Message boards :
ATLAS application :
Is this task dead ?
Message board moderation
Author | Message |
---|---|
Send message Joined: 12 Jul 11 Posts: 79 Credit: 1,067,251 RAC: 0 ![]() ![]() |
Hi I have this task which ran for more than a day on my iMac but now it's not using any CPU, it's stuck at 96,947%, then I try to open the "show graphic" from the BM I'll get a never ending "loading" on the http://localhost:65028/ link it provided, and the console shows this : ![]() It seems it is dead, isn't it ? |
Send message Joined: 12 Jul 11 Posts: 79 Credit: 1,067,251 RAC: 0 ![]() ![]() |
No change in the situation, it is obviously more than dead :( I had to finish it off... 1,5 day x 6 core lost :( |
![]() Send message Joined: 15 Jun 08 Posts: 2141 Credit: 175,544,069 RAC: 104,209 ![]() ![]() ![]() |
... It seems it is dead, isn't it ? Most likely yes. You may try to shutdown the VM gracefully. I'm not sure if it works for Mac, but on linux I would do the following: 1. open a console 2. cd to the BOINC/project/lhcathome.cern.ch_lhcathome folder 3. check the ATLAS*.xml file for a line like "<completion_trigger_file>xyz</completion_trigger_file>" 4. cd to the BOINC/slots/x subdir that contains the running VM 5. from there cd to the subfolder "shared" 6. run "touch xyz" If this doesn't work either wait until the watchdog shuts it down or cancel the VM via you BOINC manager. |
Send message Joined: 14 Jan 10 Posts: 1155 Credit: 7,103,928 RAC: 1,140 ![]() ![]() ![]() |
This is a typical file system error on your virtual hard drive. Since we cannot login into the VM, we can't repair the file system. Maybe you could save your credits by placing an empty file called ''atlas_done'' in the shared folder of the corresponding slot. |
Send message Joined: 12 Jul 11 Posts: 79 Credit: 1,067,251 RAC: 0 ![]() ![]() |
Sorry I didn't see your answers and I had cancelled it already... thanks anyway. Especially the trick regarding the credit, I'll try to note it down for the future ! :) What is the "watchdog" ? do you mean this task wouldn't have stayed forever ? |
©2023 CERN