IMHO there was no need to update, but I did say that as no lower end GPU's are being used.

Cheers.

Or does each aistub only have the appropriate info for the app it accompanies so that all aistub info gets merged into the app_info.xml

If I remember correctly, the original .aistub file from the original Lunatics installer had all the apps and plan classes listed. So it produced an app_info for everything from CUDA23 to CUDA50 along with sections for 32 bit x86 to 64 bit x86 and everything in between. You should really do some weeding of the .aistub files to limit the size of the app_info. The later installers generate a pretty efficient app_info just from the selections made in the installer. But there still can be some superfluous entries that can be trimmed down or out.

. . OK, I ran the aimerge and everything seemed to go OK, it was too quick to see the aistub names but I m sure there were 3 of them. I downloaded about 20 new WUs expecting to see them as v8.22 but they are still v8.20. It seems there is something else to do ...

Stephen

??

IMHO there was no need to update, but I did say that as no lower end GPU's are being used.

Cheers.

. . Yep, that is what everyone is saying but Miklos seems to feel that v8,22 is faster. It will sort itself out in the end.

Stephen

. .

All your old work will show as being done by the old app, but new work will show as correct.

It's the same as going from stock to Lunatics.

Cheers.

All your old work will show as being done by the old app, but new work will show as correct.

It's the same as going from stock to Lunatics.

Cheers.

. . I expected that with existing tasks, but I was expecting the new downloads to be v8.22.

. . I stopped BOINC and went through app_info.xml (Keith is right I need to clean it up and create a more efficient version) but I discovered the problem. The download Keith linked me to has a slight discrepancy, it has set the app info to v8.21 .... ( was that a transient version at some stage?) so when I had attempted to rename some files as v8.22 to get r3584 to kick in I sadly joined Miklos and trashed them. So I will now manually change that to v8.22.

. . Miklos, if you are reading this you might want to scan through your app_info.xml (it should be right near the end) and see if you have the same problem. If it says v8.21 you will need to replace all occurrences of v8.21 with v8.22.

Stephen

:(

All your old work will show as being done by the old app, but new work will show as correct.

It's the same as going from stock to Lunatics.

Cheers.

. . I expected that with existing tasks, but I was expecting the new downloads to be v8.22.

. . I stopped BOINC and went through app_info.xml (Keith is right I need to clean it up and create a more efficient version) but I discovered the problem. The download Keith linked me to has a slight discrepancy, it has set the app info to v8.21 .... ( was that a transient version at some stage?) so when I had attempted to rename some files as v8.22 to get r3584 to kick in I sadly joined Miklos and trashed them. So I will now manually change that to v8.22.

. . Miklos, if you are reading this you might want to scan through your app_info.xml (it should be right near the end) and see if you have the same problem. If it says v8.21 you will need to replace all occurrences of v8.21 with v8.22.

Stephen

:(

. . That download is a real DUD! The aistub file also fails to reference the file Multibeam_Kernels_r3584.cl so it does NOT create the kernel files for the GPU or the wisdom file. Still no joy. I will have to rebuild it: and I was trying to avoid all of this :(

Stephen

:(

All your old work will show as being done by the old app, but new work will show as correct.

It's the same as going from stock to Lunatics.

Cheers.

. . I expected that with existing tasks, but I was expecting the new downloads to be v8.22.

. . I stopped BOINC and went through app_info.xml (Keith is right I need to clean it up and create a more efficient version) but I discovered the problem. The download Keith linked me to has a slight discrepancy, it has set the app info to v8.21 .... ( was that a transient version at some stage?) so when I had attempted to rename some files as v8.22 to get r3584 to kick in I sadly joined Miklos and trashed them. So I will now manually change that to v8.22.

. . Miklos, if you are reading this you might want to scan through your app_info.xml (it should be right near the end) and see if you have the same problem. If it says v8.21 you will need to replace all occurrences of v8.21 with v8.22.

