There is also the problem that the "average credit" (RAC) is not being calculated for IFS tasks. I don't that there's any significance to a particular credit run having been interrupted: ...

I agree.

Dave, I believe you've already informed them more than once, you did your part. I suspect it'll be a more complicated fix as it involves 2 different model types and 2 different credit types, and who knows if the upload issue is going to add its own complication to it. I also suspect it won't get the proper attention at least until after the upload issue is resolved. It's just another thing we have to wait on.

Credit still hasn't been granted.

---

Credit still hasn't been granted.

Where are you reading that? There was another credit run two days ago (Sunday 29 January), and everywhere I look, the total credit has been updated.

Do you have "Do you consent to exporting your data to BOINC statistics aggregation Web sites?" checked in your project preferences?

Firstly I'm looking at my results page. Secondly I'm looking at the aggregation sites (and yes, I have "Do you consent to exporting your data to BOINC statistics aggregation Web sites?" ticked, and have had for a very long time). The last change in credit was at the beginning of December, despite a task completing without error mid January.

---

Credit still hasn't been granted.

Where are you reading that? There was another credit run two days ago (Sunday 29 January), and everywhere I look, the total credit has been updated.

Do you have "Do you consent to exporting your data to BOINC statistics aggregation Web sites?" checked in your project preferences?

On the CPDN pages, I have credit for all but the latest task I have run up to 29 January.
I also have credit on the signature box that will appear below.

Where I do not see it is on the Project tab of the boinc manager, and on the Statistics tab of the boinc manager. I am running 7.20.2 of the Boinc client.

I have given consent to exporting ... for so long I no longer remember when I gave it. Many years, I suppose.

---

There is also the problem that the "average credit" (RAC) is not being calculated for IFS tasks. I don't that there's any significance to a particular credit run having been interrupted: ...

Dave, I believe you've already informed them more than once, you did your part. I suspect it'll be a more complicated fix as it involves 2 different model types and 2 different credit types, and who knows if the upload issue is going to add its own complication to it. I also suspect it won't get the proper attention at least until after the upload issue is resolved. It's just another thing we have to wait on.

I'm over there next week, I'll ask him about it.
I suspect it's just a cron script that's stopped working for some reason (it's not related to JASMIN/upload11 AFAIK). Andy showed me once just how much technical setup, VMs, servers, scripts etc he's managing that a boinc-based project needs. I nearly fell off my chair, because it really needs a small team. No wonder things stop working and he doesn't have time, he's firefighting alot.

I nearly fell off my chair, because it really needs a small team. No wonder things stop working and he doesn't have time, he's firefighting alot.

It's the "high data rate, multiple application, multiple science partner" projects that need the teamwork (and yes, they do). That complaint needs to be passed up to the bean counters and the resource allocators, not dumped on the shoulders of the fallguy who said "OK, I can do that".

I see that tasks completed on 16th and 27th January have both been awarded credit. I can't see the historical data for your total credit. Just to be clear, I think you are saying the total credit hasn't increased either on your account page or on the aggregation sites. As Richard has noted, the OIFS tasks don't seem to be going into the recent average credit even though total credit is going up. That is consistent with what I see in the statistics tab ofBOINC manager for my own account. Total credit has been going up but average credit has been going steadily down despite last week uploading a backlog of completed tasks.

On the CPDN pages, I have credit for all but the latest task I have run up to 29 January.
I also have credit on the signature box that will appear below.

I'm assuming that different applications (OpenIFS vs Hadley) have different pipelines/processes/scripts for reporting credit. All of the Hadley tasks that I have completed since December 13th show 0.00 credit. I can't speak for OpenIFS task credit because I don't run BOINC tasks on Linux boxes.

Just to put some meat on the bones, the task I think Dave is referring to that completed on the 27th January is one of mine (22136276) that completed a year ago, not in 2023, so yes, that got fully credited. Since then two immensely long-running HadAM4 tasks have been running on that host, both slowly gaining credit throughout 2022. These started at the same time, but I had to pause one of them (22197506) for a few weeks due to virtual memory issues (a known WSL2 bug, I suspect, as the problem quietly went away). However, it has been slowly aquiriing the same credit as the other, a few weeks behind. They currently both have the exact same credit, even though the front-runner (22197505) completed on the 16th January 2023 and the other (22197506) is still running. Credit didn't go up after completion, and no extra credit appeared on the aggregation sites. I estimate 22197506 will complete within a week of this post, possibly on Friday.

---

Just to put some meat on the bones, the task I think Dave is referring to that completed on the 27th January is one of mine (22136276) that completed a year ago, not in 2023, so yes, that got fully credited. Since then two immensely long-running HadAM4 tasks have been running on that host, both slowly gaining credit throughout 2022. These started at the same time, but I had to pause one of them (22197506) for a few weeks due to virtual memory issues (a known WSL2 bug, I suspect, as the problem quietly went away). However, it has been slowly aquiriing the same credit as the other, a few weeks behind. They currently both have the exact same credit, even though the front-runner (22197505) completed on the 16th January 2023 and the other (22197506) is still running. Credit didn't go up after completion, and no extra credit appeared on the aggregation sites. I estimate 22197506 will complete within a week of this post, possibly on Friday.

