linux-mips
[Top] [All Lists]

Re: uptime is too high. Is it normal?

To: colin <colin@realtek.com.tw>
Subject: Re: uptime is too high. Is it normal?
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 21 Mar 2006 13:36:09 +0000
Cc: linux-mips@linux-mips.org
In-reply-to: <00a101c64ce9$554b3180$106215ac@realtek.com.tw>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <009101c64ce4$72d78820$106215ac@realtek.com.tw> <20060321125255.GA8779@linux-mips.org> <00a101c64ce9$554b3180$106215ac@realtek.com.tw>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.4.2.1i
On Tue, Mar 21, 2006 at 09:14:06PM +0800, colin wrote:

>   646 root            DW  [646]
>   647 root            DW  [647]
> 
> The uptime value keeps on going higher and higher.
> It seems that both 646 & 647 process has the same parent "1".
> Their "utime" are "0".

Yes, these two are the offenders.  A process should never be for a long
time in "D" state.  You'll have to figure out what these processes are
doing, but they something wrong with these two.

The load average btw is a fairly artificial meassure.  On a properly
working system it describes the burn of the workload of the system
reasonably well - but if things go wrong and cause processes to get
stuck in D state (aka uninterruptible sleep) then a system may well
still be perfectly responsive even though the load is very high.

  Ralf

<Prev in Thread] Current Thread [Next in Thread>