Stephen

:(

. . That download is a real DUD! The aistub file also fails to reference the file Multibeam_Kernels_r3584.cl so it does NOT create the kernel files for the GPU or the wisdom file. Still no joy. I will have to rebuild it: and I was trying to avoid all of this :(

Stephen

:(

You don`t necessarily need the Multibeam_Kernels_r3584.cl in appinfo.xml.
I had to remove it to get my own appinfo running.

---

With each crime and every kindness we birth our future.

Thank you everyone, but looking at my manager I do not see any SETI wu's. On the other hand when I saw under SETI Computers there they were in progress and waiting for credit. As well as a couple of Astroplulse 7 units returned today. Yet, I see nothing among all tasks listed, not a one.

Sorry, very confused here. Should I go back to lunatics and download another app?

Miklos

---

which computer are you talking about? The windows 7 or windows 10?

edit..

Looking at the windows 7 machine, there are 78 work units in progress. The send date is 1 day before you posted in this forum. So whatever happen there, happen before we gave you any advice on changing your applications. Most likely these are ghosts. Either you can attempt to have the server resend them to you or allow them to error out when the deadline passes. Reloading the lunatics or applications is not going to change this.

Zalster

---

Thank you Zalster. Perhaps I should reset the project on the Windows 7 only? Other possibilities mentioned here are out of my league.

Miklos

---

I don't think resetting the project will force a download of the lost work units. If anything, it might force them to go to error.

The recovery method is probably your best bet.

I know there is a detailed description on how to do it somewhere in the NC forum but the condensed version goes something like this

For this you will need both the BOINC manager open and Event log.

in the BOINC manager, select the seti@home project and click Update on the left hand side of the manager.

Once it has connected to the server, you should get a report back in the event log manager saying connecting to server and either you got work or no work available.

Immediately hit update again, you should see in event log, connecting to server, but before you get a response from the server, you have to close out the BOINC manager. This requires a lot of timing. ( I usually just hit update then move to close the BOINC manager.) Next. Walk away. This is the hardest thing to do..... Walk away from the machine.

Come back in 10 minutes (I know they say 5 minutes but give it a little extra)

Relaunch the BOINC manager. It should immediately attempt to connect to the server, when it does, it MAY download 20 of the previously ghosted work units.

Once it does. You will have to repeat this entire steps to download most if not all of the ghosted work units.

I've done it, so I know it works but it's time consuming.

Good luck

---

Could I just abandon those ghost units and get new ones instead?

---

4/29/2018 1:54:23 PM | SETI@home | update requested by user
4/29/2018 1:54:23 PM | SETI@home | Sending scheduler request: Requested by user.
4/29/2018 1:54:23 PM | SETI@home | Requesting new tasks for NVIDIA GPU
4/29/2018 1:54:25 PM | SETI@home | Scheduler request completed: got 0 new tasks
4/29/2018 1:54:25 PM | SETI@home | Not sending work - last request too recent: 284 sec

This is what I got when I asked for more tasks.

---

284 secs is 4.7 minutes... Minimal time between updates is 5 minutes, unless you are using a really old BOINC manager (which you are not). I recommended 10 minutes. Remember what I said about walking away from the machine?? I understand the urge to relaunch and hit update.

I'm not sure but if you reset it might, might abandon those tasks but then you get hit with the limiting number of work units because you thrashed so many.

---

I did not run the suggested steps, but I RESET Project. The new message was:

