Help is appreciated! My host 6782743 has recently reported several failed MB/guppi CPU WUs ending with exit code -1073741205 (0xffffffffc000026b) Unknown error number. I noticed the WUs are trashed when I shutdown my computer.

Some more information:
- Changes done to the system before the trashing appeared: I used the Samsung Magician to check my SSD. I didn't actively change any system settings there, but it messed up Window's power options.
- The host is using the Lunatics optimized apps (v0.44).
- BOINC's directories in c:\Program Files and c:\Program Data are excluded from Avira's scan list.

Any idea on this? :(

Hi Atlov
My desktop is a slightly older version of what you have.
I run various work units from 3 projects.
Your power settings are in the control panel
Try resetting them to default.
Another thing I have learned along my years of number crunching is to snooze all tasks, then exit boinc manger before shutting down or rebooting; likewise before installing windows updates or defragging.
There is also the option of doing a system restore, choosing a a restore point prior to running the Magician.
Hope that helps solve your issues.


Kind thoughts

Vicki

---

A city destroyed by an earthquake is an opportunity to Rebuild, redeign & make it a better place to be. Better, stronger, faster like the 6 Million Dollar Man

Help is appreciated! My host 6782743 has recently reported several failed MB/guppi CPU WUs ending with exit code -1073741205 (0xffffffffc000026b) Unknown error number. I noticed the WUs are trashed when I shutdown my computer.

Some more information:
- Changes done to the system before the trashing appeared: I used the Samsung Magician to check my SSD. I didn't actively change any system settings there, but it messed up Window's power options.
- The host is using the Lunatics optimized apps (v0.44).
- BOINC's directories in c:\Program Files and c:\Program Data are excluded from Avira's scan list.

Any idea on this? :(

The same problem has been happening on two of my machines, but I just haven't had the time yet to fully research it. There are currently still 3 in my task list for 6980751. It appears that during BOINC shutdown, some of the running tasks keep trying to restart themselves before BOINC gets completely shut down. If you look in your stored Event Log (stdoutdae.txt) for the entries during a recent shutdown, you'll probably see some lines like:

20-Mar-2016 22:23:15 [SETI@home] Task 18mr10af.21629.6611.5.32.14_0 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:15 [SETI@home] Task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:15 [SETI@home] Task 17my10aa.26761.6227.14.41.84_1 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task 18mr10af.21629.6611.5.32.14_0 using setiathome_v8 version 800 in slot 1
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 using astropulse_v7 version 710 (opencl_nvidia_100) in slot 2
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task 17my10aa.26761.6227.14.41.84_1 using setiathome_v8 version 800 in slot 0
20-Mar-2016 22:23:17 [SETI@home] Task 18mr10af.21629.6611.5.32.14_0 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] Task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] Task 17my10aa.26761.6227.14.41.84_1 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task 18mr10af.21629.6611.5.32.14_0 using setiathome_v8 version 800 in slot 1
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 using astropulse_v7 version 710 (opencl_nvidia_100) in slot 2
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task 17my10aa.26761.6227.14.41.84_1 using setiathome_v8 version 800 in slot 0
21-Mar-2016 08:03:59 [---] Starting BOINC client version 7.6.9 for windows_intelx86
21-Mar-2016 08:03:59 [---] log flags: file_xfer, sched_ops, task, cpu_sched

If you don't have the cpu_sched log flag set, however, you won't see those lines.

I don't think the tasks actually fail completely, though, until after BOINC starts back up again, and not all tasks do ultimately fail. Only CPU tasks seem to end up with computation errors, and even then, not all of those actually fail. As I said, I haven't had the time to fully dig into this yet myself. I have some bits and pieces that I've saved but haven't yet organized.

For now, I've found the simplest thing to do is to make sure I either suspend the project or exit BOINC manually immediately before I shut down either of the two affected machines. By the way, for me it only seems to be currently happening on my two Windows 7 boxes, which are both running BOINC 7.6.9 or later, and I can trace the "task restart during BOINC shutdown" back to at least last December (so I know it's not just a S@h v8 issue). However, I don't think I started getting actual computation errors until late last month.

