On Fri, May 25, 2018 at 02:16:39AM +0800, Zorro Lang wrote: > Hi, > > I just hit a XFS crash [1] when I tested XFS on a 500T storage. And a XFS > corruption [2] can be found by "xfs_repair -n". > > Tested on 4.17-rc6, xfsprogs-4.16.1. mkfs.xfs options are "-m crc=1,finobt=1, > rmapbt=1,reflink=1 -i sparse=1". > > I've report a bug as below: > https://bugzilla.kernel.org/show_bug.cgi?id=199825 > > [1] > kernel: XFS: Assertion failed: XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0, file: fs/xfs/xfs_super.c, line: 984 Maybe it's time I cleaned up & sent out my patch that dumps the data and cow forks to ftrace when we hit this assert. Also, I think i_dlayed_blks should be xfs_filblks_t, not unsigned int... --D > kernel: WARNING: CPU: 39 PID: 487 at fs/xfs/xfs_message.c:105 asswarn+0x1e/0x30 [xfs] > kernel: Modules linked in: loop intel_powerclamp coretemp kvm_intel sunrpc kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel pcbc aesni_intel crypto_simd cryptd glue_helper iTCO_wdt gpio_ich iTCO_vendor_support dm_service_time pcspkr ipmi_ssif ipmi_si ipmi_devintf hpilo hpwdt sg i7core_edac lpc_ich ipmi_msghandler shpchp acpi_power_meter pcc_cpufreq dm_multipath ip_tables xfs libcrc32c sr_mod cdrom ata_generic pata_acpi radeon i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm ata_piix sd_mod drm mpt3sas libata crc32c_intel serio_raw hpsa netxen_nic i2c_core raid_class scsi_transport_sas dm_mirror dm_region_hash dm_log dm_mod > kernel: CPU: 39 PID: 487 Comm: kswapd0 Not tainted 4.17.0-rc6+ #2 > kernel: Hardware name: HP ProLiant DL580 G7, BIOS P65 10/01/2013 > kernel: RIP: 0010:asswarn+0x1e/0x30 [xfs] > kernel: RSP: 0018:ffff9b9887d9fbc0 EFLAGS: 00010246 > kernel: RAX: 0000000000000000 RBX: ffff882b91e009d8 RCX: 0000000000000000 > kernel: RDX: ffff9b9887d9fae0 RSI: 000000000000000a RDI: ffffffffc05ecff9 > kernel: RBP: ffff882b91e008c8 R08: 0000000000000000 R09: 0000000000000021 > kernel: R10: 0000000000000000 R11: 000000000000000a R12: ffff882b91e00780 > kernel: R13: 000000000000ce58 R14: 000000000000000c R15: 00000000000000e5 > kernel: FS: 0000000000000000(0000) GS:ffff882f6fbc0000(0000) knlGS:0000000000000000 > kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > kernel: CR2: 00007f832588eff0 CR3: 0000000a2dc0a002 CR4: 00000000000206e0 > kernel: Call Trace: > kernel: xfs_fs_destroy_inode+0x1db/0x200 [xfs] > kernel: dispose_list+0x48/0x60 > kernel: prune_icache_sb+0x52/0x70 > kernel: super_cache_scan+0x139/0x190 > kernel: shrink_slab+0x249/0x430 > kernel: shrink_node+0x44a/0x490 > kernel: balance_pgdat+0x188/0x360 > kernel: kswapd+0x171/0x410 > kernel: ? remove_wait_queue+0x60/0x60 > kernel: kthread+0xf8/0x130 > kernel: ? balance_pgdat+0x360/0x360 > kernel: ? kthread_bind+0x10/0x10 > kernel: ret_from_fork+0x35/0x40 > kernel: Code: 99 f9 ff ff 0f 0b e8 c2 d4 4e c8 66 90 66 66 66 66 90 48 89 f1 41 89 d0 48 c7 c6 08 7f 5f c0 48 89 fa 31 c0 31 ff e8 b2 fc ff ff <0f> 0b c3 0f 1f 44 00 00 66 2e 0f 1f 84 00 00 00 00 00 66 66 66 > kernel: ---[ end trace 6d72efdc406c28bc ]--- > > [2] > > Phase 1 - find and verify superblock... > - reporting progress in intervals of 15 minutes > Memory available for repair (48300MB) may not be sufficient. > At least 64121MB is needed to repair this filesystem efficiently > If repair fails due to lack of memory, please > turn prefetching off (-P) to reduce the memory footprint. > Phase 2 - using internal log > - zero log... > - scan filesystem freespace and inode maps... > sb_fdblocks 2570207456, counted 2570207543 > - 10:12:40: scanning filesystem freespace - 501 of 501 allocation groups done > - found root inode chunk > Phase 3 - for each AG... > - scan (but don't clear) agi unlinked lists... > - 10:12:40: scanning agi unlinked lists - 501 of 501 allocation groups done > - process known inodes and perform inode discovery... > - agno = 0 > - agno = 15 > ... > (501 AGs) > ... > - agno = 254 > - 10:57:14: process known inodes and inode discovery - 21719296 of 19087328 inodes done > - process newly discovered inodes... > - 10:57:14: process newly discovered inodes - 501 of 501 allocation groups done > Phase 4 - check for duplicate blocks... > - setting up duplicate extent list... > - 10:57:17: setting up duplicate extent list - 501 of 501 allocation groups done > - check for inodes claiming duplicate blocks... > - agno = 30 > - agno = 0 > - agno = 45 > ... > (501 AGs) > ... > - agno = 253 > - agno = 254 > - 11:42:57: check for inodes claiming duplicate blocks - 21719296 of 19087328 inodes done > No modify flag set, skipping phase 5 > Phase 6 - check inode connectivity... > - traversing filesystem ... > - traversal finished ... > - moving disconnected inodes to lost+found ... > Phase 7 - verify link counts... > - 11:52:50: verify and correct link counts - 501 of 501 allocation groups done > No modify flag set, skipping filesystem flush and exiting. > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
