Lustre 2.14.0 Changelog: Difference between revisions

From Lustre Wiki
Jump to navigation Jump to search
Tag: Replaced
Line 14: Line 14:


== Issues fixed between 2.13.0 and 2.14.0 ==
== Issues fixed between 2.13.0 and 2.14.0 ==
''' New Feature '''
*[https://jira.whamcloud.com/browse/LU-1365 LU-1365]: Implement ldiskfs LARGEDIR support for e2fsprogs
Release Notes - Lustre - Version Lustre 2.14.0
''' Bugs '''
*[https://jira.whamcloud.com/browse/LU-1742] - Fix 'Timed out tx' error message
*[https://jira.whamcloud.com/browse/LU-4165] - Interop 2.4.1<->2.5 failure on test suite sanity-lfsck test_2c
*[https://jira.whamcloud.com/browse/LU-4671] - sanity test_43: multiop still running
*[https://jira.whamcloud.com/browse/LU-5338] - sanity-lfsck test_11b: (9) expect lastid1 0x100000000:569
*[https://jira.whamcloud.com/browse/LU-5855] - sanity-lfsck does not work under ZFS-based DNE mode
*[https://jira.whamcloud.com/browse/LU-6006] - replay-dual test_22a: Remote creation failed 1
*[https://jira.whamcloud.com/browse/LU-7653] - replay-single/test_110f test failed Lustre: DEBUG MARKER: replay-single test_110f: FAIL: 1 != 2 after recovery
*[https://jira.whamcloud.com/browse/LU-7791] - signal vs CP callback race
*[https://jira.whamcloud.com/browse/LU-8221] - Changes required to allow USE_LU_REF/lu_ref feature enabling and usage
*[https://jira.whamcloud.com/browse/LU-8304] - libcfs_debug_dumplog calls potentially sleeping function in wrong state
*[https://jira.whamcloud.com/browse/LU-8454] - non-root user is able to change stripe size for mount point
*[https://jira.whamcloud.com/browse/LU-8522] - sanity 76 fails on clients with CONFIG_SLAB unset (inode slab grew from 400 to 525)
*[https://jira.whamcloud.com/browse/LU-9215] - build: Re-add the lustre-devel package.
*[https://jira.whamcloud.com/browse/LU-9812] - Source rpm does not cleanly build with --rebuild
*[https://jira.whamcloud.com/browse/LU-10157] - LNET_MAX_IOV hard coded to 256
*[https://jira.whamcloud.com/browse/LU-10198] - GPF llog_osd_declare_write_rec+0xb6/0x3d0
*[https://jira.whamcloud.com/browse/LU-10395] - ASSERTION( osd->od_oi_table != NULL && osd->od_oi_count >= 1 ) failed
*[https://jira.whamcloud.com/browse/LU-10401] - sanity test_133g: timeout during MDT mount
*[https://jira.whamcloud.com/browse/LU-10589] - sanity-dom test_251: test_sanity failed with 1
*[https://jira.whamcloud.com/browse/LU-10728] - lnet/utils/lnetconfig/liblnetconfig.c uses sizeof on pointers
*[https://jira.whamcloud.com/browse/LU-10753] - sanity test 300c fails with 'create 5k files failed'
*[https://jira.whamcloud.com/browse/LU-10958] - brw rpc reordering causes data corruption when the writethrough cache is disabled
*[https://jira.whamcloud.com/browse/LU-11114] - Track creates in MDC and llite stats
*[https://jira.whamcloud.com/browse/LU-11152] - sanity test_133g: ost1 find  /proc/fs/lustre/ /proc/sys/lnet/ /proc/sys/lustre/ failed
*[https://jira.whamcloud.com/browse/LU-11185] - MGC lock leak
*[https://jira.whamcloud.com/browse/LU-11269] - ptlrpc_set_add_req()) ASSERTION( req->rq_import->imp_state != LUSTRE_IMP_IDLE ) failed
*[https://jira.whamcloud.com/browse/LU-11276] - racer: mdc_dev.c:1346:mdc_req_attr_set()) uncovered page
*[https://jira.whamcloud.com/browse/LU-11385] - client hit BUG: unable to handle kernel NULL pointer dereference at 0000000000000028
*[https://jira.whamcloud.com/browse/LU-11490] - run_rr_alloc() defect: FSNAME should be used instead of lustre
*[https://jira.whamcloud.com/browse/LU-11510] - preserve PFL/FLR/DoM layout with lfs_migrate
*[https://jira.whamcloud.com/browse/LU-11518] - lock_count is exceeding lru_size
*[https://jira.whamcloud.com/browse/LU-11546] - enable large_dir support for MDTs
*[https://jira.whamcloud.com/browse/LU-11548] - increase default readahead tunables to match common settings
*[https://jira.whamcloud.com/browse/LU-11656] - "lfs getstripe" on directory does not show default root layout
*[https://jira.whamcloud.com/browse/LU-11762] - replay-single test 0d fails with  'post-failover df failed'
*[https://jira.whamcloud.com/browse/LU-11803] - sanity test 255c fails with 'Ladvise test 12, bad lock count, returned 1, actual 0'
*[https://jira.whamcloud.com/browse/LU-11814] - conf-sanity test_93 osd_handler.c:7132:osd_device_init0()) ASSERTION( info ) failed:
*[https://jira.whamcloud.com/browse/LU-11832] - ARM servers crashing on MDS startup
*[https://jira.whamcloud.com/browse/LU-11848] - filefrag/FIEMAP doesn't work for PFL or FLR files
*[https://jira.whamcloud.com/browse/LU-11893] - doesn't handle logical network interface properly.
