Date: Mon, 26 Jun 2006 13:54:14 +0100 From: "Joao Barros" <joao.barros@gmail.com> To: "Scott Mitchell" <scott+lists.freebsd@fishballoon.org> Cc: freebsd-questions <freebsd-questions@freebsd.org>, Alex Zbyslaw <xfb52@dial.pipex.com> Subject: Re: Hot-Swapping hard drives on Dell PowerEdge 2850 running FBSD 5.5-PRE Message-ID: <70e8236f0606260554i71131819o5e776dea33fa61a5@mail.gmail.com> In-Reply-To: <20060626104957.GB67667@llama.fishballoon.org> References: <2f488c030606221106q4183de17gbff80d696f704505@mail.gmail.com> <449BAE45.7030705@dial.pipex.com> <2f488c030606231113i2b9b4bdcsa1b4192be54b9011@mail.gmail.com> <70e8236f0606231240t77e1b510xd5ecd98c093af559@mail.gmail.com> <2f488c030606231623x662d93afl8ff4875989d143c8@mail.gmail.com> <70e8236f0606240917m2d872902sce15c7cbacbdbc36@mail.gmail.com> <449FB7B2.7030808@dial.pipex.com> <20060626104957.GB67667@llama.fishballoon.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 6/26/06, Scott Mitchell <scott+lists.freebsd@fishballoon.org> wrote:
> Yeah, megarc does have possibly the worst interface I've seen in quite a
> long time. Allegedly the Linux monitoring tools for these adapters will
> work with 6.1, although I haven't tried this myself yet.
Yes, I think it was Doug Ambrisko who put in the shims for that, but
haven't tested it.
And yes, megarc should be on wikipedia as a bad example for bad interfaces.
>
> I've attached a couple of scripts I use for monitoring amr(4) adapters.
> One (amr-check-status) is run hourly from crontab to alert of any change
> in drive or array status. The other (700.amr-status) is a daily periodic
> script.
amrstat from ports also include a daily crontab script which I'm hapilly using.
For the hourly job, there is a function in the amr(4) driver that's
supposed to do a constant check of the controller but alas the
function is just declared, empty in function. If this was to work, a
message to syslog would be enough to send out an email with an alert
for example.
I started this weekend(again) my port of bio from OpenBSD and I really
like what they did with the sysctl variables and with sensorsd like:
$ sysctl hw.sensors
hw.sensors.0=sd0, ami0 0, drive online, OK
hw.sensors.1=sd1, ami0 1, degraded, WARN
hw.sensors.2=sd2, ami0 2, failed, CRITICAL
With sensorsd on top of that, monitoring is a breeze.
I think after bio I know what I will do next ;-)
--
Joao Barros
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?70e8236f0606260554i71131819o5e776dea33fa61a5>