Date: Thu, 10 Jun 2010 13:06:09 +0200 From: Anders Nordby <anders@FreeBSD.org> To: Peter Jeremy <peter@vk2pj.dyndns.org> Cc: freebsd-fs@FreeBSD.org Subject: Re: Odd network issues on ZFS based NFS server Message-ID: <20100610110609.GA87243@fupp.net> In-Reply-To: <20100610081710.GA64350@server.vk2pj.dyndns.org> References: <20100608083649.GA77452@fupp.net> <Pine.GSO.4.63.1006081946040.8742@muncher.cs.uoguelph.ca> <20100609122517.GA16231@fupp.net> <20100610081710.GA64350@server.vk2pj.dyndns.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi,
On Thu, Jun 10, 2010 at 06:17:10PM +1000, Peter Jeremy wrote:
> I wonder if your system is running out of free RAM. How would you
> like to monitor "inactive", "cache" and "free" from either "systat -v"
> or "vmstat -s" whilst the problem is occurring.
>
> Does something like
> perl -e '$x = "x" x 10000000;'
> temporarily correct the problem?
While the problem is happening:
root@unixfile:~# vmstat -s
511745441 cpu context switches
151635080 device interrupts
14028218 software interrupts
11549957 traps
974939023 system calls
22 kernel threads created
77512 fork() calls
6097 vfork() calls
0 rfork() calls
0 swap pager pageins
0 swap pager pages paged in
0 swap pager pageouts
0 swap pager pages paged out
699 vnode pager pageins
4777 vnode pager pages paged in
2024 vnode pager pageouts
2471 vnode pager pages paged out
0 page daemon wakeups
0 pages examined by the page daemon
318 pages reactivated
4738808 copy-on-write faults
4957 copy-on-write optimized faults
3843376 zero fill pages zeroed
0 zero fill pages prezeroed
2273 intransit blocking page faults
11236873 total VM faults taken
0 pages affected by kernel thread creation
20699066 pages affected by fork()
1707164 pages affected by vfork()
0 pages affected by rfork()
363 pages cached
27229532 pages freed
0 pages freed by daemon
6618712 pages freed by exiting processes
6054 pages active
37307 pages inactive
28 pages in VM cache
261148 pages wired down
456560 pages free
4096 bytes per page
43744208 total name lookups
cache hits (19% pos + 1% neg) system 0% per-directory
deletions 2%, falsehits 0%, toolong 0%
And from systat -v:
Disks da0 da1 pass0 pass1 1045240 wire
KB/t 0.00 0.00 0.00 0.00 25240 act
tps 0 0 0 0 149344 inact
MB/s 0.00 0.00 0.00 0.00 112 cache
%busy 0 0 0 0 1824452 free
323680 buf
> Does something like
> perl -e '$x = "x" x 10000000;'
> temporarily correct the problem?
No.
Regards,
--
Anders.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100610110609.GA87243>