*[https://jira.whamcloud.com/browse/LU-11939] - ASSERTION( tgd->tgd_tot_granted >= ted->ted_grant ) on OSS
*[https://jira.whamcloud.com/browse/LU-11961] - nodemap_config_set_active()) ASSERTION( config->nmc_default_nodemap ) failed
*[https://jira.whamcloud.com/browse/LU-11986] - After partial lustre_rmmod, lnet panics on debugfs read
*[https://jira.whamcloud.com/browse/LU-12003] - Access to invalid semaphore in osd_trunc_unlock_all (ldiskfs)
*[https://jira.whamcloud.com/browse/LU-12100] - sanity-quota test_2: user create failure, but expect success
*[https://jira.whamcloud.com/browse/LU-12191] - lctl list_param -R does not display everything
*[https://jira.whamcloud.com/browse/LU-12193] - possible scheduling with spinlocks held in the quota paths
*[https://jira.whamcloud.com/browse/LU-12198] - lnetctl peer show hangs for ~2600 clients, ioctl getting E2BIG
*[https://jira.whamcloud.com/browse/LU-12205] - host_nids_address() fix for MR setup
*[https://jira.whamcloud.com/browse/LU-12207] - sanity test_76: inode slab grew from 69804 to 69816
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-12232] - replay-ost-single test 6 fails with ''space grew after dd: before:13442048 after_dd:13442048''
*[https://jira.whamcloud.com/browse/LU-12233] - Deadlock on LNet shutdown
*[https://jira.whamcloud.com/browse/LU-12287] - Unable to detect device faults from IB event queue
*[https://jira.whamcloud.com/browse/LU-12292] - Decrement Health Value even if recovery processing fault
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-12295] - MDS Panic on DNE2 directory removing
*[https://jira.whamcloud.com/browse/LU-12321] - Unlink speed needs to be improved in DoM configuration
*[https://jira.whamcloud.com/browse/LU-12370] - Please fix grammar
*[https://jira.whamcloud.com/browse/LU-12378] - sanity-quota test 1 fails with 'project quota isn't released after deletion'
*[https://jira.whamcloud.com/browse/LU-12390] - customise mdsrate tests
*[https://jira.whamcloud.com/browse/LU-12397] - osp: race around opd_new_connection
*[https://jira.whamcloud.com/browse/LU-12460] - lli_trunc_sem can lead to a readlock
*[https://jira.whamcloud.com/browse/LU-12469] - sanity test_230b crash: LBUG of lu_buf_free in mdd_iterate_xattrs
*[https://jira.whamcloud.com/browse/LU-12470] - sanityn test_47b: create isn't blocked
*[https://jira.whamcloud.com/browse/LU-12488] - strange breakage in sanityn test 93
*[https://jira.whamcloud.com/browse/LU-12503] - LustreError: 19435:0:(vvp_io.c:1056:vvp_io_write_start()) LBUG
*[https://jira.whamcloud.com/browse/LU-12506] - Client unable to mount filesystem with very large number of MDTs
*[https://jira.whamcloud.com/browse/LU-12530] - udev add/change rule loads zfs module on clients
*[https://jira.whamcloud.com/browse/LU-12549] - Lustre project PID 32-bit overflow
*[https://jira.whamcloud.com/browse/LU-12564] - ptlrpcd daemon sleeps while holding imp_lock spinlock
*[https://jira.whamcloud.com/browse/LU-12580] - usercopy exposure attempt detected in LL_IOC_LOV_GETSTRIPE ioctl
*[https://jira.whamcloud.com/browse/LU-12586] - truncate to start of uninstantiated PFL component silently fails
*[https://jira.whamcloud.com/browse/LU-12593] - update_log corruption
*[https://jira.whamcloud.com/browse/LU-12598] - osd_ios_lf_fill() returns 0 on some error paths.
