Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
Package: linux-2.6 Severity: important Tags: patch Due to a NFS attribute revalidation problem it might happen that lseek(fd, 0, SEEK_END) returns a stale file size. Please see https://bugzilla.redhat.com/show_bug.cgi?id=672981 and http://archives.postgresql.org/pgsql-hackers/2011-01/msg02611.php for details. It would be appreciated if the bug could be fixed in the next point-releases for Lenny and Squeeze. Thanks and cheers, Bernd -- Bernd ZeimetzDebian GNU/Linux Developer http://bzed.dehttp://www.debian.org GPG Fingerprints: ECA1 E3F2 8E11 2432 D485 DD95 EB36 171A 6FF9 435F -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4d762f78.4070...@bzed.de
Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
On Tue, Mar 08, 2011 at 02:30:32PM +0100, Bernd Zeimetz wrote: > Package: linux-2.6 > Severity: important > Tags: patch > > Due to a NFS attribute revalidation problem it might happen that > lseek(fd, 0, SEEK_END) returns a stale file size. > > Please see https://bugzilla.redhat.com/show_bug.cgi?id=672981 and > http://archives.postgresql.org/pgsql-hackers/2011-01/msg02611.php for > details. It would be appreciated if the bug could be fixed in the next > point-releases for Lenny and Squeeze. > > Thanks and cheers, The ideal way to get this fixed in Squeeze is to get the patches merged into the 2.6.32 long term kernel, on which Squeeze is based. Could you check back with Joe Conway and ask for commits IDs of the upstream fixes? Cheers, Moritz -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110401195427.GA13123@pisco.westfalen.local
Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
Hi, > Could you check back with Joe Conway and ask for commits IDs of the > upstream fixes? I've got the following info from Trond Myklebust: The upstream fix is commit 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 (NFS: nfs_wcc_update_inode() should set nfsi->attr_gencount). Cheers, Bernd -- Bernd ZeimetzDebian GNU/Linux Developer http://bzed.dehttp://www.debian.org GPG Fingerprints: ECA1 E3F2 8E11 2432 D485 DD95 EB36 171A 6FF9 435F -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4dac102f.6050...@bzed.de
Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
>> Could you check back with Joe Conway and ask for commits IDs of the >> upstream fixes? > > I've got the following info from Trond Myklebust: > > The upstream fix is commit 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 > (NFS: nfs_wcc_update_inode() should set nfsi->attr_gencount). Greg, please merge 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 into 2.6.32.x (and other supported longterm kernels) It fixes the bug described in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=617364 Cheers, Moritz -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110422202822.GA5376@pisco.westfalen.local
Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
tags 617364 + upstream fixed-upstream fixed 617364 linux-2.6/2.6.38~rc6-1~experimental.1 quit Bernd Zeimetz wrote: > The upstream fix is commit 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 > (NFS: nfs_wcc_update_inode() should set nfsi->attr_gencount). Quick update: it's part of 2.6.35.13 now and there's a backport to 2.6.32.x in the longterm queue[1] under the name[2] queue-2.6.32/nfs-nfs_wcc_update_inode-should-set-nfsi-attr_gencount.patch [1] git://git.kernel.org/pub/scm/linux/kernel/git/longterm/longterm-queue-2.6.32.git [2] http://git.kernel.org/?p=linux/kernel/git/longterm/longterm-queue-2.6.32.git;a=blob;f=queue-2.6.32/nfs-nfs_wcc_update_inode-should-set-nfsi-attr_gencount.patch;hb=HEAD -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110430062327.GA8748@elie
Bug#617364: [stable] Bug#617364: linux-2.6: lseek() over NFS is returning an incorrect file length under some, circumstances
On Fri, Apr 22, 2011 at 10:28:22PM +0200, Moritz Muehlenhoff wrote: > > > >> Could you check back with Joe Conway and ask for commits IDs of the > >> upstream fixes? > > > > I've got the following info from Trond Myklebust: > > > > The upstream fix is commit 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 > > (NFS: nfs_wcc_update_inode() should set nfsi->attr_gencount). > > Greg, > please merge 27dc1cd3ad9300f81e1219e5fc305d91d85353f8 into 2.6.32.x > (and other supported longterm kernels) > > It fixes the bug described in > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=617364 Now added to the .32 and .33 longterm kernel branches, thanks. greg k-h -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110422203301.ga32...@kroah.com