Posts by Ageless

21) Questions and Answers : Windows : Is this a new error message (Message 1830398)
Posted 19 days ago by Profile Ageless
No, it's not a new message, it's been around for a while already. Best explanation is by Joe Segur, here:
Thanks for seeing that, a sanity check within the app has seen a pulse too strong to be possible, so something went wrong. The Lunatics CPU and OpenCL GPU apps for MB all have those checks now for each of the 5 signal types. The apps generally do a BOINC temporary exit with a message like the one you saw, and a time value so BOINC will restart the task from the last checkpoint 5 minutes later. That restart from checkpoint rereads the WU file, etc., so is starting fresh from a point in time before the problem. That may often be all that's needed and the task can go on and finish normally (as appears to have happened for your case). If not, another temporary exit will happen, and after 100 of those BOINC will abort the task. We hope the user will notice before that 8 hours and 20 minutes, at least try the suggested reboot and consider whether some cleaning or other maintenance is due.

There's one exception which doesn't use that temporary exit approach. For the OpenCL GPU apps the sanity check for Autocorr signals needed to be at a lower level. Therefore that sanity check also requires that a result_overflow happen, and since it has run all the way to that outcome the last checkpoint isn't a reliable restart point. Hence, for that combination of unusually strong Autocorrs with overflow the OpenCL GPU application will do an error exit.

It should be understood that these sanity checks have not been fully tested. They're really meant to catch things caused by a memory bit flip or other random occurrence which is nearly impossible to reproduce. So we had to simply do normal testing of the apps and thereby gain confidence the checks would not kick in when they shouldn't.
22) Questions and Answers : Web site : Link which I can not explain?? (Message 1830397)
Posted 19 days ago by Profile Ageless
Neither BOINC nor Seti make use of that address. You can remove the connection/allowance through the firewall.

I've just reattached my device to Seti, the BOINC entry in the firewall is still without a custom filter. So it isn't Seti that did it, and it certainly isn't BOINC that did it. At least, not the BOINC by Berkeley, downloaded from the Google Play Store, here.

Do you use Berkeley's BOINC, or is it downloaded from elsewhere and built by someone else?

Also, as far as I can figure from the different sites reviewing NoRoot Firewall, and their own documentary, only the user of the phone can add a custom filter for an application, the application cannot do that itself. So, has your phone been in the hands of someone else at some point?
23) Questions and Answers : Web site : Link which I can not explain?? (Message 1830323)
Posted 19 days ago by Profile Ageless
I have NoRoot firewall on my phone as well. What you show is a custom filter under the BOINC entry, so probably something you added yourself.
24) Questions and Answers : Windows : User XXXXX (ID: YYYYYY) is not accepting private messages from you. (Message 1830312)
Posted 19 days ago by Profile Ageless
What it means is that the other person has your account(s) on ignore through the message filtering option of the forum preferences.
25) Questions and Answers : Web site : Link which I can not explain?? (Message 1830089)
Posted 20 days ago by Profile Ageless
A connection by the BOINC client, or by the Seti web site/forums? If the latter, it may be something someone has in their signature.
26) Questions and Answers : Macintosh : Status: Postponed (Message 1829955)
Posted 21 days ago by Profile Ageless
"Postponed: Waiting to acquire slot directory lock. Another instance may be running ".

This means that BOINC finds that another task is already occupying the slot that it tries to assign the new starting task to.

Best try with our latest development version if you can reproduce that error:
Macintosh 7.6.33
- boinc_7.6.33_macOSX_x86_64.zip
- boinc_7.6.33_x86_64-apple-darwin.zip
- boinc_7.6.33_macOSX_SymbolTables.zip
27) Questions and Answers : Web site : Server unreachable (Message 1829458)
Posted 23 days ago by Profile Ageless
Bookmark http://setiathome.berkeley.edu/sah_status.html, it's a page that will give the status of all servers for the project. If for instance you see that the database servers are off line, or the scheduling/scheduler server, that means the project isn't reachable.

Of course, if such a catastrophe happens that the main database server resets itself and isn't allowing anyone to get onto it (accompanied by the message You don't have permission to access /forum_forum.php on this server.) the server status page won't be reachable either.

