Posts by Agentb

21) Message boards : Questions and problems : Location of Event Log (file)? (Message 78548) Posted 3 Jun 2017 by Agentb Post: Perhaps you might get better information about tasks if you look at a) job_log_<project_name>.txt file in the boinc data directory, which records all completed tasks. b) or use the various boinccmd command options such as --get_tasks # show tasks --get_old_tasks # show reported tasks from last 24 hours Event logs are quite dynamic, easily overwritten and liable to be tuned by user settings.
22) Message boards : The Lounge : The Seti is Slumbering Cafe (Message 78528) Posted 2 Jun 2017 by Agentb Post: anniet wrote: start looking for another tree to bark up @anniet there was a similar problem-gtx1070-upgrade-570 and whilst the OP never published a solution it looks like a driver upgrade / change may have helped (no longer getting errors) - at least worth a try or asking the OP.
23) Message boards : The Lounge : The Seti is Slumbering Cafe (Message 78339) Posted 28 May 2017 by Agentb Post: Yep, Einstein is down. :-( E@H is now up.
24) Message boards : GPUs : which PPA for Ubuntu Mesa openCL for Raden 7750??? (Message 78239) Posted 26 May 2017 by Agentb Post: I did a ppa-purge on the mesa opencl, and re-reviewed the AMD documentation. Both good ideas, running two drivers on the one device often ends badly. BOINC recognized the GPU with OpenCL. But the Einstein WU failed. I'll pursue this with the Einstein Forum - since building the AMD code for the release seems (to me) to be the least weird of the builds and the most predictable in how it could go wrong. (Trying not to laugh.) I'd say this probably the only way it will go right unless you are handy at c++ and feel happy to help with debugging and writing the missing 1.2 bits at mesa. Here is what the WU task log said of the failure.... <snip up to the first error> Using OpenCL platform provided by: Advanced Micro Devices, Inc. Using OpenCL device "Capeverde" by: Advanced Micro Devices, Inc. Max allocation limit: 1350254592 Global mem size: 2040578048 Couldn't create OpenCL command queue (error: -6)! OpenCL shutdown complete! Google for OpenCL error codes -6 CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host. The apps at E@H take at least .9GBB VRAM - clinfo may be revealing and it is reported there are some environment variables that can be set to eek a little extra, but note from the AMDGPUPRO driver page the risk of pushing OpenCL past the maximum. Known Issues System hangs when OpenCL™ attempts to allocate more than available system memory. I see some valid tasks at E@H so seems good.
25) Message boards : Questions and problems : boinctui on Ubuntu 16.04 gives "unauthorized!" and won't run (Message 78096) Posted 21 May 2017 by Agentb Post: How exactly did you install boinc and boinctui (only once from repo?). You don't say if the GUI manager is running properly. i would suggest restarting boinc and posting the output from the log files if you have set a a password boincctui needs it set in File / Config Host 127.0.0.1 Port 31416 pwd <password value>
26) Message boards : Questions and problems : BOINC and amdgpu-pro (Message 77530) Posted 22 Apr 2017 by Agentb Post: OK, got it figured out. The problem was that my /etc/ld.so.cache was out of date and also my LD_LIBRARY_PATH was set wrong. Both errors had the same effect: opencl code would pick up the wrong version of libamdocl64.so. The proprietary driver puts its libraries in /opt/amdgpu-pro/lib64 while the open source amdgpu driver has its libraries in /usr/lib64. When the /usr/lib64 version is picked up, opencl code doesn't work. Got that all sorted and the GPU is happily crunching away now... OK that's good, was it as simple as re-running ldconfig? The ubuntu install drops a couple of conf files amdgpu-pro-i386.conf and amdgpu-pro-x86_64.conf in the /etc/ld.so.conf.d/ directory and does not use the LD_LIBRARY_PATH environment vars. These conf files are picked up with a wildcard include by ldconfig. I guess that is a missing / broken step in the install process to pickup the libraries. PS - amdgpu-pro is officially still not supported under openSUSE, but you can download the SLES version and then apply the patch outlined here: https://community.amd.com/thread/212558 You may wish to mention there the ldconfig problem, as it might affect libGLand other libraries.
