Message boards :
ATLAS application :
No tasks are available for ATLAS Simulation
Message board moderation
| Author | Message |
|---|---|
|
Send message Joined: 27 Sep 08 Posts: 844 Credit: 690,647,962 RAC: 105,073    |
I'm seeing the above message but there is plenty of tasks? |
 Nils HøimyrSend message Joined: 15 Jul 05 Posts: 248 Credit: 5,974,599 RAC: 0  |
Your BOINC client should get one soon when updating. My client issued the same message at 5:58, but got one ATLAS task at 6am. We still have scope for improvement on the scheduler setup with respect to highly variable task quantities and duration for the different applications. Currently we rely on fairly standard BOINC server parameters and distribution of tasks between applications. |
|
Send message Joined: 2 May 07 Posts: 2240 Credit: 173,894,884 RAC: 3,757 
|
The deadline of the Atlas-Tasks show in boinc and in the LHC-Website of LHC a difference of ONE DAY in the momemt. Only for the Tasks in progress and not finished. The Time is ok. |
Yeti Send message Joined: 2 Sep 04 Posts: 455 Credit: 201,072,819 RAC: 36,013 |
Your BOINC client should get one soon when updating. My client issued the same message at 5:58, but got one ATLAS task at 6am. Nope, my clients are asking for work since hours but they don't get work. Could you please give the scheduler a kick ...  Supporting BOINC, a great concept ! |
|
Nils Høimyr Send message Joined: 15 Jul 05 Posts: 248 Credit: 5,974,599 RAC: 0 |
Scheduler and daemons restarted. We have reduced the RPC delay, let's see if this has a positive effect, or leads to an overload of the scheduler. |
|
Yeti Send message Joined: 2 Sep 04 Posts: 455 Credit: 201,072,819 RAC: 36,013 |
Scheduler and daemons restarted. We have reduced the RPC delay, let's see if this has a positive effect, or leads to an overload of the scheduler. Short after this I got several Atlas-Tasks, thx. We will see how it looks in several hours Supporting BOINC, a great concept ! |
 Send message Joined: 30 Aug 14 Posts: 145 Credit: 10,847,070 RAC: 0
|
I'm still getting Atlas tasks very irregular. Boinc often says "no Atlas tasks available". Therefore my machine runs out of work very often and just sits ideling :-( Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us |
|
Send message Joined: 2 May 07 Posts: 2240 Credit: 173,894,884 RAC: 3,757
|
The deadline of the Atlas-Tasks show in boinc and in the LHC-Website of LHC a difference of ONE DAY in the momemt. Now the deadline is one week shorter, but also is ONE DAY difference. |
|
Send message Joined: 27 Sep 08 Posts: 844 Credit: 690,647,962 RAC: 105,073 |
I wanted to switch one of my computers to just run ATLAS, so I created a preference with just ATLAS. I am still seeing No tasks Any sugestions on how to make it work? I would have though it's no reliant on the scheduling if I force just ATLAS tasks? |
|
Yeti Send message Joined: 2 Sep 04 Posts: 455 Credit: 201,072,819 RAC: 36,013 |
Toby, can you provide more information about the machine ? Which of your computers is it ? What are messages in the Event-Log regarding Atlas ? Are you using any kind of ...config.xml ? Supporting BOINC, a great concept ! |
|
Send message Joined: 27 Sep 08 Posts: 844 Credit: 690,647,962 RAC: 105,073 |
Hi Yeti, thanks. I tried on the Boron computer a while ago and the Desktop one more reciently. I created a preference School to split out that computer from the rest using settings, Use CPU, run test, Only Atlas apps, accept work from other if none, no limt on jobs and max CPU, 3. my app_config is:
The most common lines in the logs are:
