From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gavin Flower Subject: Re: RAID6 data-check took almost 2 hours, clicking sounds, system unresponsive Date: Thu, 28 Apr 2011 13:03:39 -0700 (PDT) Message-ID: <327628.84887.qm@web65101.mail.ac2.yahoo.com> References: <4DA773EB.5030005@turmel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <4DA773EB.5030005@turmel.org> Sender: linux-raid-owner@vger.kernel.org To: Phil Turmel Cc: =?iso-8859-1?Q?Mathias_Bur=E9n?= , neilb@suse.de, linux-raid@vger.kernel.org List-Id: linux-raid.ids --- On Fri, 15/4/11, Phil Turmel wrote: > From: Phil Turmel > Subject: Re: RAID6 data-check took almost 2 hours, clicking sounds, s= ystem unresponsive > To: "Gavin Flower" > Cc: "Mathias Bur=E9n" , neilb@suse.de, linux= -raid@vger.kernel.org > Date: Friday, 15 April, 2011, 10:23 > On 04/14/2011 05:12 PM, Gavin Flower > wrote: >=20 > >=20 > > Hi Phil, > >=20 > > I was under the impression that I had an adequate > power supply, so I checked all 5 drives.=A0=20 [...] > >=20 > > Note that Power_Cycle_Count is anomalous only for > /dev/sdc, so would this suggest cable problems? >=20 > No two drives are perfectly identical, so when the drive's > power rail is only slightly overloaded, the least tolerant > drive chokes as the voltage declines (we're talking tens of > milliseconds, here).=A0 As soon as it chokes, the extra > load disappears, and the power supply recovers.=A0 The > other drives carry on.=A0 The drive that choked resets > (*Click*) in time for the block driver to try again, and the > cycle repeats. >=20 > As a test, borrow another power supply and hook just that > one drive to it.=A0 If the problem continues, the drive > is toast.=A0 If the problem goes away, look for a better > power supply.=A0 Note:=A0 for the Barracuda with the > problem, the detailed spec says the 5V load spikes on > activity, not the 12V load.=A0 So make sure the current > capacity of the power supply meets your needs for both 5V > & 12V (plus your motherboard).=A0 Also check if the > power supply has multiple regulators for drive power, and if > you need to re-arrange the connectors to spread the load > evenly amongst them. >=20 > As another test, you can swap all your cables around.=A0 > If the problem is in the cables, the problem will follow the > cables to the drive you moved them to. >=20 > > I am not sure what to make of the other > discrepancies. > >=20 > > Note that sda, sdb, sdd, & sde were bought and put > in at the same time, while sdc was only obtained and > inserted recently. >=20 > So sdc came from a different manufacturing batch, which is > likely to have slightly different tolerances. >=20 > HTH, >=20 > Phil Thanks Phil, A few days ago, I noticed that 2 of my 3 RAID arrays were down to 4 out= of 5 drives - /dev/sdc had been dropped out, the one which made clicki= ng sounds when I ran badblocks. A couple of days ago, my friend Mario brought over his oscilloscope and= a volt meter. The 5 volt rail was showing about 4.7 volts, typically = it should be 5.2 - 5.4 (from memory of what he said), and the voltage l= ooked shaky on the oscilloscope. The old power supply rated at 400 Wat= ts. =20 Mario suggested that power supplies greater than 500 Watts had signific= antly better quality, also he and others said that power supplies tende= d to have reduced capability to supply their maximum power as they age.= So while, 400 Watts seemed nominally adequate for my system, I starte= d looking for ones that wee at least 500 Watts, I also looked at other = features, such as reliability and the ability to support at least 5 sat= a drives without using adapters. I was in the process of checking out various power supplies, when my de= velopment machine ('saturn') refused to complete the boot process due t= o RAID problems. There are many power supplies that would have met my requirements, but = I told Mario that I was prepared to pay a bit extra, if there was real = benefit, as I saw no point in being penny wise and pound foolish as the= y say in England. If the time Mario and I (let alone that of the other= s who advised me) had spent on this problem was costed, it would have b= een more than double the price of the power supply, so I figured paying= a bit extra was a good investment. The one Mario obtained for me was t= he one in stock that met my needs without being too expensive. The new= one is 700 Watts with reasonably robust specifications: Cooler Master = Extreme Power Plus 700W. MTBF > 100,000 hours (11 years), high efficie= ncy 80% at typical load... Reassembling the 2 defective RAID-6 partitions went okay, now all 3 RAI= D partitions are complete. Been running over 16 hours now and no apparent problems. I ran badbloc= ks on all 5 disks concurrently - no clicking sounds were heard, nor wer= e any errors reported. Also the 'ata' errors previously seen on the sys= tems log are absent. I very much appreciate the help provided to me by the people on this li= st. Regards, Gavin -- To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html