27) Message boards : Questions and problems : BOINC and amdgpu-pro (Message 77527) Posted 21 Apr 2017 by Agentb Post: Is amdgpu-pro supported, and if so, how do I get it to work with BOINC? Proabably, although my SUSE knowledge is very limited, and AMD-GPUPRO is fairly new to SUSE You will need to add boinc to the video group, if not done already. usermod -A video boinc and restart boinc Good luck.
28) Message boards : Questions and problems : Boinc Linux Problems (Message 77281) Posted 10 Apr 2017 by Agentb Post: Hello guys and girls, this is my first post - sorry if this had been already posted but didn't find this because I guess it's very specific. I have some problems with Boinc on Linux (Mint) 18.1. Boinc installed due to repository on Linux Mint (sudo apt-get install boinc) Version: 7.6.31 1. When PC is booting the Boinc Manager should start up automatically and connect to the Core client automatically because it's set into the "start-up-list" on Linux . Even when I set the delay to a few seconds the Manager always says "can not connect" or "disconnected". I have to manually start Boinc Manager. 2. I set the "use CPU time" to 50 % on my Notebook. But it uses 100 % all the time. Only when I close and restart Boinc it's crunching by presetting. 3. Sometimes when I close the Manager I can not reopen it by clicking on the icon. I have to start the Manager via shell/terminal by typing in "Boinc". Any solutions? Thank you! Hi ShootY welcome to boinc forums. You'll need to be a little more specific if possible. Confirm you only installed once from repos (and no other source) The boinc manager (GUI) is started as boincmgr The boinc client - which schedules tasks is normally started as a service at boot time, and should be named boinc If you need to start the GUI from a command line use boincmgr not Boinc, i don't know why that restarts the manager. How exactly are you restarting boinc (2.) (not the GUI but boinc client) ? There may be some error messages in a) boinc event log b) in the ~/.BOINC/ directory I might suggest you post them as described in the sticky threads. There are a few of ways to autostart applications, you will need to be very specific about what you are try to autostart and when, I would leave that until you solve the other issues. As a guess you need to add the user account you are using to login - to the group that is created during install called boinc, this solves a couple of issues for some distros. Good luck.
29) Message boards : GPUs : Fedora 25 GPUs sometimes there but reported missing (Message 76451) Posted 14 Mar 2017 by Agentb Post: But look at the second log, after the BOINC restart - BOINC detects the GPU, and even starts running an Einstein task. OP asked why no CUDA tasks. I don't disagree there is probably a timing issue here, however the first log is likely to have been done by systemd, the second later in x-windows client session with sudo privs. To "detect the GPU" the boinc client forks a copy of itself and makes makes several dlopen calls. OpenCL and CUDA libraries are different and are often installed separately (depends on the tribe). Older versions of boinc client make old assumptions where these libraries are found, and so these calls fail. hth
30) Message boards : GPUs : Fedora 25 GPUs sometimes there but reported missing (Message 76448) Posted 14 Mar 2017 by Agentb Post: I'm not well versed in Fedora but searching for Fedora in the advanced search reveals GPU CUDA issues on Fedora 23 with NVIDIA GTX 970 If the same issue, then it looks like you have not installed the cuda libraries, or boinc client cannot find them, but you are crunching using the OpenCL libraries. There are likely to be several other issues (which explains the no detection at start) these may be addressed with permissions settings and/or adding a delay in the start script. Upgrading boinc to something more recent (7.6.31 or later) which has better GPU detection, may also help.
31) Message boards : GPUs : boinc can't find (asus) AMD R7 250 on Ubuntu 14.04.05 (Message 76397) Posted 13 Mar 2017 by Agentb Post: Fri 10 Mar 2017 10:43:45 AM EET \| \| Starting BOINC client version 7.2.42 for x86_64-pc-linux-gnu Any help would be appreciated ! Thx. Apologies for the delay - This is quite an old version (7.2.x), and a lot of improvements have been made recently in GPU detection. This forum has many posts about this. The repos for ubuntu should be much more up to date 7.6.31 iirc, and boinc should in almost all cases be installed via the repo. sudo apt-get install boinc not from the unmaintained download on the boinc web site.
