker pool and file_upload_handler integration with Kafka through the librdkafka library.

We will remove the ACL restrictions that are causing 403 forbidden when users try to contact the scheduler or upload files shortly.

We are still validating the end-to-end upload path with our local BOINC clients. Once we confirm it, we will open up traffic to the upload servers, start serving the downloads that were available when we shutdown, and we can start turning the BOINC daemons back on.