Hi, Sorry to resurrect an old issue, but I've just come across the same (or very similar-looking) problem. I'm also on an Openstack Swift storage node with lots of small writes to SSDs as in the OP, running on Debian Stretch in our case with the following kernel:
Linux swift-storage-1 4.9.0-6-amd64 #1 SMP Debian 4.9.82-1+deb9u3 (2018-03-02) x86_64 GNU/Linux Kernel logs said: [4769736.560752] XFS (sdc1): Metadata corruption detected at xfs_attr3_leaf_write_verify+0xe8/0x100 [xfs], xfs_attr3_leaf block 0xe7dd89b0 [4769736.563285] XFS (sdc1): Unmount and run xfs_repair [4769736.564554] XFS (sdc1): First 64 bytes of corrupted metadata buffer: [4769736.565818] ffff960ab1d0d000: 00 00 00 00 00 00 00 00 fb ee 00 00 00 00 00 00 ................ [4769736.567064] ffff960ab1d0d010: 10 00 00 00 00 20 0f e0 00 00 00 00 00 00 00 00 ..... .......... [4769736.568272] ffff960ab1d0d020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [4769736.569446] ffff960ab1d0d030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [4769736.570611] XFS (sdc1): xfs_do_force_shutdown(0x8) called from line 1339 of file /build/linux-YDazDa/linux-4.9.82/fs/xfs/xfs_buf.c. Return address = 0xffffffffc06c1ada [4769736.573226] XFS (sdc1): Corruption of in-memory data detected. Shutting down filesystem [4769736.574419] XFS (sdc1): Please umount the filesystem and rectify the problem(s) As per the message, I unmounted the filesystem and ran xfs_repair on it. The first run of xfs_repair told me to mount the filesystem to replay the log, which I did. I then unmounted it and ran xfs_repair again: ~$ sudo xfs_repair /dev/sdc1 Phase 1 - find and verify superblock... Phase 2 - using internal log - zero log... - scan filesystem freespace and inode maps... - found root inode chunk Phase 3 - for each AG... - scan and clear agi unlinked lists... - process known inodes and perform inode discovery... - agno = 0 - agno = 1 - agno = 2 - agno = 3 - process newly discovered inodes... Phase 4 - check for duplicate blocks... - setting up duplicate extent list... - check for inodes claiming duplicate blocks... - agno = 0 - agno = 1 - agno = 2 - agno = 3 Phase 5 - rebuild AG headers and trees... - reset superblock... Phase 6 - check inode connectivity... - resetting contents of realtime bitmap and summary inodes - traversing filesystem ... - traversal finished ... - moving disconnected inodes to lost+found ... Phase 7 - verify and correct link counts... done The filesystem now seems to be back up and running OK. I don't know if there's any more information I could provide to help track down this issue? Thanks, Chris -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1596550 Title: Metadata corruption detected at xfs_attr3_leaf_write_verify+0xd7/0xf0 Status in linux package in Ubuntu: Confirmed Bug description: We noticed a XFS metadata corruption once we ran a lot of small write IOs on SSDs in our OpenStack swift environment: [1468860.211158] XFS (sdax): Metadata corruption detected at xfs_attr3_leaf_write_verify+0xd7/0xf0 [xfs], block 0x7c99480 [1468860.211195] XFS (sdax): Unmount and run xfs_repair [1468860.211215] XFS (sdax): First 64 bytes of corrupted metadata buffer: [1468860.211247] ffff880630f66000: 00 00 00 00 00 00 00 00 fb ee 00 00 00 00 00 00 ................ [1468860.211268] ffff880630f66010: 10 00 00 00 00 20 0f e0 00 00 00 00 00 00 00 00 ..... .......... [1468860.211289] ffff880630f66020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [1468860.211309] ffff880630f66030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [1468860.211328] XFS (sdax): xfs_do_force_shutdown(0x8) called from line 1254 of file /build/linux-lts-xenial-7RlTta/linux-lts-xenial-4.4.0/fs/xfs/xfs_buf.c. Return address = 0x ffffffffc068f616 [1468860.212214] XFS (sdax): Corruption of in-memory data detected. Shutting down filesystem [1468860.212232] XFS (sdax): Please umount the filesystem and rectify the problem(s) [1468860.212323] XFS (sdax): xfs_do_force_shutdown(0x1) called from line 315 of file /build/linux-lts-xenial-7RlTta/linux-lts-xenial-4.4.0/fs/xfs/xfs_trans_buf.c. Return address = 0xffffffffc06bdda2 [1468860.261436] XFS (sdax): xfs_log_force: error -5 returned. This error is reported with linux-generic-lts-xenial @4.4.0.22.12 on a XFS filesystem formatted with 1024 as inode size and mounted with rw,noatime,nodiratime,attr2,nobarrier,inode64,logbufs=8,sunit=512,swidth=512,noquota For us this issue seems to be reproducible after several hours of stress testing. cat /proc/version_signature Ubuntu 4.4.0-22.40~14.04.1-generic 4.4.8 Description: Ubuntu 14.04.3 LTS Release: 14.04 --- AlsaDevices: total 0 crw-rw---- 1 root audio 116, 1 Jun 10 13:19 seq crw-rw---- 1 root audio 116, 33 Jun 10 13:19 timer AplayDevices: Error: [Errno 2] No such file or directory ApportVersion: 2.14.1-0ubuntu3.11 Architecture: amd64 ArecordDevices: Error: [Errno 2] No such file or directory AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1: CRDA: Error: [Errno 2] No such file or directory DistroRelease: Ubuntu 14.04 IwConfig: Error: [Errno 2] No such file or directory MachineType: HP ProLiant DL380 Gen9 Package: linux (not installed) PciMultimedia: ProcEnviron: TERM=screen PATH=(custom, no user) XDG_RUNTIME_DIR=<set> LANG=en_US.UTF-8 SHELL=/bin/bash ProcFB: ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-4.4.0-22-generic root=/dev/mapper/lxc-root00 ro biosdevname=1 net.ifnames=0 usbcore.autosuspend=-1 vga=normal nomodeset nomdmonddf nomdmonisw crashkernel=1024M-:128M ProcVersionSignature: Ubuntu 4.4.0-22.40~14.04.1-generic 4.4.8 RelatedPackageVersions: linux-restricted-modules-4.4.0-22-generic N/A linux-backports-modules-4.4.0-22-generic N/A linux-firmware 1.127.15 RfKill: Error: [Errno 2] No such file or directory Tags: trusty Uname: Linux 4.4.0-22-generic x86_64 UpgradeStatus: No upgrade log present (probably fresh install) UserGroups: _MarkForUpload: True dmi.bios.date: 07/20/2015 dmi.bios.vendor: HP dmi.bios.version: P89 dmi.chassis.type: 23 dmi.chassis.vendor: HP dmi.modalias: dmi:bvnHP:bvrP89:bd07/20/2015:svnHP:pnProLiantDL380Gen9:pvr:cvnHP:ct23:cvr: dmi.product.name: ProLiant DL380 Gen9 dmi.sys.vendor: HP To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1596550/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp