some days ago i saw the announcement of a new a7 core (viewtopic.php?f=24&t=31960), so i just tried it out on my main machine (Win7 32 bit, Intel Pentium Dual Core E5800) and i successfully completed a p11911 WU without an issue. After that i decided to do the same on my Linux folding machine (a Lenovo 8808-94G running Linux Mint 18.3 Sylvia KDE) which is running A4/A7 WUs for years without any issue. Unfortunately the core fails to start (FahCore returned: FAILED_2 (1 = 0x1)). Any thoughts?
I've seen a report from a Mac user about a missing library. I don't know if this is the same problem, but it might be. (The Mac code is very closely related to the Linux code) Has the Linux system, itself, captured any additional information about this error?
bruce wrote: Has the Linux system, itself, captured any additional information about this error?
I'm not sure where to look for clues - i checked /var/log/kern.log and /var/log/syslog, but i can't see anything related to FAHcore or FAHClient. Are there other logfiles i should check?
FAH keeps two logs, one intended for you to read an another that gives more detailed information that might help. (There may not be any useful information there.) Both FAH logs are uploaded and deleted so you'll probably have to run another WU with your internet upload blocked. You may be able to find that log inside of the numbered subdirectory of "work"
I reported the missing library for the Mac OS version of the core. For that OS the crash log was located elsewhere in the same upper level folder that the data files for the client were stored. The crash log name began with "FAHCore_a7" with a time stamp appended. Maybe that will help you in locating where the crash report is on your linux system.
iMac 2.8 i7 12 GB smp8, Mac Pro 2.8 quad 12 GB smp6
MacBook Pro 2.9 i7 8 GB smp3
i had to wait until this machine finished a normal a7 WU before i let it download another beta WU and copied the whole /var/lib/fahclient directory while getting the errors in the log. These are the contents:
Edit 2: This new core does not link to any non-standard dynamic libraries but it does require GLIBCXX_3.4.21 which may not be supported on some older systems. Although, the current non-beta a7 core also requires GLIBCXX_3.4.21 so that may not be it.
Edit 3: A further thing you can try is just running the core manually like this:
machine ~ # ldd --version
ldd (Ubuntu GLIBC 2.23-0ubuntu9) 2.23
Copyright (C) 2016 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Written by Roland McGrath and Ulrich Drepper.
jcoffland wrote:
Edit2: This new core does not link to any non-standard dynamic libraries but it does require GLIBCXX_3.4.21 which may not be supported on some older systems. Although, the current non-beta a7 core also requires GLIBCXX_3.4.21 so that may not be it.
Ok, the problem is that the new core is requiring GLIBC_2.25 & GLIBC_2.27 which your system does not support. The previous core required GLIBC_2.17 at most. I'm looking into why it changed.
08:12:31:FS00:Unpaused
08:12:32:WU00:FS00:Connecting to 65.254.110.245:8080
08:12:32:WU00:FS00:Assigned to work server 207.218.130.23
08:12:32:WU00:FS00:Requesting new work unit for slot 00: READY cpu:2 from 207.218.130.23
08:12:32:WU00:FS00:Connecting to 207.218.130.23:8080
08:12:36:WU00:FS00:Downloading 4.53MiB
08:12:40:WU00:FS00:Download complete
08:12:40:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:11911 run:0 clone:0 gen:6 core:0xa7 unit:0x00000008cfda82175c87739bd509f989
08:12:40:WU00:FS00:Downloading core from http://cores.foldingathome.org/v7/0.0.18/lin/64bit/Core_a7.fah
08:12:40:WU00:FS00:Connecting to cores.foldingathome.org:80
08:12:41:WU00:FS00:FahCore a7: Downloading 7.85MiB
08:12:47:WU00:FS00:FahCore a7: 68.45%
08:12:49:WU00:FS00:FahCore a7: Download complete
08:12:49:WU00:FS00:Valid core signature
08:12:49:WU00:FS00:Unpacked 18.96MiB to cores/cores.foldingathome.org/v7/0.0.18/lin/64bit/Core_a7.fah/FahCore_a7
08:12:49:WU00:FS00:Starting
08:12:49:WU00:FS00:Running FahCore: /usr/bin/FAHCoreWrapper /var/lib/fahclient/cores/cores.foldingathome.org/v7/0.0.18/lin/64bit/Core_a7.fah/FahCore_a7 -dir 00 -suffix 01 -version 704 -lifeline 29276 -checkpoint 15 -np 2
