hold still and do nothing while the Project Staff works on the issues ?

That's certainly my personal position - but I'm not staff, so that doesn't completely answer your question.

I am letting tasks run. I ran a manual update by clicking the button and my credit went up, completed tasks also reported. I am not detaching and reattaching whatever any notice says till I run out of work. The only thing I am doing differently is checking logs a bit more frequently and in a VM where BOINC has no tasks currently checking to see what happens there. I think sitting on our hands is the best idea for a while.

Status here as of 14:00 PDT US Time: Two tasks (1024, 1027) in progress at the URL transition completed and updated their data this morning. I rebooted (cold start) and did 2 manual Updates from BOINC. Tasks still show In Progress and Credit is not updated. BOINC is attempting to access climateprediction.net and reports it cannot access the server. I will await further instruction.

9/17/2024 12:48:51 PM | climateprediction.net | Fetching scheduler list
9/17/2024 12:48:52 PM | climateprediction.net | Scheduler list fetch from https://climateprediction.net/ failed: transient HTTP error
9/17/2024 12:49:02 PM | | Project communication failed: attempting access to reference site
9/17/2024 12:49:03 PM | | Internet access OK - project servers may be temporarily down.
9/17/2024 12:52:41 PM | climateprediction.net | update requested by user
9/17/2024 12:52:44 PM | climateprediction.net | Fetching scheduler list
9/17/2024 12:52:45 PM | climateprediction.net | Scheduler list fetch from https://climateprediction.net/ failed: transient HTTP error
9/17/2024 12:52:47 PM | | Project communication failed: attempting access to reference site
9/17/2024 12:52:48 PM | | Internet access OK - project servers may be temporarily down.
9/17/2024 1:25:50 PM | climateprediction.net | update requested by user
9/17/2024 1:25:51 PM | climateprediction.net | Fetching scheduler list
9/17/2024 1:25:52 PM | | Project communication failed: attempting access to reference site
9/17/2024 1:25:52 PM | climateprediction.net | Scheduler list fetch from https://climateprediction.net/ failed: transient HTTP error
9/17/2024 1:25:54 PM | | Internet access OK - project servers may be temporarily down.

I know these things can be difficult But it would have been a good idea to have sent a BIONC notice before this whole thing started.

Yes, I agree and I've made that point already which was accepted. Though in defence of CPDN, if it had all gone to plan, the changeover would have been hardly noticeable unless someone looked at the URL.

I have not personally heard any more. I'm sure there will be an update. I will check tomorrow and raise the point that a BOINC notice about the status would be appreciated. I also agree finding a post on the right thread is not ideal.

---

---
CPDN Visiting Scientist

A completed batch #1026 model of mine uploaded its Zip files earlier today and reported this evening at 21:50 BST. A new model downloaded at the same time, so the site appears now to be working for me, at least.

(The trickles have the same time because the model was running offline in the hope of surviving possible power disruption during building work.)

if it had all gone to plan

Just spent an entire afternoon debugging some bizarre bug at work. Nothing in software ever goes according to the plan. LMAO.

if it had all gone to plan

Just spent an entire afternoon debugging some bizarre bug at work. Nothing in software ever goes according to the plan. LMAO.

True, but when you have several thousands users it's a different story!

A completed batch #1026 model of mine uploaded its Zip files earlier today and reported this evening at 21:50 BST. A new model downloaded at the same time, so the site appears now to be working for me, at least.

I've just confirmed with David@CPDN that the site is still down so I'm confused how you managed to get a new task. Can you double check it really was a new task the client downloaded at 23:00 last night? I have reported it back to CPDN so they can check.

The latest I have is CPDN are in conversation with David Anderson@BOINC to understand what's going on. But the site is still down, no new signups and no-one should be able to get new tasks or report completed tasks. Uploads of model results zip file to the S.Korean server are unaffected.

---

---
CPDN Visiting Scientist

[Uploads of model results zip file to the S.Korean server are unaffected.

Thanks for that comment, Glenn. At least I now know that the trickle uploads that I'm seeing are going somewhere.

The Box, that has already resetted CPDN managed to download 3 new WUs from CPDN.ORG:

---

Supporting BOINC, a great concept !

A completed batch #1026 model of mine uploaded its Zip files earlier today and reported this evening at 21:50 BST. A new model downloaded at the same time, so the site appears now to be working for me, at least.

I've just confirmed with David@CPDN that the site is still down so I'm confused how you managed to get a new task. Can you double check it really was a new task the client downloaded at 23:00 last night? I have reported it back to CPDN so they can check.

The latest I have is CPDN are in conversation with David Anderson@BOINC to understand what's going on. But the site is still down, no new signups and no-one should be able to get new tasks or report completed tasks. Uploads of model results zip file to the S.Korean server are unaffected.

