Re: Why so many kprocs?

From: Yard, John (jyard@AIS.UCLA.EDU)
Date: Wed Aug 20 2003 - 14:53:52 EDT

• Next message: Bill Verzal: "SCSI Shark question"
• Previous message: Aaron W Morris: "Re: echo to stderr"
• Maybe in reply to: Patrick B. O'Brien: "Why so many kprocs?"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

We have 1200, heavy, heavy async i/o -
data warehouse
 
JYard
UCLA

        -----Original Message-----
        From: Colin [mailto:colin@THEDOMAIN.CO.ZA]
        Sent: Wednesday, August 20, 2003 10:42 AM
        To: aix-l@Princeton.EDU
        Subject: Re: Why so many kprocs?
        
        

        Look at your asynchronous IO 'lsattr -El aio0'. how many
maxservers are running and is autoconfig=available ?

        The kprocs will be relative to your maxservers in use.

        Usually when there are large numbers of disks, one should
increase the maxservers.

        Don't kill the kproc's

        -----Original Message-----
        From: IBM AIX Discussion List [mailto:aix-l@Princeton.EDU] On
Behalf Of Patrick B. O'Brien
        Sent: 20 August 2003 07:22 PM
        To: aix-l@Princeton.EDU
        Subject: Why so many kprocs?

        One of my RS6000 P660 ML11 Oracle Boxes has 133 Kproc jobs and
another only has 33. Can any one explain to me why there are so many
Kproc's on that system? Can I kill some of them ECT?

        TIA!!!

• Next message: Bill Verzal: "SCSI Shark question"
• Previous message: Aaron W Morris: "Re: echo to stderr"
• Maybe in reply to: Patrick B. O'Brien: "Why so many kprocs?"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 22:17:09 EDT