From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Hesse Subject: Re: iSCSI regression with linux 3.9 and 4.0 Date: Mon, 23 Mar 2015 09:53:08 +0100 Message-ID: <20150323095308.196dd451@leda.localdomain> References: <20150320135742.63df219b@leda.localdomain> <1426859486.19806.30.camel@localhost.localdomain> <20150320153104.086f3b46@leda.localdomain> <1426863883.19806.65.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; boundary="Sig_/VMAkbbRv3yY6kZKT6f+FoPG"; protocol="application/pgp-signature" Return-path: Received: from mx.mylinuxtime.de ([148.251.109.235]:35623 "EHLO mx.mylinuxtime.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751853AbbCWIxP (ORCPT ); Mon, 23 Mar 2015 04:53:15 -0400 In-Reply-To: <1426863883.19806.65.camel@localhost.localdomain> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Ewan Milne Cc: linux-scsi@vger.kernel.org --Sig_/VMAkbbRv3yY6kZKT6f+FoPG Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Ewan Milne on Fri, 2015/03/20 11:04: > On Fri, 2015-03-20 at 15:31 +0100, Christian Hesse wrote: > > Ewan Milne on Fri, 2015/03/20 09:51: > > > On Fri, 2015-03-20 at 13:57 +0100, Christian Hesse wrote: > > > > Hello everybody! > > > >=20 > > > > I reported this issue at LKML [0] but received no answer. Hopefully > > > > linux-scsi is a better place... > > > >=20 > > > > Beginning with linux 3.19 I see an iSCSI regressen. This works > > > > perfectly with linux 3.18.x (tested with 3.18.6) and before. Effect= ed > > > > kernels I tested are 3.19.0, 3.19.2 and 4.0rc4.r199.gb314aca. > > > >=20 > > > > The logs tell the story: > > > >=20 > > > > Feb 19 11:26:49 thebe kernel: scsi host6: iSCSI Initiator over TCP/= IP > > > > Feb 19 11:26:49 thebe kernel: scsi 6:0:0:0: Direct-Access QNAP > > > > iSCSI Storage 4.0 PQ: 0 ANSI: 5 Feb 19 11:26:49 thebe kernel: sd > > > > 6:0:0:0: [sdb] 1073741824 512-byte logical blocks: (549 GB/512 GiB) > > > > Feb 19 11:26:49 thebe kernel: sd 6:0:0:0: [sdb] Write Protect is off > > > > Feb 19 11:26:49 thebe kernel: sd 6:0:0:0: [sdb] Mode Sense: 2f 00 00 > > > > 00 Feb 19 11:26:49 thebe kernel: sd 6:0:0:0: [sdb] Write cache: > > > > disabled, read cache: enabled, doesn't support DPO or FUA Feb 19 > > > > 11:26:49 thebe kernel: sdb: unknown partition table Feb 19 11:26:49 > > > > thebe kernel: sd 6:0:0:0: [sdb] Attached SCSI disk Feb 19 11:26:49 > > > > thebe iscsid[10804]: Connection1:0 to [target: > > > > iqn.2004-04.com.qnap:ts-859:iscsi.xxxxxxx.c40a18, portal: > > > > xx.xx.xx.xx,3260] through [iface: default] is operational now Feb 19 > > > > 11:26:57 thebe kernel: sdb: unknown partition table Feb 19 11:28:20 > > > > thebe kernel: EXT4-fs (dm-8): mounting with "discard" option, but t= he > > > > device does not support discard Feb 19 11:28:20 thebe kernel: EXT4-= fs > > > > (dm-8): mounted filesystem with ordered data mode. Opts: (null) Feb= 19 > > > > 11:28:24 thebe kernel: sd 6:0:0:0: [sdb] UNKNOWN Result: hostbyte= =3D0x00 > > > > driverbyte=3D0x08 Feb 19 11:28:24 thebe kernel: sd 6:0:0:0: [sdb] S= ense > > > > Key : 0x5 [current] Feb 19 11:28:24 thebe kernel: sd 6:0:0:0: [sdb] > > > > ASC=3D0x24 ASCQ=3D0x0 Feb 19 11:28:24 thebe kernel: sd 6:0:0:0: [sd= b] CDB: > > > > Feb 19 11:28:24 thebe kernel: cdb[0]=3D0x2a: 2a 00 34 5b 07 ff 00 2= f 88 > > > > 00 Feb 19 11:28:24 thebe kernel: blk_update_request: critical target > > > > error, dev sdb, sector 878381055 Feb 19 11:28:24 thebe kernel: > > > > EXT4-fs warning (device dm-8): ext4_end_bio:317: I/O error -121 > > > > writing to inode 33196503 (offset 8388608 size 7278592 starting blo= ck > > > > 108749056) Feb 19 11:28:24 thebe kernel: Buffer I/O error on device > > > > dm-8, logical block 108749056 Feb 19 11:28:24 thebe kernel: Buffer > > > > I/O error on device dm-8, logical block 108749057 Feb 19 11:28:24 > > > > thebe kernel: Buffer I/O error on device dm-8, logical block > > > > 108749058 Feb 19 11:28:24 thebe kernel: Buffer I/O error on device > > > > dm-8, logical block 108749059 Feb 19 11:28:24 thebe kernel: Buffer > > > > I/O error on device dm-8, logical block 108749060 Feb 19 11:28:24 > > > > thebe kernel: Buffer I/O error on device dm-8, logical block > > > > 108749061 Feb 19 11:28:24 thebe kernel: Buffer I/O error on device > > > > dm-8, logical block 108749062 Feb 19 11:28:24 thebe kernel: Buffer > > > > I/O error on device dm-8, logical block 108749063 Feb 19 11:28:24 > > > > thebe kernel: Buffer I/O error on device dm-8, logical block > > > > 108749064 Feb 19 11:28:24 thebe kernel: Buffer I/O error on device > > > > dm-8, logical block 108749065 Feb 19 11:28:24 thebe kernel: EXT4-fs > > > > warning (device dm-8): ext4_end_bio:317: I/O error -121 writing to > > > > inode 33196503 (offset 8388608 size 7278592 starting block 10874931= 2) > > > > Feb 19 11:28:24 thebe kernel: EXT4-fs warning (device dm-8): > > > > ext4_end_bio:317: I/O error -121 writing to inode 33196503 (offset > > > > 8388608 size 7278592 starting block 108749568) Feb 19 11:28:24 thebe > > > > kernel: EXT4-fs warning (device dm-8): ext4_end_bio:317: I/O error > > > > -121 writing to inode 33196503 (offset 8388608 size 7278592 starting > > > > block 108749824) Feb 19 11:28:24 thebe kernel: EXT4-fs warning > > > > (device dm-8): ext4_end_bio:317: I/O error -121 writing to inode > > > > 33196503 (offset 8388608 size 7278592 starting block 108750080) Feb > > > > 19 11:28:24 thebe kernel: EXT4-fs warning (device dm-8): > > > > ext4_end_bio:317: I/O error -121 writing to inode 33196503 (offset > > > > 8388608 size 7278592 starting block 108750336) Feb 19 11:29:10 thebe > > > > kernel: sd 6:0:0:0: [sdb] UNKNOWN Result: hostbyte=3D0x00 > > > > driverbyte=3D0x08 Feb 19 11:29:10 thebe kernel: sd 6:0:0:0: [sdb] S= ense > > > > Key : 0x5 [current] Feb 19 11:29:10 thebe kernel: sd 6:0:0:0: [sdb] > > > > ASC=3D0x24 ASCQ=3D0x0 Feb 19 11:29:10 thebe kernel: sd 6:0:0:0: [sd= b] > > > > CDB: Feb 19 11:29:10 thebe kernel: cdb[0]=3D0x2a: 2a 00 20 44 89 17= 00 > > > > 20 50 00 Feb 19 11:29:10 thebe kernel: blk_update_request: critical > > > > target error, dev sdb, sector 541362455 Feb 19 11:29:10 thebe kerne= l: > > > > Buffer I/O error on dev dm-8, logical block 66621731, lost sync page > > > > write Feb 19 11:29:10 thebe kernel: Aborting journal on device > > > > dm-8-8. Feb 19 11:29:10 thebe kernel: EXT4-fs error (device dm-8): > > > > ext4_journal_check_start:56: Detected aborted journal Feb 19 11:29:= 10 > > > > thebe kernel: EXT4-fs (dm-8): Remounting filesystem read-only Feb 19 > > > > 11:29:20 thebe kernel: EXT4-fs error (device dm-8): > > > > ext4_put_super:780: Couldn't clean up the journal > > > >=20 > > > > [0] https://lkml.org/lkml/2015/2/19/91 > > >=20 > > > Sense key 0x5 ASC/ASCQ 