It seems to have come down at the same time as the reported task:

17/09/2024 21:49:58 | climateprediction.net | Sending scheduler request: To send trickle-up message.
17/09/2024 21:49:58 | climateprediction.net | Reporting 1 completed tasks
17/09/2024 21:49:58 | climateprediction.net | Requesting new tasks for CPU
17/09/2024 21:50:00 | climateprediction.net | Scheduler request completed: got 1 new tasks
17/09/2024 21:50:00 | climateprediction.net | Project requested delay of 3636 seconds
17/09/2024 21:50:02 | climateprediction.net | Started download of wah2_eas25_31gs_209212_24_1027_012340283.zip
17/09/2024 21:50:02 | climateprediction.net | [file_xfer] URL: http://download.cpdn.org/download//batch_1027/workunits/wah2_eas25_31gs_209212_24_1027_012340283.zip

The completed task was running offline for about six days before uploading the output files with a backup in case of a power outage (in which case I would have restarted from the backup): the backup has two batch #1026 files, and the new file is batch #1027.

I don't pretend to understand why yesterday tasks were handed out by climateprediction.net and seemingly cpdn.org. Looking at the server stats, tasks ready to send has gone down by 100 since this morning. I've asked CPDN to check again the scheduler is still running, doesn't look right.

The project is going to revert back to climateprediction.net which presumably means the cpdn.org tasks will abort (again).

---

---
CPDN Visiting Scientist

I've had confirmation from Andy that CPDN is now back up and running but under their original name climateprediction.net not cpdn.org. Trickles, data uploads and reporting completed tasks should now be working ok.

There will be a more formal Notice sent to all clients soon.

---

---
CPDN Visiting Scientist

Signup to CPDN is still broken. Have reported it.

---

---
CPDN Visiting Scientist

The project is going to revert back to climateprediction.net which presumably means the cpdn.org tasks will abort (again).

I don't think that will be the case By now, I'm pretty sure that the previous forced switch (from climateprediction.net to www.cpdn.org) was caused by the DNS redirect. They must avoid doing that in reverse, or all hell will break loose again.

If anyone who has suffered the forced transfer and is actually running a task (or tasks) could signpost it here, we could keep an eye on progress. In particular, do the trickles appear here, and is the credit added to the right account? If it seems to be proceeding normally, I'd suggest letting it complete, but set 'no new work'. When the machine is quiescent, detach/reattach to return things to normal.

If, on the other hand, error messages appear in the logs and trickles fail to get through, abort it as soon as possible to cut your losses.

That advice is, of course, unofficial and provisional. If in doubt, wait for the official announcement.

I don't think that will be the case By now, I'm pretty sure that the previous forced switch (from climateprediction.net to www.cpdn.org) was caused by the DNS redirect.

Hmm, going back to my message 71443, if DNS re-direct was the problem, why did at the time master URL switched to www.climateprediction.net when trying to open-up this URL in web-browser instead open-up cpdn.org (not main.cpdn.org but cpdn.org)?
Meaning, if DNS re-direct was the reason, shouldn't BOINC instead immediately have switched from www.climateprediction.net to cpdn.org, something that clearly didn't happen.
Instead, some time later switched from www.climateprediction.net to climateprediction.net, despite this still opened-up cpdn.org.

To me it's much more likely master URL is grabbed from a server configuration file used by both main.cpdn.org (as seen as URL to use to join climateprediction.net if you're not logged in) and in scheduler reply, here by using web-browser:

<scheduler_reply>
<scheduler_version>721</scheduler_version>
<master_url>https://climateprediction.net/</master_url>
<request_delay>3636.000000</request_delay>
<message priority="low">Error in request message: xp.get_tag() failed </message>
<project_name>climateprediction.net</project_name>
</scheduler_reply>

Please see post in News (if you haven't already) on recent events: https://main.cpdn.org/forum_thread.php?id=9315#71481

---

---
CPDN Visiting Scientist

Attaching to https://climateprediction.net/ still not possible results in AddProject: Error -224

Usually CPDN has some periods of no new work in a year. In addition to more testing, I think we can take advantage of those windows to do rebranding in the future. Once we get enough results for all outstanding batches, if we pause resend for timeouts until all deadline expires, we might just get a window where we can mess with project setup with little consequence even if things don't go as planned.

Attaching to https://climateprediction.net/ still not possible results in AddProject: Error -224

That translates into

#define ERR_HTTP_PERMANENT  -224
    // represents HTTP 404 or 416 error

We'll have to ask the admins to look into that. It's probably trying to access one of the servers whose address has changed - e.g.from www.cpdn.org to main.cpdn.org

There are probably a few of those still lying around - report them here as you find them.