Date: Wed, 29 Jan 2014 08:33:30 +1000 From: Andy Farkas <andyf@andyit.com.au> To: freebsd-stable@freebsd.org Subject: Re: Processes are incorrectly marked as swapped out Message-ID: <52E8303A.9020501@andyit.com.au> In-Reply-To: <201401281142.10317.jhb@freebsd.org> References: <2D47B79E-C171-4B91-B0AB-4DD2212770C6@gmail.com> <op.w9446fkykndu52@ronaldradial.radialsg.local> <201401281142.10317.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 29/01/14 02:42, John Baldwin wrote:
> On Thursday, January 23, 2014 8:02:29 am Ronald Klop wrote:
>> On Thu, 23 Jan 2014 13:19:36 +0100, Dmitry Sivachenko
>> <trtrmitya@gmail.com> wrote:
>>
>>> Hello!
>>>
>>> After upgrade from stable/9 to stable/10 I see the following regression.
>>> Some processes are marked as swapped out in top(1) output:
>>>
>>> 1436 root 1 43 0 16524K 0K nanslp 14 1:14 0.00%
>>> <cron>
>>> 1381 smmsp 1 20 0 23988K 0K pause 18 0:04 0.00%
>>> <sendma
>>> 99348 mitya 1 21 0 23492K 0K pause 16 0:00 0.00%
>>> <csh>
>>>
>>> ps(1) also shows them as swapped out (W as second character in state
>>> field):
>>> 1381 - IWs 0:00.00 sendmail: Queue runner at 00:30:00 for
>>> /var/spool/clie
>>> 1436 - IWs 0:00.00 /usr/sbin/cron -s
>>> 80231 - IWs 0:00.00 /usr/local/sbin/collectdmon -c
>>> /usr/local/sbin/coll
>>> 99348 1 IWs 0:00.00 -csh (csh)
>>>
>>> Though swapinfo reports that zero swap is used and even if I turn swap
>>> completely off (swapoff -a)
>>> the output of both top(1) and ps(1) does not change: these processes are
>>> still marked as swapped out.
>> The code of an application can get removed from memory, because there
>> still is an image of it in the executable on disk. It can be 'swapped' in
>> by reading the executable again. The program is memory mapped (mmap).
>> See VN PAGER vs SWAP PAGER in 'systat -vm'.
> However, a swapped out process always uses swap (for kernel stacks), so this
> seems like a real bug.
>
Probably not related, but on a VirtualBox VM guest running:
FreeBSD 10.0-STABLE (GENERIC) #0 r261215: Tue Jan 28 18:13:37 EST 2014
having been freshly rebooted:
% uptime
7:59AM up 43 secs, 2 users, load averages: 0.79, 0.30, 0.11
I see this strange anomaly in the "TIME" column of top(1):
% top -CHSb -otime | grep swap
0 root -16 0 0K 160K swapin 0 47.8H 0.00%
kernel{swapper}
No swap is being used (of ~3GB). Filesystem is UFS.
Possibly a formatting error although earlier values have been 45.4H & 47.2H
with reboots in between.
-andyf
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52E8303A.9020501>