Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 4 Mar 2002 08:15:15 -0600
From:      mikea <mikea@mikea.ath.cx>
To:        stable@FreeBSD.ORG, freebsd-isp@FreeBSD.ORG
Subject:   Re: Machine rebooting frequently
Message-ID:  <20020304081515.A43223@mikea.ath.cx>
In-Reply-To: <001f01c1c37f$e073e880$8301a8c0@aubonne.virtua.ch>; from marcel-lists@virtua.ch on Mon, Mar 04, 2002 at 02:24:13PM %2B0100
References:  <001f01c1c37f$e073e880$8301a8c0@aubonne.virtua.ch>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 04, 2002 at 02:24:13PM +0100, Marcel Prisi wrote:
> Help !!
> 
> We have a postgresql server (bi-PIII 833, 1Gb RAM, Adaptec SCSI RAID)
> rebooting by itself more and more frequently.
> 
> I can not find any message anywhere, as if someone just presses the "reboot"
> button every now and then. It used to reboot every week or so, but it
> rebooted just six times today !
> 
> What can I do to fond the cause ? The machine runs 4.5-RELEASE, but used to
> run 4.3-PRERELEASE with the same trouble.
> 
> I partially checked RAM through memtest86 (www.memtest86.com) but did not
> find any trouble. What might it be ? faulty processor ? Faulty power-supply
> ? any clues ? how to test ?

The program-driven memory testers are at best rather bad at 
finding memory problems. 

Candidates, not in any particular order:
o	Overheated processor(s)
o	Overheated or faulty memory
o	Faulty or overloaded power supply
o	Faulty line power from the wall
o	Flaky motherbord (_Highly_ unlikely)

and, inevitably, 

o	"Other"

Problem Determination and Isolation techniques:

o Processor: run healthd or another voltage/temperature monitor
to watch voltages and temperatures. Check all fans for good
speed. Remove lint, cat-hair, etc., from heat sinks.[1] 

o Faulty or overheated memory: Check fans, etc., as above Take . 
memory to a store that has a hardware memory tester Remove one . 
SIMM at a time, reboot, wait                                   . 

o Power Supply: Check DC power with voltmeter and oscilloscope.
Replace PS if you _think_ there might be a problem with it or if
it is anywhere near capacity. They're cheap here: about $100 will
get you a very-high-capacity PS.

o Faulty line power: Put the server on a known-good UPS. 

o Flaky motherboard: Replace motherboard (absolute last resort).


[1]		Cat hair in my house appears to be attracted preferentially
		to CPU heat sinks. 

-- 
Mike Andrews
mikea@mikea.ath.cx
Tired old sysadmin since 1964

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020304081515.A43223>