I think that the reason the finished task and the unfinished one have the same credit is because of the credit issue with Hadley models that's been ongoing since at least the end of November. Based on your previous tasks, the models of that batch should have 5 trickles total when completed and the credit is awarded per trickle. Because of the issue, the finished task's 5th trickle is not showing up and therefore no credit is being awarded. The same is very likely going to be the case with the unfinished one. When it finishes and reports, no more credit will be awarded as the 5th trickles will not register. They both uploaded their 4th trickles before the problem started and thus have the same credit.

Looks likely. Another task I looked at from the same batch that finished had 33854 credit awarded.

My Mac's completed and — after the server eventually accepted the trickles — reported three HadCM3 tasks in recent weeks. They're shown on the site as completed and valid, but the trickles aren't shown and no credit has been awarded.

https://www.cpdn.org/result.php?resultid=22249638
https://www.cpdn.org/result.php?resultid=22293889
https://www.cpdn.org/result.php?resultid=22294323

I was assigned another such task a couple of days ago, but the server's yet to send (or finish sending) the files.

https://www.cpdn.org/result.php?resultid=22304961

---

NG

I was assigned another such task a couple of days ago, but the server's yet to send (or finish sending) the files.

https://www.cpdn.org/result.php?resultid=22304961

This task is almost a year old. It's marked as timed out on the previous user. I wonder if that batch has already been closed and so you won't get anything (i.e. actual files of the workunit). Strange that it'd be assigned to you still though.

I was assigned another such task a couple of days ago, but the server's yet to send (or finish sending) the files.

https://www.cpdn.org/result.php?resultid=22304961

This task is almost a year old. It's marked as timed out on the previous user. I wonder if that batch has already been closed and so you won't get anything (i.e. actual files of the workunit). Strange that it'd be assigned to you still though.

Just checked, batch 927 has not been closed though it may well be soon. In the past credit has been given even if batches are closed.

I was assigned another such task a couple of days ago, but the server's yet to send (or finish sending) the files.

https://www.cpdn.org/result.php?resultid=22304961

This task is almost a year old. It's marked as timed out on the previous user. I wonder if that batch has already been closed and so you won't get anything (i.e. actual files of the workunit). Strange that it'd be assigned to you still though.

Just checked, batch 927 has not been closed though it may well be soon. In the past credit has been given even if batches are closed.

The question then is why aren't the files being sent?

I was assigned another such task a couple of days ago, but the server's yet to send (or finish sending) the files.

https://www.cpdn.org/result.php?resultid=22304961

This task is almost a year old. It's marked as timed out on the previous user. I wonder if that batch has already been closed and so you won't get anything (i.e. actual files of the workunit). Strange that it'd be assigned to you still though.

Just checked, batch 927 has not been closed though it may well be soon. In the past credit has been given even if batches are closed.

The question then is why aren't the files being sent?

The server never 'sends' a file: it sends you a url from which your BOINC client can (or will attempt to) download it.

Downloading is subject to the usual vagaries of internet access at your location.

Downloads have been broken for a while now for me as well. This isn't a local network issue.

Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: Hostname download.cpdn.org was found in DNS cache
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: Trying 129.67.193.0...
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: TCP_NODELAY set
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: Found bundle for host download.cpdn.org: 0x7fee70c95690 [serially]
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: Hostname download.cpdn.org was found in DNS cache
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: Trying 129.67.193.0...
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: TCP_NODELAY set
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: connect to 129.67.193.0 port 80 failed: Connection refused
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: Failed to connect to download.cpdn.org port 80: Connection refused
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2461] Info: Closing connection 1024
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: connect to 129.67.193.0 port 80 failed: Connection refused
Thu Feb 2 07:41:11 2023 | climateprediction.net | [http] [ID#2462] Info: Failed to connect to download.cpdn.org port 80: Connection refused

Downloading is subject to the usual vagaries of internet access at your location.

There's no problem with Internet access here. Only with CPDN. I'm getting "Project communication failed" errors in the Event Log, so either the URL's dead or the server hosting it is. I think I'll ditch the task. Maybe even ditch this project altogether as there's very little work for Macs now and I'm not confident that the days of work spent on the previous three tasks have contributed anything to the research.

Thanks, both, for your replies.

---

NG

Dave - I have emailed CPDN and told them this is happening. These old batches need closing. The 1yr deadline is a hangover from the original days of CPDN when the tasks would/could run for many weeks, needs changing for the Hadley models (has been changed for OpenIFS tasks).

I do have a mac version of OpenIFS that could be rolled out with a bit more work. Unfortunately I don't have any the new Apple silicon to test it out on and neither do CPDN (it's built on High Sierra).

Downloading is subject to the usual vagaries of internet access at your location.

There's no problem with Internet access here. Only with CPDN. I'm getting "Project communication failed" errors in the Event Log, so either the URL's dead or the server hosting it is. I think I'll ditch the task. Maybe even ditch this project altogether as there's very little work for Macs now and I'm not confident that the days of work spent on the previous three tasks have contributed anything to the research.

Thanks, both, for your replies.