In any case, all those are happening at the project, there is no need for your ISP to do anything. Most of the times when only one website isn't reachable, it's due to something having happened to that website. If all websites are unreachable, it's time to ask the ISP and check cables, DNS server settings and such things.

One further helpful site is http://www.downforeveryoneorjustme.com/
28) Questions and Answers : Windows : No work available to process (Message 1829393)
Posted 23 days ago by Profile Ageless
The server status page shows the scheduling server and upload server as off line, so the message you get is correct, there is no work available from this project at this time.

Not an error, just an informational message.
29) Questions and Answers : Macintosh : Mac Pro 12 Core (24 treads) and only 8 working, others saying Wait for shared Memory (Message 1829214)
Posted 24 days ago by Profile Ageless
Having loads of memory doesn't mean you have enough shared memory.

From this BOINC FAQ:
BOINC science applications use shared memory to communicate with the core client, and a certain amount is reserved for each current task, whether running or waiting. The default configuration of a multi-CPU Mac (regardless of how much RAM is installed) is sometimes inadequate to support several projects at once.

If you get this error message, which is usually followed by a “Couldn't start or resume: -144” error in the current computation, it may help to set your BOINC general preference “Leave applications in memory while suspended?” to no; however, the problem will probably just become less frequent rather than being solved, and moreover some projects’ applications, those that save checkpoints infrequently (or erratically), may not take well to this setting.

A better solution is available—if you don’t mind reconfiguring your system’s kernel a little. The method, accompanied by a more detailed description of the issue, is outlined in Configuring Shared Memory on Mac OS X from Spy Hill Research (hosts of the BOINC Pirates@home project). It involves creating a text file “/etc/sysctl.conf” that contains the following commands:

kern.sysv.shmmax=16777216
kern.sysv.shmmin=1
kern.sysv.shmmni=128
kern.sysv.shmseg=32
kern.sysv.shmall=4096

and rebooting the computer. This will quadruple the default allocation of shared memory.

Note that the problem may manifest with slightly different error messages: “Can't create shared memory: system shmat” followed by “Couldn't start or resume: -146”.
30) Questions and Answers : Preferences : Is there a way to have SETI@Home send data more often? (Message 1829213)
Posted 24 days ago by Profile Ageless
In BOINC 7 not the additional days amount decides when to contact the project, but the store at least amount. Having it set to 10 days isn't really useful here, although I understand the thought behind it: because of the maximum of 100 tasks per hardware source, setting store at least to 10 days will probably bring more contact points asking for more work (10 days) than a way lower amount does.

However, because Seti has the 100 tasks per hardware resource, what it does is tell BOINC to back off for an X-amount of time when it's reached the 100 tasks in progress. Now BOINC will probably run the cache empty before asking for more work, because of the 10 days being the low water mark and only at that point will BOINC report the uploaded tasks and ask for new work.

Reducing the store at least value to a lower value allows the storing of 100 tasks and will top it off again once more tasks are run, uploaded and reported.
31) Questions and Answers : Macintosh : "Wont get new tasks" (Message 1828318)
Posted 28 days ago by Profile Ageless
"Won't get new tasks" is the message you get when you have set "No new tasks" in BOINC Manager. The button shows the option you have, now 'Allow new tasks'.

BOINC Manager->View->Advanced view->Projects tab->Select Seti@Home->click 'Allow new tasks'. (This will change to 'No new tasks')
32) Questions and Answers : Preferences : Why am I not getting enough tasks? (Message 1828181)
Posted 29 days ago by Profile Ageless
Seti has a hard limit on tasks it allows computers to store. This limit is 100 tasks per CPU and 100 per GPU. This is done so that the database doesn't grow too much out of proportions, which causes instability problems through which the whole thing can crash - happened before.

At the moment there's no work, because the database server is down. See the server status page for that. As far as I know, they had trouble with getting the kernel updated, and are going back to the previous version.
33) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1827131)
Posted 28 Oct 2016 by Profile Ageless
I think you will find that the first line over-rules the second line, so the effect is no space usage.

