All Asteroid@Home tasks on my old Mac OSX machine crash almost immediately reporting 'Computation Error'. It is a GenuineIntel Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz [x86 Family 6 Model 15 Stepping 11](2 processors. I am running Darwin 9.8.0 OS and BOINC 6.12.35 (the latest version for this processor, I believe). Is there a fix?

Alan Barnes

I'm not at all familiar with Mac OS version numbering but I have a hunch your OSX is too old. Almost every BOINC project website has a page named Applications which lists the platforms the project supports and the name and version of the application they send to each platform. The home page typically has a link to Applications.

On the Applications page for this project it says OS 10.4 and 10.5 are supported so I gather your 9.8 is too old. Mac is not well supported in the BOINC world, mostly because developers frequently don't have access to a Mac.

---

BOINC FAQ Service
Official BOINC wiki
Installing BOINC on Linux

The question now is why the scheduler sends tasks to his (or anyone else's) Mac OS Computers
when the OS don't meet the app requirements (there have to be check for this server side):

<stderr_txt>
dyld: lazy symbol binding failed: Symbol not found: ___bzero
Referenced from: /Library/Application Support/BOINC Data/slots/4/../../projects/asteroidsathome.net_boinc/period_search_10210_x86_64-apple-darwin__sse2
Expected in: /usr/lib/libSystem.B.dylib

http://asteroidsathome.net/boinc/result.php?resultid=24079883

---

- ALF - "Find out what you don't do well ..... then don't do it!" :)

The application is send because there is no definition for the version of OS. But this definition is default nowhere, usually everything should work on the current os but I think that not much people expect that someone is using so old version. I think that many projects doesn't check version of os. And without having any Mac I can't do proper testing. I already tried some vm, but not so old version, and without the right hardware I also can't be sure that everything will work on the specific hardware.

I have spent some time to get Mac PC but not luck here.

Apple has no office here in Czech republic. Mailing with AppStore failed. One volunteer in Czech rep. didn't respond to mail.

And finally, I cann't manage to run Hackintosh on my hw. It's very unstable.

So we are still looking for any mac for xCode 3.6 and with SSE3 (+AVX better) CPU.

Error is: Estimated computation size 1,380,023 GFLOPs
Computer is: jiMac Pro 2017, 3.2GHz 8-Core Intel Xeon W
Graphics: Radeon Pro Vega 64X 16GB
running: macOS 13.4.1(22F82)

I am getting the same error on all 34 jobs and have stopped processing. Any ideas?
*****************
Application
Period Search Application 102.11 (sse3)
Name
ps_230619_input_5187_143
State
Computation error
Received
Sunday, June 25, 2023 at 06:35:22 PM
Report deadline
Thursday, July 06, 2023 at 06:35:21 AM
Estimated computation size
1,380,023 GFLOPs
CPU time
---
Elapsed time
---
Executable
period_search_10211_x86_64-apple-darwin__sse3I

---

I'm not at all familiar with Mac OS version numbering but I have a hunch your OSX is too old. Almost every BOINC project website has a page named Applications which lists the platforms the project supports and the name and version of the application they send to each platform. The home page typically has a link to Applications.

On the Applications page for this project it says OS 10.4 and 10.5 are supported so I gather your 9.8 is too old. Mac is not well supported in the BOINC world, mostly because developers frequently don't have access to a Mac.

Please don't confuse Darwin 9.8 with OS 10.4 or 10.5. They are not related in the way that you think. Darwin 9 *is* OS 10.5. Darwin is the underlying UNIX code and MacOS refers to the UI riding on top of it.

I'm not sure if the OP's issue is age of the hardware or software. I'm seeing ~ 60% failure rate on tasks on my two Mac computers. One is old (2011 i5) and one is new-ish (2020 M1). Both are running > 10.5. The stderr readout always shows:

<![CDATA[
<message>
process exited with code 2 (0x2, -254)</message>
<stderr_txt>

Error: Number of lc points is greater than POINTS_MAX = 1000

</stderr_txt>
]]>

I have not seen any errors on my Windows Laptop. (2018 i5).