*[https://jira.whamcloud.com/browse/LU-12632] - sanity-hsm test_90: FAIL: requests did not complete
*[https://jira.whamcloud.com/browse/LU-12637] - Support RHEL 8.1
*[https://jira.whamcloud.com/browse/LU-12642] - bad sequence output on OST
*[https://jira.whamcloud.com/browse/LU-12644] - correct fast read & strided readahead interaction
*[https://jira.whamcloud.com/browse/LU-12646] - Restore performance enhancement from  LU-11413
*[https://jira.whamcloud.com/browse/LU-12651] - High kworker CPU usage (osc_grant_work_handler) on IDLE connections
*[https://jira.whamcloud.com/browse/LU-12652] - osd_attr_set() fails with ENOSPC on linux-4.18
*[https://jira.whamcloud.com/browse/LU-12670] - osd_oxc_get(): Shrink rcu critical section to not include EXIT
*[https://jira.whamcloud.com/browse/LU-12677] - FIxing conf-sanity.sh test_21d (zfs)
*[https://jira.whamcloud.com/browse/LU-12679] - large-lun test_2 should check / warn if OSTSIZE is too small (zfs)
*[https://jira.whamcloud.com/browse/LU-12687] - Fast ENOSPC on direct I/O
*[https://jira.whamcloud.com/browse/LU-12702] - sanity-quota test_4a: create after timer expired, but expect EDQUOT
*[https://jira.whamcloud.com/browse/LU-12704] - racer test_1: Invalid layout: The component end must be aligned by the stripe size
*[https://jira.whamcloud.com/browse/LU-12718] - Readahead can trigger an assertion failire in cl_io_read_ahead
*[https://jira.whamcloud.com/browse/LU-12726] - sanity-sec "error: set_param: setting /sys/fs/lustre/mdt/lustre-MDT0000/identity_upcall=NONE: wrote only 1"
*[https://jira.whamcloud.com/browse/LU-12730] - sanity test_807: f807.sanity expected blocks: 4103, got: 2051
*[https://jira.whamcloud.com/browse/LU-12741] - crash in osd_object_delete at end of sanity
*[https://jira.whamcloud.com/browse/LU-12747] - sanity: test 811 fail with "MDD orphan cleanup thread not quit"
*[https://jira.whamcloud.com/browse/LU-12748] - parallel readahead needs to be optimized at high number of process
*[https://jira.whamcloud.com/browse/LU-12757] - sanity-lfsck test 36a fails with '(N) Fail to resync /mnt/lustre/d36a.sanity-lfsck/f2'
*[https://jira.whamcloud.com/browse/[LU-12759] - parameter grant_shrink gets reset to 1 after client reconnects
*[https://jira.whamcloud.com/browse/LU-12761] - sanity test 103a syntax error using version_code
*[https://jira.whamcloud.com/browse/LU-12768] - readcache_max_filesize in 2.12?
*[https://jira.whamcloud.com/browse/LU-12770] - Sanity quota test fixes for zfs
*[https://jira.whamcloud.com/browse/LU-12775] - conf-sanity test 32c fails with ‘mv remote dir failed’
*[https://jira.whamcloud.com/browse/LU-12778] - sanity-sec test 12 fails with 'nodemap_add failed with 1'
*[https://jira.whamcloud.com/browse/LU-12781] - sanity test_272a crashes with SSK
*[https://jira.whamcloud.com/browse/LU-12787] - sanity-quota: test 3 failure "write success, but expect EDQUOT"
*[https://jira.whamcloud.com/browse/LU-12791] - kernel update [RHEL8.0 4.18.0-80.11.2.el8_0]
*[https://jira.whamcloud.com/browse/LU-12799] - NULL pointer dereference IP: ptlrpc_disconnect_and_idle_import
*[https://jira.whamcloud.com/browse/LU-12809] - recovery-small/110k gets stuck at client umount
*[https://jira.whamcloud.com/browse/LU-12811] - lustre_swab_fiemap needs to check fm_mapped_extents
*[https://jira.whamcloud.com/browse/LU-12816] - LBUG: (niobuf.c:350:ptlrpc_register_bulk()) ASSERTION( !(desc->bd_registered && req->rq_send_state != LUSTRE_IMP_REPLAY) || mbits != desc->bd_last_mbits )
*[https://jira.whamcloud.com/browse/LU-12822] - gcc9 support
*[https://jira.whamcloud.com/browse/LU-12823] - basename in test-framework to support capital alphabets
*[https://jira.whamcloud.com/browse/LU-12826] - Project quotas: users can change project IDs
*[https://jira.whamcloud.com/browse/LU-12828] - FLOCK request can be processed twice during resend
*[https://jira.whamcloud.com/browse/LU-12833] - lustre_lwp_add_conn can't find lwp device
*[https://jira.whamcloud.com/browse/LU-12846] - could not delete orphan [0x2001ac5ff:0x3a30:0x0]: rc = -2
