All Eintein@home tasks segfaults
7 Oct 2019 19:46:22 UTC
Topic 219738
(moderation:
Hello,
I wanted to get back to compute so I have BOINC set on one of my machines:
CPU type:AuthenticAMD AMD Ryzen 3 2200G with Radeon Vega Graphics [Family 23 Model 17 Stepping 0]Operating system:Linux Debian Debian GNU/Linux 10 (buster) [4.19.0-5-amd64|libc 2.28 (Debian GLIBC 2.28-10)]
However all tasks that come from Einstein@home ends in
<core_client_version>7.14.2</core_client_version> <![CDATA[ <message> process got signal 11</message> <stderr_txt> putenv 'LAL_DEBUG_LEVEL=3' SIGSEGV: segmentation violation
It was happening for a while now, so I decided to finally try to solve it. I'm also computing rosetta@home and there tasks are executing fine, so it seems as something Einstein specific. Any ideas what could be causing things, or where to start with debugging?
You don't have use LIBC215
)
You don't have use LIBC215 checked in computing preferences do you?
Ludo wrote:Any ideas what
)
As Keith mentions, it's likely something to do with what setting you have for using (or not) the LIBC215 version of the app. Whatever setting you currently have, try changing it to the opposite setting.
For Debian buster, I believe you need the option to use the LIBC215 version selected.
Your computers are hidden so I can't see which particular searches you are trying to run. If it's either O2AS20-500 (which is finishing up) or O2MD1 (which is just starting) then both of those have a LIBC215 version of the app.
Cheers,
Gary.
Thanks for response, that
)
Thanks for response, that would make sense why it segfaults if it tries to access some function which is not present anymore. I turned on the libc215 setting, now need to wait for some task to arrive to check if it helps.
EDIT: so I confirm the setting use of libc2.15 in project settings helps, if anybody walks around this - https://einsteinathome.org/account/prefs/project