4/29/2018 2:27:18 PM | SETI@home | Resetting project
4/29/2018 2:27:24 PM | SETI@home | Master file download succeeded
4/29/2018 2:27:29 PM | SETI@home | Sending scheduler request: To fetch work.
4/29/2018 2:27:29 PM | SETI@home | Requesting new tasks for NVIDIA GPU
4/29/2018 2:27:31 PM | SETI@home | Scheduler request completed: got 0 new tasks
4/29/2018 2:27:31 PM | SETI@home | No tasks sent
4/29/2018 2:27:31 PM | SETI@home | No tasks are available for AstroPulse v7
4/29/2018 2:27:31 PM | SETI@home | No tasks are available for SETI@home v8
4/29/2018 2:27:31 PM | SETI@home | Tasks for CPU are available, but your preferences are set to not accept them
4/29/2018 2:27:31 PM | SETI@home | Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them
4/29/2018 2:27:33 PM | SETI@home | Started download of arecibo_181.png
4/29/2018 2:27:33 PM | SETI@home | Started download of sah_40.png
4/29/2018 2:27:33 PM | SETI@home | Started download of sah_banner_290.png
4/29/2018 2:27:33 PM | SETI@home | Started download of sah_ss_290.png
4/29/2018 2:27:34 PM | SETI@home | Finished download of sah_40.png
4/29/2018 2:27:34 PM | SETI@home | Finished download of sah_banner_290.png
4/29/2018 2:27:36 PM | SETI@home | Finished download of arecibo_181.png
4/29/2018 2:27:36 PM | SETI@home | Finished download of sah_ss_290.png
4/29/2018 2:29:06 PM | SETI@home | update requested by user
4/29/2018 2:29:07 PM | SETI@home | Sending scheduler request: Requested by user.
4/29/2018 2:29:07 PM | SETI@home | Requesting new tasks for NVIDIA GPU
4/29/2018 2:29:09 PM | SETI@home | Scheduler request completed: got 0 new tasks
4/29/2018 2:29:09 PM | SETI@home | Not sending work - last request too recent: 97 sec

None of the above mentioned downloaded units are visible among my tasks. I looked at all tasks.

---

. . That download is a real DUD! The aistub file also fails to reference the file Multibeam_Kernels_r3584.cl so it does NOT create the kernel files for the GPU or the wisdom file. Still no joy. I will have to rebuild it: and I was trying to avoid all of this :(
(

You don`t necessarily need the Multibeam_Kernels_r3584.cl in appinfo.xml.
I had to remove it to get my own appinfo running.

. . OK, so how do I get it to work? Something is not happy ... :(

Stephen

? ?

As I said earlier, you do not need the 8.22 app, I'm quite happy with the 8.20 app and some actually report that the older app is slightly faster than the newer 1.

A fix would be to either edit the app_info or rerun the Lunatics installer.

Cheers.

I deleted SETI and after an hour reinstalled it. Result: got 1 8.22 unit that is visible. When it was returned received many cuda 42's. Is there a next step for me? But not a one with 8.22.

thanks

---

Thank you everyone, but looking at my manager I do not see any SETI wu's. On the other hand when I saw under SETI Computers there they were in progress and waiting for credit. As well as a couple of Astroplulse 7 units returned today. Yet, I see nothing among all tasks listed, not a one.

Sorry, very confused here. Should I go back to lunatics and download another app?

Miklos

. . When you use the "Tasks" button on the LEFT you are NOT looking at your computer, you are seeing your account details AT SETI.

. . The only tasks from that information that should appear on your computer are those under "In Progress" which means tasks sent to your computer but no result returned yet.

. . Any tasks that are "In Progress" should appear under the "Tasks" tab at the TOP of your BOINC Manager as long as the TOP button of the commands section on the LEFT says "Show Active Tasks". If it says "Show All Tasks" then you need to click on it.

. . Having said that the only completed tasks from your computer with the 2 x 780ti cards which IT "received" in the last 5 days are AstroPulse tasks. So when you ran the Lunatics install, was that about the 25th of April? If so then did you go through each screen of the installer and select the app to use for ALL types of tasks that you want to and CAN run?

. . Try running the installer again and read each screen carefully making the appropriate selection on every page, and this includes unchecking any box that indicated hardware you do not have. Only when ONLY the box that matches your hardware is selected on each page go the next. Some pages, such as Intel GPU and AMD GPU will have NO boxes selected, take special note of the first option at the top of each page. You are still receiving AP work so you need to make sure you are still set to receive MB tasks for your Nvidia GPU.

. . Hope that helps ...

Stephen

> >

Would you please send me a link so I can get to the correct site, just in case?

Thanks

---