08:12:49:WU00:FS00:Started FahCore on PID 29519
08:12:49:WU00:FS00:Core PID:29523
08:12:49:WU00:FS00:FahCore 0xa7 started
08:12:50:WU00:FS00:0xa7:*********************** Log Started 2019-11-05T08:12:49Z ***********************
08:12:50:WU00:FS00:0xa7:************************** Gromacs Folding@home Core ***************************
08:12:50:WU00:FS00:0xa7: Type: 0xa7
08:12:50:WU00:FS00:0xa7: Core: Gromacs
08:12:50:WU00:FS00:0xa7: Args: -dir 00 -suffix 01 -version 704 -lifeline 29519 -checkpoint 15 -np
08:12:50:WU00:FS00:0xa7: 2
08:12:50:WU00:FS00:0xa7:************************************ CBang *************************************
08:12:50:WU00:FS00:0xa7: Date: Nov 5 2019
08:12:50:WU00:FS00:0xa7: Time: 05:57:01
08:12:50:WU00:FS00:0xa7: Revision: 46c96f1aa8419571d83f3e63f9c99a0d602f6da9
08:12:50:WU00:FS00:0xa7: Branch: master
08:12:50:WU00:FS00:0xa7: Compiler: GNU 8.3.0
08:12:50:WU00:FS00:0xa7: Options: -std=c++11 -O3 -funroll-loops -fno-pie -fPIC
08:12:50:WU00:FS00:0xa7: Platform: linux2 4.19.0-5-amd64
08:12:50:WU00:FS00:0xa7: Bits: 64
08:12:50:WU00:FS00:0xa7: Mode: Release
08:12:50:WU00:FS00:0xa7:************************************ System ************************************
08:12:50:WU00:FS00:0xa7: CPU: Intel(R) Core(TM)2 CPU 6300 @ 1.86GHz
08:12:50:WU00:FS00:0xa7: CPU ID: GenuineIntel Family 6 Model 15 Stepping 2
08:12:50:WU00:FS00:0xa7: CPUs: 2
08:12:50:WU00:FS00:0xa7: Memory: 980.85MiB
08:12:50:WU00:FS00:0xa7:Free Memory: 94.33MiB
08:12:50:WU00:FS00:0xa7: Threads: POSIX_THREADS
08:12:50:WU00:FS00:0xa7: OS Version: 4.10
08:12:50:WU00:FS00:0xa7:Has Battery: false
08:12:50:WU00:FS00:0xa7: On Battery: false
08:12:50:WU00:FS00:0xa7: UTC Offset: 1
08:12:50:WU00:FS00:0xa7: PID: 29523
08:12:50:WU00:FS00:0xa7: CWD: /var/lib/fahclient/work
08:12:50:WU00:FS00:0xa7:******************************** Build - libFAH ********************************
08:12:50:WU00:FS00:0xa7: Version: 0.0.18
08:12:50:WU00:FS00:0xa7: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
08:12:50:WU00:FS00:0xa7: Copyright: 2019 foldingathome.org
08:12:50:WU00:FS00:0xa7: Homepage: https://foldingathome.org/
08:12:50:WU00:FS00:0xa7: Date: Nov 5 2019
08:12:50:WU00:FS00:0xa7: Time: 06:13:26
08:12:50:WU00:FS00:0xa7: Revision: 490c9aa2957b725af319379424d5c5cb36efb656
08:12:50:WU00:FS00:0xa7: Branch: master
08:12:50:WU00:FS00:0xa7: Compiler: GNU 8.3.0
08:12:50:WU00:FS00:0xa7: Options: -std=c++11 -O3 -funroll-loops -fno-pie
08:12:50:WU00:FS00:0xa7: Platform: linux2 4.19.0-5-amd64
08:12:50:WU00:FS00:0xa7: Bits: 64
08:12:50:WU00:FS00:0xa7: Mode: Release
08:12:50:WU00:FS00:0xa7:************************************ Build *************************************
08:12:50:WU00:FS00:0xa7: SIMD: sse2
08:12:50:WU00:FS00:0xa7:********************************************************************************
08:12:50:WU00:FS00:0xa7:Project: 11911 (Run 0, Clone 0, Gen 6)
08:12:50:WU00:FS00:0xa7:Unit: 0x00000008cfda82175c87739bd509f989
08:12:50:WU00:FS00:0xa7:Reading tar file core.xml
08:12:50:WU00:FS00:0xa7:Reading tar file frame6.tpr
08:12:50:WU00:FS00:0xa7:Digital signatures verified
08:12:50:WU00:FS00:0xa7:Calling: mdrun -s frame6.tpr -o frame6.trr -x frame6.xtc -cpt 15 -nt 2
08:12:50:WU00:FS00:0xa7:Steps: first=1500000 total=250000
08:12:53:WU00:FS00:0xa7:Completed 1 out of 250000 steps (0%)
08:12:57:FS00:Finishing
08:32:21:WU00:FS00:0xa7:Completed 2500 out of 250000 steps (1%)
08:51:49:WU00:FS00:0xa7:Completed 5000 out of 250000 steps (2%)
09:11:26:WU00:FS00:0xa7:Completed 7500 out of 250000 steps (3%)
09:30:54:WU00:FS00:0xa7:Completed 10000 out of 250000 steps (4%)
By the way, i get around 10-20 percent more PPD running p11911 WUs with the beta core than normal A7 WUs. Is the beta A7 core in fact faster or does it depend on the project 11911 benchmark parameters?
Great. Thanks for the confirmation. The PPD does depend on the benchmark. P11911 is a test project and was not benchmarked carefully. However, the new core was built with a newer compiler and so might be slightly faster.