From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arkadiusz Miskiewicz Subject: Re: I/O errors without erros from underlying device Date: Mon, 7 Dec 2015 21:46:28 +0100 Message-ID: <201512072146.28378.a.miskiewicz@gmail.com> References: <201512071705.27177.a.miskiewicz@gmail.com> <201512071803.26434.arekm@maven.pl> <22117.49283.546268.719858@quad.stoffel.home> Reply-To: arekm@maven.pl Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-2 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <22117.49283.546268.719858@quad.stoffel.home> Sender: linux-raid-owner@vger.kernel.org To: John Stoffel Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On Monday 07 of December 2015, John Stoffel wrote: > >>>>> "Arkadiusz" =3D=3D Arkadiusz Mi=B6kiewicz writ= es: > Arkadiusz> On Monday 07 of December 2015, John Stoffel wrote: >=20 > Arkadiusz> 4.3.0 kernel, raid6 array: > >> I think there's a bug in the 4.3.x and 4.4-rc3 and lower with bloc= k > >> merges. I ran into these over the weekend, where v4.2.6 was stabl= e, > >> but anything higher would lock up and crash on me. >=20 > Arkadiusz> Well, no crashes here. >=20 > That's good. It was hard(er) to hit when I wasn't running KVM VMs at > the same time on the server, and I was running strictly RAID1 disks, > so it's hard to know. >=20 > >> So first step would be to make sure you get and test v4.4-rc4. >=20 > Arkadiusz> Do you know which commit there? >=20 > Try this, from the master lkml git repository: >=20 > 2873d32ff493ecbfb7d2c7f56812ab941dda42f4 It's merge commit. Don't see any obvious patch in that merge that would= help=20 my case. Anyway I would expect my problem to be related to badblock lists which = numbers=20 are close to dmesg error message: [ 848.988518] Buffer I/O error on de= v md7,=20 logical block 3907148544, async page read > >> http://sprunge.us/XSWI But how to repair these if write() also fails and=20 http://www.spinics.net/lists/raid/msg49325.html suggests that write sho= uld=20 "fix" these (by using replacement blocks I guess) ? =20 > Arkadiusz> md7 : active raid6 sdg[10] sdad1[9] sdac1[8] sdag1[7] sdaf= 1[6] >=20 > >> sdae1[5] sdaj1[4] sdai1[3] sdah1[2] sdn1[1] Arkadiusz> 31255= 089152 > >> blocks super 1.2 level 6, 512k chunk, algorithm 2 [10/10] [UUUUUUU= UUU] >=20 > Arkadiusz> bitmap: 1/30 pages [4KB], 65536KB chunk >=20 > Arkadiusz> array had weird failure where many disks went into failed = state >=20 > >> but Arkadiusz> remove && adding these disks "fixed" it (turns out = not > >> really fixed it). >=20 > Arkadiusz> Unfortunately now some reads fail: >=20 > Arkadiusz> pread(4, 0x1483a00, 4096, 16003680464896) =3D -1 EIO (Inpu= t/output >=20 > >> error) >=20 > Arkadiusz> To reproduce used xfs_io > Arkadiusz> xfs_io -d -c "pread 16003680464896 4096" /dev/md7 > Arkadiusz> pread64: Input/output error > Arkadiusz> which does pread exactly as shown above. >=20 > Arkadiusz> write also fails for that area: > Arkadiusz> xfs_io -d -c "pwrite 16003680464896 4096" /dev/md7 > Arkadiusz> pwrite64: Input/output error >=20 > Arkadiusz> Note that nothing is written in dmesg when that happens. >=20 > Arkadiusz> I've tried various offsets and sizes of pread and at some = point >=20 > >> that was logged: Arkadiusz> [ 848.988518] Buffer I/O error on dev= md7, > >> logical block 3907148544, async page read >=20 > Arkadiusz> but no error from underlying devices. >=20 > Arkadiusz> List of bad blocks: > Arkadiusz> http://sprunge.us/XSWI >=20 > Arkadiusz> What can I do now? >=20 > Arkadiusz> (loosing data from that few sectors is acceptable if the r= est >=20 > >> will be readable) >=20 > Arkadiusz> Thanks, > Arkadiusz> -- > Arkadiusz> Arkadiusz Mi=B6kiewicz, arekm / ( maven.pl | pld-linux.org= ) > Arkadiusz> -- > Arkadiusz> To unsubscribe from this list: send the line "unsubscribe >=20 > >> linux-raid" in Arkadiusz> the body of a message to > >> majordomo@vger.kernel.org >=20 > Arkadiusz> More majordomo info at >=20 > >> http://vger.kernel.org/majordomo-info.html >=20 > Arkadiusz> -- > Arkadiusz> Arkadiusz Mi=B6kiewicz, arekm / ( maven.pl | pld-linux.org= ) --=20 Arkadiusz Mi=B6kiewicz, arekm / ( maven.pl | pld-linux.org ) --=20 Arkadiusz Mi=B6kiewicz, arekm / ( maven.pl | pld-linux.org ) -- 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