XPost: comp.os.linux.misc   
   From: a_eder_muc@web.de   
      
   On Fr 24 Mai 2024 at 12:24, Woozy Song  wrote:   
      
   > Woozy Song wrote:   
   >> So the atd supposedly will not start another job until load factor falls   
   >> below a limit. Different documentation gives the default as 0.8 or 1.5   
   >> Now I launch a job that uses 4 cores on a 6-core CPU. If I run top   
   >> command, I see four processes running close to 100%.   
   >> Now if I submit another job 10 seconds later, that starts thereby   
   >> overloading the CPU. Documentation suggests setting load limit to more   
   >> than n-1 for n CPU cores, but I think that is intended for single-thread   
   >> jobs. I have tried altering the load limit in atd.service file to all   
   >> sorts of values, but second job keeps starting while the first is flogging   
   >> the CPU. I check with 'ps -ef|grep atd' to see it is using the desired   
   >> load limit. I am aware that the load factor is an average, you can see it   
   >> changes slowly in top/htop/glances. So I also increase the delay between   
   >> jobs to 30 seconds, but still nothing works. So it looks like I have to   
   >> specify a time like 'now+60 minutes' when I submit, requiring some guess   
   >> how long first job runs. I know I can install a proper job scheduler such   
   >> as Some Grid Engine, but that is more work.   
   >> This is on Debian 11, by the way.   
   >   
   > I found the trick: you have to add '-q B' to command, then load-limit rule   
   > applies (it behaves like batch command instead of at). Otherwise it uses   
   > default queue 'a' that only uses time without load limit.   
      
   I think ot is '-q b', the small letter b for the batch queue (a is for at).   
   The other letters are not used by default and just serve to indicate niceness.   
      
   'Andreas   
      
   --   
   ceterum censeo redmondinem esse delendam   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)