Maybe with the work buffer of 0.5 day then it's 1 out 1 in for ATLAS, vs the other project will buffer work, which is sort of strange as typically ATLAS task are shorter than Theory for example. e.g. on another machine with all sub projects I see buffer of 18 Theory, 7 LHCb, 2 CMS and 2 ATLAS. I don't have any error for ATLAS, I just wanted to get my RAC for ATLAS up a bit higher. |
|
Yeti Send message Joined: 2 Sep 04 Posts: 455 Credit: 201,072,819 RAC: 36,013 |
Toby, first please work through my checklist V3 Number 4 and 5 If this doesn't help, try my app_config.xml:
<app_config>
<project_max_concurrent>8</project_max_concurrent>
<app>
<name>ATLAS</name>
<fraction_done_exact/>
<max_concurrent>2</max_concurrent>
</app>
<app>
<name>Theory</name>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>ATLAS</app_name>
<version_num>100</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>4.000000</avg_ncpus>
<max_ncpus>4.000000</max_ncpus>
<plan_class>vbox64_mt_mcore_atlas</plan_class>
<api_version>7.7.0</api_version>
<cmdline>--memory_size_mb 7500</cmdline>
<dont_throttle/>
<is_wrapper/>
<needs_network/>
</app_version>
</app_config>
For now, I have to go to bed, I will come back tomorrow to see, if it helped Supporting BOINC, a great concept ! |
|
Send message Joined: 27 Sep 08 Posts: 844 Credit: 690,647,962 RAC: 105,073 |
I think I worked it out, for the other subprojects no limit = no limit, where as for Atlas no limt = 1. I set it to 24 an it just grabbed 24 tasks straight away. There is a complex inter play of web and app config settings 🙄 |
|
Send message Joined: 2 May 07 Posts: 2240 Credit: 173,894,884 RAC: 3,757
|
I created a preference School to split out that computer from the rest using settings, Use CPU, run test, Only Atlas apps, accept work from other if none, no limt on jobs and max CPU, 3. You have it now running, but you don't need the Test flag. |
|
Send message Joined: 27 Sep 08 Posts: 844 Credit: 690,647,962 RAC: 105,073 |
I'll toggle the test flag and see what happens, I assume it won't make any difference? Yeh didn't effect the outcome. Req for ATLAS team to understand no limits correctly |
|
Send message Joined: 2 May 07 Posts: 2240 Credit: 173,894,884 RAC: 3,757
|
The deadline of the Atlas-Tasks show in boinc and in the LHC-Website of LHC a difference of ONE DAY in the momemt. Sixtrack is now ok, but Atlas and Theory have this difference of one day still. |
|
Send message Joined: 11 Aug 11 Posts: 6 Credit: 16,714,519 RAC: 0
|
Server Status shows ATLAS Simulation 1016 3712 available/in process but for the last hour all I get is 9/1/2017 5:19:40 PM | LHC@home | Sending scheduler request: To fetch work. 9/1/2017 5:19:40 PM | LHC@home | Requesting new tasks for CPU and AMD/ATI GPU 9/1/2017 5:19:42 PM | LHC@home | Scheduler request completed: got 0 new tasks 9/1/2017 5:19:42 PM | LHC@home | No tasks sent 9/1/2017 5:19:42 PM | LHC@home | No tasks are available for ATLAS Simulation Anyone Else?? |
|
Send message Joined: 20 Feb 16 Posts: 3 Credit: 46,306 RAC: 0 
|
yes I have the same problem, the server has a lot of atlas tasks but I get none... so bad |
|
Send message Joined: 20 Feb 16 Posts: 3 Credit: 46,306 RAC: 0
|
04.09.2017 12:09:51 | LHC@home | update requested by user 04.09.2017 12:09:51 | LHC@home | Sending scheduler request: Requested by user. 04.09.2017 12:09:51 | LHC@home | Requesting new tasks for CPU 04.09.2017 12:09:53 | LHC@home | Scheduler request completed: got 0 new tasks 04.09.2017 12:09:53 | LHC@home | No tasks sent 04.09.2017 12:09:53 | LHC@home | No tasks are available for ATLAS Simulation Server shows full with work, can LHC Team fix it or gives any news of this? |
 Send message Joined: 15 Jun 08 Posts: 2528 Credit: 253,722,187 RAC: 73,124  |
04.09.2017 12:09:51 | LHC@home | update requested by user Same here. |
©2024 CERN