Message boards : Questions and problems : Are there any current issues with BOINC and Ubuntu 18.04 LTS release?
Message board moderation
Author | Message |
---|---|
![]() ![]() Send message Joined: 17 Nov 16 Posts: 906 ![]() |
I was wondering if there is are any issues with the new Ubuntu 18.04LTS release and BOINC. The new release made several under the hood changes from 16.04LTS primarily a new Desktop Manager with the switch from Unity to Gnome. Has that caused any issues? Also it has new GCC and GTK libraries. Has that caused any issues? Anybody currently running Ubuntu 18.04LTS want to chime in with their comments and experiences? I am running the TBar 7.4.44 BOINC release currently. TIA, Keith |
Send message Joined: 5 Oct 06 Posts: 5149 ![]() |
We are in the final stages of testing BOINC v7.10.2 as a release candidate. You are more than welcome to test it, but you would need to adjust your viewpoint and get used to running BOINC in service mode from a repository distribution - quite a lot will be different (not least, file locations, ownership, and permissions). I believe you're used to running applications under Anonymous Platform, so that change will be significant for you. On the plus side, this release contains a fix for TBar's long-standing complaint about formatting errors for the timestamps in the Event Log: #2420. Please ask him to thank Juha Sointusalo for that. |
Send message Joined: 8 Nov 10 Posts: 310 ![]() |
- quite a lot will be different (not least, file locations, ownership, and permissions). I believe you're used to running applications under Anonymous Platform, so that change will be significant for you. If we upgrade from 7.8.3/7.9.3, should we allow all projects to complete, detach, and start over? |
Send message Joined: 5 Mar 08 Posts: 272 ![]() |
Well I upgraded one cruncher from 7.8.2 to 7.10.2 without issue. This was on Debian, not Ubuntu though. The BOINC data directory is now in /var/lib/boinc but there’s a symlink so the old one still works. Running kernel 4.16 that came down a couple of days ago (was 4.15 before that). I let apt update it, didn’t detach but did finish off cached work first. Also there have been reports from one user on Einstein about mouse jerkiness - he was using Ubuntu 18 so the OS or drivers might not be that stable. MarkJ |
Send message Joined: 1 Jul 16 Posts: 147 ![]() |
I have two PCs on 18.04 and only issue I've seen is that 7.9 is in the repository. |
![]() ![]() Send message Joined: 17 Nov 16 Posts: 906 ![]() |
Well I am going to stick with BOINC 7.4.44 because of its unique abilities. I haven't had any issues with either SETI or Einstein. OTOH, I have had problems with MilkyWay. All tasks error out immediately now. I have posted in every forum available there and have only one response that was of no use. From the stderr.txt output it looks like the MilkyWay app is having issues compiling the OpenCL wisdom file. This is the output. Stderr output <core_client_version>7.4.44</core_client_version> <![CDATA[ <message> process exited with code 1 (0x1, -255) </message> <stderr_txt> <search_application> milkyway_separation 1.46 Linux x86_64 double OpenCL </search_application> Reading preferences ended prematurely BOINC GPU type suggests using OpenCL vendor 'NVIDIA Corporation' Setting process priority to 0 (13): Permission denied Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4' Switching to Parameter File 'astronomy_parameters.txt' <number_WUs> 5 </number_WUs> <number_params_per_WU> 20 </number_params_per_WU> Using AVX path Found 1 platform Platform 0 information: Name: NVIDIA CUDA Version: OpenCL 1.2 CUDA 9.2.101 Vendor: NVIDIA Corporation Extensions: cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer Profile: FULL_PROFILE Using device 2 on platform 0 Found 3 CL devices Device 'GeForce GTX 1070' (NVIDIA Corporation:0x10de) (CL_DEVICE_TYPE_GPU) Board: Driver version: 396.24 Version: OpenCL 1.2 CUDA Compute capability: 6.1 Max compute units: 15 Clock frequency: 1784 Mhz Global mem size: 8513978368 Local mem size: 49152 Max const buf size: 65536 Double extension: cl_khr_fp64 Build log: -------------------------------------------------------------------------------- <kernel>:183:72: warning: unknown attribute 'max_constant_size' ignored __constant real* _ap_consts __attribute__((max_constant_size(18 * sizeof(real)))), ^ <kernel>:185:62: warning: unknown attribute 'max_constant_size' ignored __constant SC* sc __attribute__((max_constant_size(NSTREAM * sizeof(SC)))), ^ <kernel>:186:67: warning: unknown attribute 'max_constant_size' ignored __constant real* sg_dx __attribute__((max_constant_size(256 * sizeof(real)))), ^ <kernel>:235:26: error: use of undeclared identifier 'inf' tmp = mad((real) Q_INV_SQR, z * z, tmp); /* (q_invsqr * z^2) + (x^2 + y^2) */ ^ <built-in>:35:19: note: expanded from here #define Q_INV_SQR inf ^ -------------------------------------------------------------------------------- clBuildProgram: Build failure (-11): CL_BUILD_PROGRAM_FAILURE Error building program from source (-11): CL_BUILD_PROGRAM_FAILURE Error creating integral program from source Failed to calculate likelihood Background Epsilon (15.441100) must be >= 0, <= 1 13:21:06 (2986): called boinc_finish(1) </stderr_txt> ]]> I have reset the project and it made no difference. I have set NNT and will likely remove the project from that machine if I can't find a solution. There is very little forum traffic at MilkyWay. I doubt I ever get any useful commentary from there. Both Seti an Einstein had no issues with compiling their respective OpenCL wisdom files for their respective applications. So I don't know why MilkyWay is having issues. Anyone want to offer up any suggestion or comments on what has gone wrong with the MilkyWay gpu app and Ubuntu 18.04 and the latest Nvidia ppa drivers? |
Send message Joined: 5 Mar 08 Posts: 272 ![]() |
The only comment I can make is from Debian. With the Nvidia drivers I install nvidia-kernel-dkms but to get OpenCL one has to install nvidia-opencl-icd as well. Given you are using a ppa version its going to be different (and probably more up to date drivers) which might have their own set of problems. MarkJ |
![]() ![]() Send message Joined: 17 Nov 16 Posts: 906 ![]() |
I always install the opencl-icd drivers. They get picked up automatically with the main driver when I use Synaptic Package Manager to install. |
Send message Joined: 1 Jul 16 Posts: 147 ![]() |
The default NVidia drivers from the additional hardware package was not enough for me on 18.04. I added the nvidia PPA and BOINC was able to process. I don't recall if that was OpenCL or CUDA that didn't work though. |
![]() ![]() Send message Joined: 17 Nov 16 Posts: 906 ![]() |
The only comment I can make is from Debian. With the Nvidia drivers I install nvidia-kernel-dkms but to get OpenCL one has to install nvidia-opencl-icd as well. Given you are using a ppa version its going to be different (and probably more up to date drivers) which might have their own set of problems. MilkyWay's response to me was "we know there is a problem with our code but we won't fix it. To fix your problem use a version of BOINC later than 7.6.31 which works with our buggy code." |
Copyright © 2025 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.