List:General Discussion« Previous MessageNext Message »
From:Dan Nelson Date:December 30 2002 9:48pm
Subject:Re: Replication priority / speed
View as plain text  
In the last episode (Dec 30), Matt Sturtz said:
> Hello, Jeremy, et al--  Thanks for the reply before...  Further questions:
> >> Is it possible to set either set the priority ('nice') of the
> >> Slave thread down so it doesn't do that?
> >
> > The slave thread only?  No, not really.  You could nice MySQL when
> > you start it up.  But I'm not sure how much effect (positive or
> > negative) that'd have.
> When I run "show [full] processlist", there's an "Id" column, but it
> doesn't corrospond with the Unix PID of the process (on OS's that use
> a seperate PID for each thread-- like Linux does)...  Is there any
> way to(easilly) figure out which PID is handling the slave thread, so
> that I might re-nice it after it's already been started up?

That probably won't help you, since I doubt you're CPU-bound.  Most
likely your slave thread is monopolizing the disk I/O.  You could try
manually stopping and starting the thread by sending it SIGSTOP and
SIGCONT signals.  Alternating the two every 5 seconds will give you a
50% slowdown.  I don't know how LinuxThreads is going to like having
threads messes with like that though.  You can probably also do SLAVE
STOP / SLAVE START and get the same effect.
	Dan Nelson
Replication priority / speedMatt Sturtz26 Dec
  • Re: Replication priority / speedJeremy Zawodny27 Dec
    • Re: Replication priority / speedMatt Sturtz30 Dec
      • Re: Replication priority / speedJeremy Zawodny30 Dec
      • Re: Replication priority / speedDan Nelson30 Dec
        • Re: Replication priority / speedJeremy Zawodny30 Dec
  • Re: Replication priority / speedSimon K. Grabowski30 Dec
    • Re: Replication priority / speedMatt Sturtz31 Dec
  • Re: Replication priority / speedSimon Grabowski1 Jan