Error is: Estimated computation size 1,380,023 GFLOPs
Computer is: jiMac Pro 2017, 3.2GHz 8-Core Intel Xeon W
Graphics: Radeon Pro Vega 64X 16GB
running: macOS 13.4.1(22F82)

I am getting the same error on all 34 jobs and have stopped processing. Any ideas?
*****************
Application
Period Search Application 102.11 (sse3)
Name
ps_230619_input_5187_143
State
Computation error
Received
Sunday, June 25, 2023 at 06:35:22 PM
Report deadline
Thursday, July 06, 2023 at 06:35:21 AM
Estimated computation size
1,380,023 GFLOPs
CPU time
---
Elapsed time
---
Executable
period_search_10211_x86_64-apple-darwin__sse3I

If you look at the Task details, you can see the stderr text, which shows the same error that I'm seeing on my Macs. Here's the link to the latest errored task from your Mac Pro.
https://asteroidsathome.net/boinc/result.php?resultid=386275105

Scroll to the bottom to see the stderr text coming back from the task. I'm hopeful that this comment in the stderr can help the developers determine what's going on:

Error: Number of lc points is greater than POINTS_MAX = 1000

FWIW, this behavior (high percentage of errors and the specific lc points message) are also showing up on Android devices.

I am running with a 2020 iMac using an Apple Silicon M. I have the most current macOS 13.4.1(22F82). Since this is being reports on different hardware but similar up to date MacOS, I suspect the issue may lay in how Astroid@Home is working with the latest MacOS update since the issue appeared. immediately following my installing the latest update.

I am running with a 2020 iMac using an Apple Silicon M. I have the most current macOS 13.4.1(22F82). Since this is being reports on different hardware but similar up to date MacOS, I suspect the issue may lay in how Astroid@Home is working with the latest MacOS update since the issue appeared. immediately following my installing the latest update.

I'm also seeing the same behavior on my 2011 iMac running Mac OS 10.13.6. Same percentage of failures with the same message in the stderr output.

bump. This is still happening and I haven't seen any engagement.

bump. This is still happening and I haven't seen any engagement.

Anybody out there that can look at lc > 1000 errors for MacOS & linux? I'm seeing tons of Computation Errors.

bump. This is still happening and I haven't seen any engagement.

Anybody out there that can look at lc > 1000 errors for MacOS & linux? I'm seeing tons of Computation Errors.

Hi Donald,
What do you mean by "& linux"? As I can see your issues id only with the MacOS application, isn't it?

And yes there are some new WUs which carries larger number of Light points than the older limit of 1000.
Having updated app for MacOS is in the Task List, but I can't say any specific time that it'll take.

Meanwhile you may want to switch your Macs to a different project and get back once we have updated version.

Best,
Georgi

---

“The good thing about science is that it's true whether or not you believe in it.” ― Neil deGrasse Tyson

I appear to labour under a similar suite of constrictions.

I only recently started Asteroids@Home, so the sample size is small, but of the 148 total WUs, ignoring the 19 still being processed and such, 84 were valid and 45 returned computation errors.

Asteroids@home is running on an up to date Mac mini M2, with BOINC calculation set at 100% for 75% of the time.
Possibly relevant: I set up A@H because with World Community Grid the Mapping Cancer Markers project WUs tend to get stuck in a loop, most often early in the processing, but not other projects (if they have WUs available). So A@H also getting the short end of the computational stick is... frustrating.

Just now I had set A@H to not get new tasks, and planned to check semi-periodically, but given that this problem seems to persist for months now, I was reluctant to get my hopes up. One last try, and: All new WUs, all eight, had computation errors within seconds. So back to "no new tasks" A@H goes...

Hi Geronimo,
I really appreciate your feedback.

As I said, during the time all other applications was updated to the requirements added by the new WUs structure, exceeding the old limit of 1000 light points. Unfortunately the team wasn't able to react quickly enough and prepare new macOS apps.

We'll do our best to fix that ASAP (in relative terms).

May I contact you later for this matter, if it is OK with you of course?

Georgi

---

“The good thing about science is that it's true whether or not you believe in it.” ― Neil deGrasse Tyson

We'll do our best to fix that ASAP (in relative terms).

May I contact you later for this matter, if it is OK with you of course?

Georgi

Sure, I set the forum to send me a mail once you PM me, or you respond here, either works.

Add me to the list. I have many, many jobs that errored out. Mac mini 2018 i7. running Monterey 12.6.9

Thanks for the support guys!

I'll keep you all in the loop.

I just need to find some spare time. Other than that, only sky is the limit.

Cheers!
Georgi

---

“The good thing about science is that it's true whether or not you believe in it.” ― Neil deGrasse Tyson

Add me to the list of folks that have been using MacOS machines that are now unable to do any work units for you. I have a Mac with an Apple M1 chipset, an now get "Computation Error" failures as fast as I download work units. I will have to suspend participation until I hear that you are supporting my OS.

---