Posts by JonS

1) Message boards : Number crunching : >96% of v2.08 aarch64-unknown-linux-gnu (mt) WUs failing (Message 765)
Posted 13 Mar 2018 by JonS
Post:
I have 4 Odroid C2s and they are failing most of their WUs after less than 2 seconds, with the following:

<core_client_version>7.4.23</core_client_version>
<![CDATA[
<message>
process exited with code 193 (0xc1, -63)
</message>
<stderr_txt>
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc
SIGABRT: abort called
Stack trace (14 frames):
[0x43174c]
linux-vdso.so.1(__kernel_rt_sigreturn+0x0)[0x7f92fc0510]
[0x45152c]
[0x50e2fc]
[0x4e0128]
[0x457f44]
[0x457f90]
[0x456808]
[0x457f34]
[0x4531c0]
[0x418114]
[0x404e48]
[0x5069b8]
[0x422060]

Exiting...

</stderr_txt>
]]>
2) Message boards : Number crunching : 2^64 WU peak memory usage now too high for my Odroid C2s (Message 567)
Posted 7 Jul 2017 by JonS
Post:
When my Odroid C2s started crunching 2^64 WUs they used 800MB to 900MB RAM, those that finish now use 1,500MB to 1,600MB.

However, WUs are now frequently either failing to finish (today 10 reported back to the project that there is not enough memory on the client), or are locking up the computer completely (SSH and BOINC Manager remote access attempts both time out) and they have to be power-cycled to regain access, or the WU does not run and is shown as "waiting for memory".

This behaviour started a few days ago, and has steadily increased in frequency to the point I've had to stop trying to run 2^64 WUs.

Are machines with 2GB RAM no longer suitable for this project?



©2022 Sergei Chernykh