From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bill Davidsen Subject: Re: mdadm 2.6.4 : How i can check out current status of reshaping ? Date: Thu, 07 Feb 2008 16:15:58 -0500 Message-ID: <47AB750E.2060202@tmr.com> References: <79188012.20080204070802@j8.com.ru> <18343.38465.112723.66522@notabene.brown> <58351009.20080205121332@j8.com.ru> <18344.13816.921912.885730@notabene.brown> <119270390.20080206221505@j8.com.ru> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <119270390.20080206221505@j8.com.ru> Sender: linux-raid-owner@vger.kernel.org To: Andreas-Sokov Cc: Neil Brown , linux-raid@vger.kernel.org List-Id: linux-raid.ids Andreas-Sokov wrote: > Hello, Neil. > > ..... > >> Possible you have bad memory, or a bad CPU, or you are overclocking >> the CPU, or it is getting hot, or something. >> > > As seems to me all my problems has been started after i have started update MDADM. > This is server worked normaly (but only not like soft-raid) more 2-3 years. > Last 6 months it worked as soft-raid. All was normaly, Even I have added successfully > 4th hdd into raid5 )when it stared was 3 hdd). And then Reshaping have been passed fine. > > Yesterday i have did memtest86 onto it server and 10 passes was WITH OUT any errors. > Temperature of server is about 25 grad celsius. > No overlocking, all set to default. > > What did you find when you loaded the module with gdb as Neil suggested? If the code in the module doesn't match the code in memory you have a hardware error. memtest86 is a useful tool, but it is not a definitive test because it doesn't use all CPUs and do i/o at the same time to load the memory bus. > Realy i do not know what to do because off wee nedd grow our storage, and we can not. > unfortunately, At this moment - Mdadm do not help us in this decision, but very want > it get. > I would pull out half my memory and retest. If it still fails I would swap to the other half of memory. If that didn't show a change I would check that the code in the module is what Neil showed in his last message (I assume you already have), and then reseat all of the cables, etc. I agree with Neil: >> But you clearly have a hardware error. >> > > >> NeilBrown >> > > > > -- Bill Davidsen "Woe unto the statesman who makes war without a reason that will still be valid when the war is over..." Otto von Bismark