Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 7 Nov 2005 22:54:57 +0100
From:      =?ISO-8859-1?Q?S=F8ren_Schmidt?= <sos@DeepCore.dk>
To:        Michael Butler <imb@protected-networks.net>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: 6-stable unstable with HighPoint HPT372N UDMA133 controller
Message-ID:  <A52B26A3-AE26-4D59-B160-43D7ED0C35C7@DeepCore.dk>
In-Reply-To: <436F8A6D.5080208@protected-networks.net>
References:  <436E7B02.9020207@protected-networks.net> <436E8290.3050604@protected-networks.net> <436F8A6D.5080208@protected-networks.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 07/11/2005, at 18:10, Michael Butler wrote:
>
> | I wrote:
> | | cvsup'd and built: FreeBSD 6.0-STABLE #4: Fri Nov  4 18:07:30 =20
> EST 2005
> |
> | ~ [ .. ]
> |
> | | Nov  6 16:43:27 mail kernel: DOH! ata_alloc_request failed!
> | | Nov  6 16:43:27 mail kernel: FAILURE - out of memory in
> | ata_raid_init_request
> | | Nov  6 16:43:27 mail last message repeated 7 times
> | | Nov  6 16:43:27 mail kernel:
>
> Looking at the output of "sysctl -a" on a now almost idle machine I =20=

> see:
>
> ITEM            SIZE     LIMIT     USED    FREE  REQUESTS
>
> ~ [ .. ]
>
> ata_composit:    192,        0,  12296,    124,    62341
> ata_request:     200,        0,  24592,    108,   920945
>
> ~ .. shouldn't these be freed at some point if there's minimal disk
> activity (and few dirty buffers, "systat -vm" says there are 7)? Or =20=

> am I
> misreading this?

That does indeed look wrong. However I cant reproduce the problem =20
here on any of the machines I have 6.0 on, and I dont think a memory =20
leeak that severe would have survived this far, but I've been wrong =20
many times before...

I'll look into it as soon as I have a little more time than today..

- S=F8ren





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?A52B26A3-AE26-4D59-B160-43D7ED0C35C7>