I don't believe that to be true anymore, and it doesn't show it in any of his logs. The opposite even.
10/25/2016 5:59:07 PM | | Disk: 111.69 GB total, 44.41 GB free
10/25/2016 5:59:07 PM | | max disk usage: 44.46GB

10/26/2016 7:03:20 PM | | Disk: 111.69 GB total, 40.94 GB free
10/26/2016 7:03:20 PM | | max disk usage: 40.99GB

10/27/2016 9:38:59 AM | | Disk: 111.69 GB total, 47.63 GB free
10/27/2016 9:38:59 AM | | max disk usage: 47.70GB

10/27/2016 12:13:05 PM | | Disk: 111.69 GB total, 47.73 GB free
10/27/2016 12:13:05 PM | | max disk usage: 47.80GB

10/27/2016 1:19:00 PM | | Disk: 111.69 GB total, 47.73 GB free
10/27/2016 1:19:00 PM | | max disk usage: 47.80GB


Of course he shouldn't install Battlefield 1 on that drive, as that takes considerable more disk space than he has available (50GB), but other than that, it isn't inhibiting the question of work, not yet at least. However, just to make sure it works with all projects out there, and some do run older back-end software that does things differently with a zero in that place, it's best to add a value above zero.

Your buffers sizes look wrong...
His present buffer settings give him 100+ tasks already, so unless something else changed since I stopped reading NC, I think he's pushing against the 100 tasks boundary with his present settings and therefore increasing the cache size won't do much good on this project.

@BilBg, thanks for that. I seem to remember that from skimming through before, but couldn't so quickly find it (and my time here is limited these days).
34) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1827068)
Posted 28 Oct 2016 by Profile Ageless
I have Use at most N% of CPU time set at 75% since you suggested having 3 of the 4 cores available and the last one for use with the GPU.

Use at most N% of CPU time is throttling.
Use at most N% of CPUs is set the amount of cores to run. I know, it can be confusing, but just mouse over the options and it'll give a hint.

About the preferences, things I notice:
<max_ncpus_pct>75.000000</max_ncpus_pct>
That's the N% of CPUs, so that's good.

<disk_max_used_gb>0.000000</disk_max_used_gb>
This however, with it you tell BOINC to use 0 GB of disk space. Best put in a real number, there are some versions out there that use 0 as in use all but I have to look up if 7.6.22 is one of those. Best add a value larger than 0, though.

<disk_max_used_pct>100.000000</disk_max_used_pct>
You're lucky you set this, as now it'll use all the free space available.
Seti isn't so demanding though on disk space, it's not as if its tasks are gigabytes big. There are some projects out there that do, yes.
In any case, set some hard boundaries. In all my years I ran Seti I've always done with Use no more than 15GB and Leave at least 0.1GB free. :)

Weird. So I aborted the 3 tasks that were suppose to use the NVIDIA GPU, updated my connection to the server, then it redownloaded 3 new tasks and now it's working properly.

Well, it probably only needed a nudge. :)
35) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1827031)
Posted 27 Oct 2016 by Profile Ageless
12au16aa.10035.12337.8.35.104_2 doesn't seem to be running, no.
Or at least, it doesn't show up in any of the checkpoints in the minute 22 seconds of that log, perhaps that it checkpointed afterwards.

But... I do see a lot of suspending/unsuspending of tasks. Are you running BOINC with throttle through Use at most N% of CPU time? What have you set that value at?
Same question for Suspend when non-BOINC CPU is above N%?

(Best would be to post the contents of the global_prefs_override.xml file from your BOINC data directory at C:\ProgramData\BOINC\ .. you can open this file in Notepad, no special XML reader is necessary.)

If when you check your log now the GPU task still hasn't checkpointed, abort it. Best abort the other SoG as well, perhaps that your GPU just cannot do them. I'm not up-to-date on the development of the SoG applications, as I hardly read Number Crunching anymore. Perhaps that one of the others who earlier posted in this thread can take that up.
36) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1826994)
Posted 27 Oct 2016 by Profile Ageless
How can you tell that installing the new drivers fixed the issue?

