Posts by mmonnin

41) Message boards : News : Theory application reaches 4 TRILLION events today !! (Message 34651) Posted 14 Mar 2018 by mmonnin Post: My Einstein@home RAC is around 25000, while all other project RAC rarely goes beyond 2000. This because Einstein@home overpays GPU tasks. Tullio GPUs do more work.
42) Message boards : ATLAS application : ATLAS native app (Message 34650) Posted 14 Mar 2018 by mmonnin Post: How do I NOT receive the native app? Run test applications is unchecked yet I received native apps on computers with separate locations. Going to have to unselect Atlas all together.
43) Message boards : ATLAS application : Unable to upload an Atlas task (Message 34587) Posted 12 Mar 2018 by mmonnin Post: Another locked: 316316 LHC@home 3/12/2018 6:41:14 AM [error] Error reported by file upload server: [E8dMDmPJSFsnDDn7oo6G73TpABFKDmABFKDmquKKDmABFKDmkyA17m_0_r860623117_ATLAS_result] locked by file_upload_handler PID=-1
44) Questions and Answers : Unix/Linux : arch linux virtual box (Message 34581) Posted 11 Mar 2018 by mmonnin Post: +1 to all your comments, except: mmonnin wrote: If you do wish to run Atlas, they run most efficient at 3-5 threads ... The VMs run through a startup phase and a shutdown phase where only 1 core is used and all other cores remain idle. Those total idle times become more relevant for WUs with short calculation phases together with a slow internet connection. WUs with long calculation phases together with a fast internet connection will be less affected. Thus, a singlecore setup will be most efficient. Of cource, there may be other reasons to run multicore VMs. e.g. a single 5core would need much less total RAM than 5 singlecores. I was referencing David Cameron's post here: https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4479&postid=32802#32802 A single thread is 5th most efficient.
45) Questions and Answers : Unix/Linux : arch linux virtual box (Message 34576) Posted 10 Mar 2018 by mmonnin Post: Your logs show that you ran only Sixtrack and ATLAS (native). Both don't need VirtualBox. I'm sure those atlas tasks only showed up after I installed virtualbox. But then again, the logs said that virtualbox wasn't installed... Maybe I'll try again tomorrow, maybe not. Native atlas doesn't require Vbox. Instead of including everything in a .vdi the native app puts those requirements on the user if they wish to not use vbox. Please read this thread for some info on how to get it all setup. I haven't done so myself as even vbox is easier. https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4396 After vbox is installed, the BOINC client must be restarted. VBox detection is done at client startup. Native apps can be disabled by unselecting test apps. Your sixtrack tasks are validating, no need to abort those. If you do wish to run Atlas, they run most efficient at 3-5 threads with an app_config file.
46) Message boards : ATLAS application : Unable to upload an Atlas task (Message 34554) Posted 6 Mar 2018 by mmonnin Post: 1 for me as well: 3570k 6313 LHC@home 3/6/2018 5:53:12 PM Temporarily failed upload of j70KDm0NUCsnDDn7oo6G73TpABFKDmABFKDmSwKKDmABFKDmohi1Vn_2_r1676589227_ATLAS_result: transient upload error
47) Message boards : ATLAS application : ATLAS native app (Message 34519) Posted 1 Mar 2018 by mmonnin Post: I made a guide on my home forum on how to do it in Window and Linux. Easier then setting up the native client I believe: http://www.overclock.net/forum/365-overclock-net-boinc-team/1628924-guide-setting-up-multiple-boinc-instances.html There are times when I really need it, as when trying to run CPU and GPU projects with different buffer settings. I will use the Linux version, and have to create “gui_rpc_auth.cfg” and place it in the /etc/boinc-client folder (for Ubuntu) in order to connect with BoincTasks over the LAN. I notice that you refer to a "/var/lib/rpc_gui_auth.cfg" file. Is that the correct name? I don't know if they are really the same file or not. Thanks. Nope, you're right. I just swapped it around. Thanks for catching it. I've updated the guide. The code lines look like crap now after the forum update to vB. :( I also use it for running GPUs on another client to make it easier to manage. Also helpful keeping the CPU busy when mt or vbox tasks don't use a whole CPU thread.
48) Message boards : ATLAS application : ATLAS native app (Message 34516) Posted 1 Mar 2018 by mmonnin Post: If you run a separate BOINC client instance for ATLAS (native)... Do you have a good (i.e., simple) way of setting up two BOINC instances? Is it stable? Can you post it? I made a guide on my home forum on how to do it in Window and Linux. Easier then setting up the native client I believe: http://www.overclock.net/forum/365-overclock-net-boinc-team/1628924-guide-setting-up-multiple-boinc-instances.html
49) Questions and Answers : Unix/Linux : ATLAS native Linux app (Message 34432) Posted 21 Feb 2018 by mmonnin Post: It'd be great if 100% of the setup was verified to work and all in one post. Including the singularity and stuff on other sites.
50) Message boards : CMS Application : Jobs draining for a WMAgent upgrade (Message 34400) Posted 19 Feb 2018 by mmonnin Post: Why are tasks still being sent out if there is no work to do? Nothing but 207 (0x000000CF) EXIT_NO_SUB_TASKS for 8 hours.
51) Message boards : Sixtrack Application : Transfer issues (Message 33540) Posted 28 Dec 2017 by mmonnin Post: Still have one locked :( Others have completed and sent. LHC@home 12/27/2017 10:07:38 PM [error] Error reported by file upload server: [LHC_2015_LHC_2015_260_BOINC_errors__19__s__62.31_60.32__5.6_5.7__5__39_1_sixvf_boinc65870_0_r1527001410_0] locked by file_upload_handler PID=-1
52) Message boards : ATLAS application : Uploads of finished tasks not possible since last night (Message 33398) Posted 15 Dec 2017 by mmonnin Post: 47290 LHC@home 12/15/2017 5:42:38 PM Temporarily failed upload of 63gNDml8GirnDDn7oo6G73TpABFKDmABFKDmrUNKDmABFKDm1XIRLo_1_r1909933188_ATLAS_result: transient HTTP error This one is still stuck for me. Others have uploaded since.
53) Message boards : ATLAS application : Uploads of finished tasks not possible since last night (Message 33306) Posted 13 Dec 2017 by mmonnin Post: Thanks for the update. I have one stuck as well. Theory and LHCb are uploading on the same PC for me.
54) Message boards : ATLAS application : Performance 4-core VM versus 3-core VM (Message 33214) Posted 6 Dec 2017 by mmonnin Post: On my 1950x I run 4 threads with 2 tasks only since 9.8gb of memory is used. I noticed it took awhile for the CPU time to reach run time and show as 100% in BoincTasks so I started running some extra tasks on another client to keep the CPU busy. While the CPU is pushed with tasks > threads Atlas runs at about 2.5 CPUs utilized. There was a period where I ran out of other work and a couple of tasks were around 3.4 - 3.5 CPUs utilized out of 4. I could run fewer extra tasks but I'm ok with it.
55) Message boards : Number crunching : Tasks Fail on Linux Host (Message 31016) Posted 24 Jun 2017 by mmonnin Post: Was BOINC restarted after the VBox installation. I think it's detected at BOINC startup like hardware. There should be a line in the event log saying it found VBox.

Previous 20

LHC@home