The same problem has been happening on two of my machines, but I just haven't had the time yet to fully research it. There are currently still 3 in my task list for 6980751. It appears that during BOINC shutdown, some of the running tasks keep trying to restart themselves before BOINC gets completely shut down. If you look in your stored Event Log (stdoutdae.txt) for the entries during a recent shutdown, you'll probably see some lines like:

20-Mar-2016 22:23:15 [SETI@home] Task 18mr10af.21629.6611.5.32.14_0 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:15 [SETI@home] Task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:15 [SETI@home] Task 17my10aa.26761.6227.14.41.84_1 exited with zero status but no 'finished' file
20-Mar-2016 22:23:15 [SETI@home] If this happens repeatedly you may need to reset the project.
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task 18mr10af.21629.6611.5.32.14_0 using setiathome_v8 version 800 in slot 1
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 using astropulse_v7 version 710 (opencl_nvidia_100) in slot 2
20-Mar-2016 22:23:16 [SETI@home] [cpu_sched] Restarting task 17my10aa.26761.6227.14.41.84_1 using setiathome_v8 version 800 in slot 0
20-Mar-2016 22:23:17 [SETI@home] Task 18mr10af.21629.6611.5.32.14_0 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] Task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] Task 17my10aa.26761.6227.14.41.84_1 exited with a DLL initialization error.
20-Mar-2016 22:23:17 [SETI@home] If this happens repeatedly you may need to reboot your computer.
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task 18mr10af.21629.6611.5.32.14_0 using setiathome_v8 version 800 in slot 1
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task ap_29au10ae_B2_P1_00352_20160319_11096.wu_0 using astropulse_v7 version 710 (opencl_nvidia_100) in slot 2
20-Mar-2016 22:23:17 [SETI@home] [cpu_sched] Restarting task 17my10aa.26761.6227.14.41.84_1 using setiathome_v8 version 800 in slot 0
21-Mar-2016 08:03:59 [---] Starting BOINC client version 7.6.9 for windows_intelx86
21-Mar-2016 08:03:59 [---] log flags: file_xfer, sched_ops, task, cpu_sched

If you don't have the cpu_sched log flag set, however, you won't see those lines.

Indeed, I see the same. I'll reset the project and reinstall Lunatics as soon as my queue is empty. For now, BOINC will be suspended before the shutdown.

There is something else to mention, maybe we also have this in common: I run a software called Genymotion (emulation of Andriod via VirtualBox), which I usually don't quit manually before the shutdown. During the shutdown VirtualBox delays things with some error messages. If Genymotion is not running, the shutdown is faster and the WUs are NOT trashed.

Indeed, I see the same. I'll reset the project and reinstall Lunatics as soon as my queue is empty. For now, BOINC will be suspended before the shutdown.

There is something else to mention, maybe we also have this in common: I run a software called Genymotion (emulation of Andriod via VirtualBox), which I usually don't quit manually before the shutdown. During the shutdown VirtualBox delays things with some error messages. If Genymotion is not running, the shutdown is faster and the WUs are NOT trashed.

I don't know that a reset and reinstall will have any effect. It appears to me that the problem only occurs with the more recent versions of BOINC, but whether the tasks are trying to restart themselves during BOINC shutdown, or whether BOINC is triggering it is hard to tell. Sometimes all the tasks try to restart, other times only a few of them, and other times it doesn't happen at all.

I rechecked all my boxes again yesterday and found that my Windows 8.1 machine, which is also on BOINC 7.6.9, had a couple examples of the "restart during shutdown" phenomenon, so it apparently isn't just a Windows 7 thing. My two Windows XP boxes and the one on Windows Vista, all of which are running earlier versions of BOINC, don't appear to have ever had the problem.

Those two instances on the Windows 8.1 box, which shuts down automatically once a day on weekdays, occurred on May 12 and May 19. Both times, all 17 running tasks attempt to restart once. However, none of those ended up with errors when the machine rebooted.

I'm not running Genymotion on any of my machines. In fact, of the 3 machines that I've now seen experiencing this problem, my daily driver is the only one that I use for anything other than crunching. It gets shut down manually and I always close everything else (browser, e-mail client, etc.) first before shutting down, so there really shouldn't be anything other than the bloated OS holding things up.