Good question! I cannot from here, but we can add another debug flag that can show that. Tasks like these should checkpoint every so often to write their state to disk. This in case somewhere along the way something goes awry and BOINC has to be restarted, that you don't have to start the task from the start.

You can add a check mark to checkpoint_debug in the Event Log options to enable this flag. There is another flag with a lot more details that we can add as well, but since that one will poll the system several times per second and dump all that information into the Event Log, I'll hold off with offering that one for now.
37) Questions and Answers : Preferences : Specify or Dissable which GPU's are able to execute tasking ? (Message 1826982)
Posted 27 Oct 2016 by Profile Ageless
When you click the "preferences" button inside of the GUI it routes to the following web page, http://setiathome.berkeley.edu/prefs.php?subset=global

But that's also what the mouse-over on that button says: View and modify your computing preferences.

The project preferences differ from project to project, and while some projects do offer a direct link to them from the manager, this project chose not to do so. You can only have so many buttons before it gets cluttering.

The project preferences are easily reached from your account, Preferences for this project.
38) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1826978)
Posted 27 Oct 2016 by Profile Ageless
I did not find cpu_op_debug in Event Log options

Sorry, that was a typo from me, meant to leave on sched_op_debug, which you have done. :)

I changed the number of CPUs thinking that if I freed up a CPU, it would get one of the tasks that included my NVIDIA GPU to start. Obviously, that is incorrect so I did switch it back to using all 4 of the CPUs.

Well, it's not incorrect, I just don't know if it's needed for CUDA. Therefore leaving that to someone who does know.

But seeing how BOINC is now executing a task on your GPU using the OpenCL application, it might be wise to tell BOINC to use 3 cores again and leave the fourth free for the GPU:
10/27/2016 12:13:05 PM | SETI@home | [coproc] Assigning NVIDIA instance 0 to 12au16aa.10035.12337.8.35.104_2
10/27/2016 12:13:06 PM | SETI@home | [task] task_state=EXECUTING for 12au16aa.10035.12337.8.35.104_2 from start
10/27/2016 12:13:06 PM | SETI@home | [cpu_sched] Restarting task 12au16aa.10035.12337.8.35.104_2 using setiathome_v8 version 812 (opencl_nvidia_SoG) in slot 4


(Earlier when I looked in your task list, I only saw the CUDA50 task, but now I see you also still have those two (one, two) SoG tasks which use OpenCL and have a deadline of the 31st of October, so still time enough to finish them).

Edit: In the end I think it was your drivers that were flummoxed. The driver reinstall fixed that.
39) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1826959)
Posted 27 Oct 2016 by Profile Ageless
Thank you.
You can turn off cpu_sched_debug and turn on cpu_sched and task_debug.
Also leave on file_xfer, sched_ops, and task (as these are the defaults).
Plus leave coproc_debug and cpu_op_debug on.
(BOINC Manager->Options->Event Log options...)
(With the above flags, wait about 5 minutes, then copy and post the log from the event log, last 50 or so lines).

I also see you changed back from your previous log:
10/25/2016 6:03:23 PM | | Number of usable CPUs has changed from 4 to 3.
10/25/2016 6:03:23 PM | | max CPUs used: 3

.. to using all four CPU cores?
Not sure if that's a good idea on an Nvidia card when using CUDA, but I'll leave that for an Nvidia prof to tell. I thought it was OpenCL that likes to have a free a CPU core, not sure if CUDA does that as well.
40) Questions and Answers : Windows : SETI@Home not using my GPU (Message 1826919)
Posted 27 Oct 2016 by Profile Ageless
Could you turn on some of the debug flags for me?
BOINC Manager->View->Advanced view->Options->Event Log options...->check coproc_debug, cpu_sched_debug and sched_op_debug->OK

At this point exit BOINC and restart it:
BOINC Manager->File->Exit->Acknowledge if you have Exit dialog on.
Start->All Programs->BOINC->BOINC Manager.

Next tell BOINC to contact the server:
BOINC Manager->Projects tab->select Seti@Home->click Update.

Wait about a minute, then open the event log, BOINC Manager->Tools->Event Log (or press CTRL+SHIFT+E).
Now copy and post all the lines (Copy All).


Previous 20 · Next 20


 
©2016 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.