32) Message boards : GPUs : [Linux] Trick to use AMDGPU-Pro OpenCL capabilities without installing the whole propertary driver stack (Message 75415) Posted 20 Jan 2017 by Agentb Post: Ellesmere is the codename for the RX-470 an RX-480, Polaris10 is a psudonym for Ellesmere - see https://en.wikipedia.org/wiki/AMD_Radeon_400_series So this host has only one graphics card or two?
33) Message boards : Questions and problems : Can I reduce I/O using memory? (Message 75347) Posted 17 Jan 2017 by Agentb Post: somitcw wrote: With BOINC pounding on hard-drives 24/7, it is nice to know that you declare "I don't think running BOINC will kill any HDD". That is what you said. And I would concur with Juha. I have a PC pounding away 7x24 BIONC for 5 years 8 months and 19 days daily backups etc. HDD SMART statistics reveal it is about 70% of it's reallocated sector allocation which suggests I have another two years left before I need to worry about anything. Another PC has a small SSD pounding silently away for nearly two years at BOINC , while running hot double GPU system, it has a has wear level indicator at 50%. If the press is to believed, this is just an indicator and the life is typically another 30-100+% longer - I'll leave it to the reader to search. If you are really worried about this, monitor the SMART drive stats monthly, and then make comment.
34) Message boards : GPUs : clinfo.exe wrong values for CL_DEVICE_ADDRESS_BITS (Message 75178) Posted 10 Jan 2017 by Agentb Post: That zip file dates back to 30-Mar-2011 13:52, so i would suggest you would be better off deleting it.
35) Message boards : GPUs : clinfo.exe wrong values for CL_DEVICE_ADDRESS_BITS (Message 75161) Posted 9 Jan 2017 by Agentb Post: Welcome to the BOINC forums Most of these questions need to be raised with nVidia. nVidia's support for OpenCL is at best average. It looks like you are running an old version see http://www.nvidia.co.uk/Download/index.aspx?lang=en-uk
36) Message boards : Questions and problems : Project Doesn't Recognize Computer (Message 74950) Posted 27 Dec 2016 by Agentb Post: Fezik wrote: I have been trying to connect to the SZTAKI desktop grid but my account doesn't seem to recognize the computer regardless of what commands I put into it. There are no tasks and no project files downloaded and I am not sure why this is. Please help - I was hoping to set this up tonight, but that doesn't look possible right now. Welcome to boinc forum This project does not have a lot of work - currently it shows no work available see http://szdg.lpds.sztaki.hu/szdg/server_status.php the event log will show something like Tue 27 Dec 2016 13:57:19 GMT \| SZTAKI Desktop Grid \| (Project has no jobs available)
37) Message boards : News : Thanks to Kathryn for 10 years of moderation (Message 74904) Posted 22 Dec 2016 by Agentb Post: Thank you Kathryn. There's nothing wrong with sobriety in moderation. - John Ciardi I'll get my coat.
38) Message boards : The Lounge : BOINC in the News (Message 74822) Posted 18 Dec 2016 by Agentb Post: Astronomy Magazine - Amateur astronomers discover a binary pulsar system on crowdsourced technology
39) Message boards : The Lounge : The Seti is Slumbering Cafe (Message 74775) Posted 14 Dec 2016 by Agentb Post: Einstein Forums are now up, in fact i didn't actually notice them down.
40) Message boards : Questions and problems : Avira Antivirus Detecting BOINC as a Virus (Message 74513) Posted 30 Nov 2016 by Agentb Post: Yeah, I figured it was probably a false positive. I was not aware of virustotal.com so thanks for the tip! AV list for submitting samples and FP has a link for Avira and other AV vendor submissions. Even ClamAV is on the list...

Previous 20 · Next 20

Copyright © 2024 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.