Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 15 Oct 2013 09:37:35 -0400
From:      Allan Jude <freebsd@allanjude.com>
To:        freebsd-current@freebsd.org
Subject:   Re: vmstat -z: zfs related failures on r255173
Message-ID:  <525D451F.3040005@allanjude.com>
In-Reply-To: <1381837350.494453682.d94z37mg@frv45.ukr.net>
References:  <1381521443.74219922.8eyuscyw@frv45.ukr.net> <1381837350.494453682.d94z37mg@frv45.ukr.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 2013-10-15 07:53, Dmitriy Makarov wrote:
> Please, any idea, thougth, help! 
> Maybe what information can be useful for diggin - anything...
>
> System what I'm talkin about has a huge problem: performance degradation in short time period (day-two). Don't know can we somehow relate this vmstat fails with degradation.
>
>
>  
>> Hi all
>>
>> On CURRENT r255173 we have some interesting values from vmstat -z : REQ = FAIL
>>
>> [server]# vmstat -z
>> ITEM                   SIZE  LIMIT     USED     FREE      REQ FAIL SLEEP
>> ....... skipped....
>> NCLNODE:                528,      0,       0,       0,       0,   0,   0
>> space_seg_cache:         64,      0,  289198,  299554,25932081,25932081,   0
>> zio_cache:              944,      0,   37512,   50124,1638254119,1638254119,   0
>> zio_link_cache:          48,      0,   50955,   38104,1306418638,1306418638,   0
>> sa_cache:                80,      0,   63694,      56,  198643,198643,   0
>> dnode_t:                864,      0,  128813,       3,  184863,184863,   0
>> dmu_buf_impl_t:         224,      0, 1610024,  314631,157119686,157119686,   0
>> arc_buf_hdr_t:          216,      0,82949975,   56107,156352659,156352659,   0
>> arc_buf_t:               72,      0, 1586866,  314374,158076670,158076670,   0
>> zil_lwb_cache:          192,      0,    6354,    7526, 2486242,2486242,   0
>> zfs_znode_cache:        368,      0,   63694,      16,  198643,198643,   0
>> ..... skipped ......
>>
>> Can anybody explain this strange failures in zfs related parameters in vmstat, can we do something with this and is this really bad signal?
>>
>> Thanks! 
>
> _______________________________________________
> freebsd-current@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
I am guessing those 'failures' are failures to allocate memory. I'd
recommend you install sysutils/zfs-stats and send the list the output of
'zfs-stats -a'

-- 
Allan Jude




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?525D451F.3040005>