0x24 0x00 is ILLEGAL REQUEST, INVALID FIELD IN > > > CDB. The CDB was 2A 00 34 5B 07 FF 00 2F 88 00, which is a WRITE_10 > > > to LBA 878381055 with a length of 12168 blocks (a little less than 6M= B). > > > It looks like this is within the reported capacity of the device, and > > > there are no other bits set in the CDB. > > >=20 > > > Looks like you could get this error if RWWP (reject without write > > > protection) is set in the control mode page. I don't see any messages > > > about the protection type, though. What does sysfs report? > >=20 > > Is that what you are interested in? > >=20 > > # cat protection_mode protection_type=20 > > none > > 0 > >=20 > > In case it matters: The iSCSI device is LUKS encrypted, that is why dev= ice > > mapper shows up. > >=20 > > I removed the discard option from filesystem's default mount option, but > > that brings no difference except the message is not printed. >=20 > It is most likely the device that is returning the error, there is a > place in the iSCSI Initiator that generates an ILLEGAL REQUEST sense, > but it is not the same ASC/ASCQ. >=20 > There was this change: >=20 > commit bcdb247c6b6a1f3e72b9b787b73f47dd509d17ec > Author: Martin K. Petersen > Date: Tue Jun 3 18:45:51 2014 -0400 >=20 > sd: Limit transfer length > =20 > Until now the per-command transfer length has exclusively been gated = by > the max_sectors parameter in the scsi_host template. Given that the s= ize > of this parameter has been bumped to an unsigned int we have to be > careful not to exceed the target device's capabilities. > =20 > If the if the device specifies a Maximum Transfer Length in the Block > Limits VPD we'll use that value. Otherwise we'll use 0xffffffff for > devices that have use_16_for_rw set and 0xffff for the rest. We then > combine the chosen disk limit with max_sectors in the host template. = The > smaller of the two will be used to set the max_hw_sectors queue limit. > =20 > Signed-off-by: Martin K. Petersen > Reviewed-by: Ewan D. Milne > Signed-off-by: Christoph Hellwig >=20 > What is the value of max_sectors_kb and queue_max_sectors_kb in sysfs > for the device? Is it different than what is reported on 3.18? I reverted these two commits: commit 3a9794d32984b67a6d8992226918618f0e51e5d5 Author: Brian King Date: Thu Jan 29 15:54:40 2015 -0600 sd: Fix max transfer length for 4k disks commit bcdb247c6b6a1f3e72b9b787b73f47dd509d17ec Author: Martin K. Petersen Date: Tue Jun 3 18:45:51 2014 -0400 sd: Limit transfer length =20 iSCSI device still fails, max_sectors_kb is still set to 32767. Looks like there is another change that introduced the regression. --=20 main(a){char*c=3D/* Schoene Gruesse */"B?IJj;MEH" "CX:;",b;for(a/* Chris get my mail address: */=3D0;b=3Dc[a+= +];) putchar(b-1/(/* gcc -o sig sig.c && ./sig */b/42*2-3)*42);} --Sig_/VMAkbbRv3yY6kZKT6f+FoPG Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJVD9R0AAoJEIlFIdfMgPR2dJ8IAMbkzQYFdGrvSzF5htPFDcBz ttdUAeJTNs+EKDz/5C1k6ELNx4chjoEhfRQurzALqdKIZjUSbPOIAtZ8QA479OBg n8/N1HlCQ6dspcCP5Hsk7rrqvW+1pP7221ECwEj/B1bb2ImNqHtbm+jeOtpK+1D9 qL2wvotK3iOPUj7jUtX0OnvQotzFqx+V6h5uzJ3lpfSriXHXHNYgMV1t7NPnCE67 enj2FnKOZPGKX9dVVx0joML8cHW9OEjiBbnseIUcnTFxVW3d66YHKbuErg/5iFA6 CzvvLQqECY0sNX5lonBLATcexUfTQGBV9CjbHavDYkjzEWxVqFEGkT2+ryaaWK0= =4XdQ -----END PGP SIGNATURE----- --Sig_/VMAkbbRv3yY6kZKT6f+FoPG--