LHC@home

There are a few comments recently about whether VirtualBox 5.2 works or not with ATLAS. According to Laurence the BOINC community suggests 5.2.16 is a good version. Please let us know if you have issues with this version.

Personally the 5.2.16 gives many unmanageable errors with ATLAS. The CMS, LHCb, Theory have no issues.

The 5.1.38 works well for all tasks.

I don't think others have computers that can run so many WU at once?

Personally the 5.2.16 gives many unmanageable errors with ATLAS.

I have started to upgrade machines (Win10 1803) to VB 5.2.16 and BOINC 7.12.1 and until now, this works fine. So far I haven't seen any unmanageable errors, Maybe it has something to do with my clients that are never allowed to use more than 75% of CPU-Power

Did you upgrade BOINC too ?

---

Supporting BOINC, a great concept !

Machine is running Windows 10 - 1803, VB 5.2.16 and BOINC 7.12.1.

Tried running a 3-core ATLAS, 2-core Theory, and a 1-core LHCb tasks at the same time.

The Stderr.txt for the ATLAS task had the following error message:

2018-08-09 12:57:11 (1988): Error creating VirtualBox instance! rc = 0x80004002

While the tasks were processing, the machine stopped running. When I restarted it, the Event Log had this message:

8/9/2018 3:41:13 PM | LHC@home | [error] no project URL in task state file

Atlas task finished and validated, Theory and LHCb tasks are still running.

Let me know if you need more information.

Have upgraded Virtualbox from 5.2.8 to 5.2.16. Windows 10 pro (10.00.16299.00).
Atlas-Task with Boinc 7.12.1 get Status unmanagable!
After changing the priority from vboxsvc.exe from "normal" to "lower than normal" it is running.

Today in the morning I upgraded a Win7 Client in the middle of two Atlas-WUs to BOINC 7.12.1 and VirtualBox 5.2.16.

The WUs have finished normal, have their HITs-File and got validated.

Until now I have crunched 100 WUs with 5.2.16 and there was no problem so far

---

Supporting BOINC, a great concept !

Yes, I did both, I allow 100% of CPU usage so it's likely to do with that?

Yes, I did both, I allow 100% of CPU usage so it's likely to do with that?

Maybe, so far I had not any unmanagable VM.

And I had to downgrade several installations of BOINC to 7.8.3, not regarding ATLAS, but because of problems with the Grafic-Card not running in additional Instances of BOINC, but VB 5.2.16 is still running fine

---

Supporting BOINC, a great concept !

If you're looking for the latest VirtualBox 5.2 packages, see VirtualBox 5.2 builds. Please also use version 5.2 if you still need support for 32-bit hosts, as this has been discontinued in 6.0. Version 5.2 will remain supported until July 2020

This is the only Computer from five with one or two tasks failed per day:
https://lhcathome.cern.ch/lhcathome/results.php?hostid=10593998
Virtualbox 5.2.32

2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.preExecute 2019-08-08 01:20:29,398 INFO Now writing wrapper for substep executor EVNTtoHITS
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe._writeAthenaWrapper 2019-08-08 01:20:29,398 INFO Valgrind not engaged
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.preExecute 2019-08-08 01:20:29,398 INFO Athena will be executed in a subshell via ['./runwrapper.EVNTtoHITS.sh']
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.execute 2019-08-08 01:20:29,399 INFO Starting execution of EVNTtoHITS (['./runwrapper.EVNTtoHITS.sh'])
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.execute 2019-08-08 01:24:51,820 INFO EVNTtoHITS executor returns 1
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.validate 2019-08-08 01:24:52,925 ERROR Validation of return code failed: Non-zero return code from EVNTtoHITS (1) (Error code 65)
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfExe.validate 2019-08-08 01:24:52,951 INFO Scanning logfile log.EVNTtoHITS for errors
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.trfValidation.scanLogFile 2019-08-08 01:24:53,099 WARNING Detected CoreDumpSvc report - activating core dump svc grabber
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.transform.execute 2019-08-08 01:24:53,101 CRITICAL Transform executor raised TransformValidationException: Non-zero return code from EVNTtoHITS (1); Logfile error in log.EVNTtoHITS: "Segmentation fault: Event counter: 0; Run: unknown; Evt: unknown; Current algorithm: ISF_Kernel_FullG4; Current Function: unknown"
2019-08-08 01:27:14 (17604): Guest Log: PyJobTransforms.transform.execute 2019-08-08 01:24:56,320 WARNING Transform now exiting early with exit code 65 (Non-zero return code from EVNTtoHITS (1); Logfile error in log.EVNTtoHITS: "Segmentation fault: Event counter: 0; Run: unknown; Evt: unknown; Current algorithm: ISF_Kernel_FullG4; Current Function: unknown")

Saved Atlas-task for a BIOS-update and saw this lines in log:
2020-02-17 15:07:54 (42100): Stopping VM.
2020-02-17 15:08:14 (42100): Error in stop VM for VM: -2147024891
Command:
VBoxManage -q controlvm "boinc_719cf43ad3e99c5c" savestate
Output:
0%...10%...20%...30%...40%...50%...60%...70%...80%...90%...
Progress state: E_ACCESSDENIED
VBoxManage.exe: error: Failed to save machine state
VBoxManage.exe: error: Assertion failed: [SUCCEEDED(rc)] at 'F:\tinderbox\win-5.2\src\VBox\Main\src-server\MachineImpl.cpp' (10670) in long __cdecl Machine::i_saveStorageControllers(struct settings::Storage &).
VBoxManage.exe: error: COM RC=E_ACCESSDENIED (0x80070005).
VBoxManage.exe: error: Please contact the product vendor!
VBoxManage.exe: error: Details: code E_FAIL (0x80004005), component SessionMachine, interface IMachine
VBoxManage.exe: error: Context: "enum RTEXITCODE __cdecl handleControlVM(struct HandlerArg *)" at line 623 of file VBoxManageControlVM.cpp

2020-02-17 15:08:14 (42100): VM did not stop when requested.
2020-02-17 15:08:14 (42100): VM was successfully terminated.
2020-02-17 15:18:10 (12332): Detected: vboxwrapper 26197

After BIOS-update task finished successful.
https://lhcathome.cern.ch/lhcathome/result.php?resultid=263875662

May be off thread, but I'll try anyway! Virtual box kills some of my own software (intercepting TCP internet stream with Wireshark/Dumpcap). I want to either kill all the annoying messages telling me what I already know (VB not installed), OR configure VB so my own code will run. I would prefer the latter so's I can do more LHC stuff. 3rd choice is to abandon LHC. Any suggestions? regards, Bob.