*[https://jira.whamcloud.com/browse/LU-12852] - growing a PFL file with last stripe as -1 fails
*[https://jira.whamcloud.com/browse/LU-12853] - general protection fault: 0000 RIP: keys_fill
*[https://jira.whamcloud.com/browse/LU-12854] - Allow "true/false" or "on/off" for audit_mode on nodemap
*[https://jira.whamcloud.com/browse/LU-12856] - LustreError: 82937:0:(ldlm_lib.c:3268:target_bulk_io()) @@@ truncated bulk READ 0(270336) 
*[https://jira.whamcloud.com/browse/LU-12865] - sanity test 160f fails with ‘mds1: User cl6 not registered’
*[https://jira.whamcloud.com/browse/LU-12870] - sanity-hsm test 9A fails with “uuid Dumping not found in agent list on mds1”
*[https://jira.whamcloud.com/browse/LU-12889] - Do not assume peers are MR capable
*[https://jira.whamcloud.com/browse/LU-12894] - SSK regression in 2.12.3
*[https://jira.whamcloud.com/browse/LU-12895] - review-dne-selinux sanity test_185a: lu_object_attr()) ASSERTION( ((o)->lo_header->loh_attr & LOHA_EXISTS) != 0 ) failed
*[https://jira.whamcloud.com/browse/LU-12898] - building lustre utils on ppc64le fails with error format expects llu but argument is __u64
*[https://jira.whamcloud.com/browse/LU-12901] - Failing to create a properly sized IB queue pair
*[https://jira.whamcloud.com/browse/LU-12911] - Setting a LOV EA can access or change outside allocated buffer
*[https://jira.whamcloud.com/browse/LU-12914] - (mdt_open.c:312:mdt_prep_ma_buf_from_rep()) ASSERTION( ma->ma_lmv == ((void *)0) && ma->ma_lmm == ((void *)0) ) failed
*[https://jira.whamcloud.com/browse/LU-12919] - Fix source specified remote destination send case
*[https://jira.whamcloud.com/browse/LU-12920] - build: lbuild script should not continue if it's missing ed
*[https://jira.whamcloud.com/browse/LU-12928] - recovery-small test_136: crash in sec2target_str() with review-dne-selinux-ssk
*[https://jira.whamcloud.com/browse/LU-12937] - make newwiretest adds strings, not connected to changes
*[https://jira.whamcloud.com/browse/LU-12944] - Migration corrupts xattrs
*[https://jira.whamcloud.com/browse/LU-12945] - Avoid sending PageSlab pages through tcp stack
*[https://jira.whamcloud.com/browse/LU-12946] - Multipath path flapping issue
*[https://jira.whamcloud.com/browse/LU-12949] - extend_recovery_timer assertion
*[https://jira.whamcloud.com/browse/LU-12951] - LL_IOC_GETOBDCOUNT return wrong MDT counts
*[https://jira.whamcloud.com/browse/LU-12955] - AST replies are dropped when servers are non-MR, clients and routers are MR
*[https://jira.whamcloud.com/browse/LU-12956] - oops in target_handle_connect()
*[https://jira.whamcloud.com/browse/LU-12960] - create striped directory on specific MDTs not working
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-12965] - class_import_put()) ASSERTION( _v >= 0 && _v  < 0x5a5a5a5a ) failed: value: 1515870810
*[https://jira.whamcloud.com/browse/LU-12967] - sanity test 80 silently fails to get sync_on_lock_cancel parameter
*[https://jira.whamcloud.com/browse/LU-12968] - KASAN: slab-out-of-bounds in mgs_wlp_lcfg+0xb3/0x4a0 [mgs]
*[https://jira.whamcloud.com/browse/LU-12969] - make sure we reset quota limits for all test ID/users
*[https://jira.whamcloud.com/browse/LU-12974] - Running lfsck with dry-run (-n) should not move orphan obhect to lost+found
*[https://jira.whamcloud.com/browse/LU-12981] - Make a maximal transfer size check accurately.
*[https://jira.whamcloud.com/browse/LU-12992] - sanity test_240: failed to mount client
*[https://jira.whamcloud.com/browse/LU-12999] - gcc8 compile warning on string overflow in name_create_mdt
*[https://jira.whamcloud.com/browse/LU-13001] - check_routers_before_use causes LNet to hang indefinitely if any router is down
*[https://jira.whamcloud.com/browse/LU-13002] - sanity-lnet stops all test suites from running
*[https://jira.whamcloud.com/browse/LU-13017] - sanity test 56ra fails with ' '/usr/bin/lfs find -size 0 -type f /mnt/lustre/d56ra.sanity' should send 12 glimpse RPCs to OST'
*[https://jira.whamcloud.com/browse/LU-13019] - sanity test_60g: usleep: command not found
*[https://jira.whamcloud.com/browse/LU-13025] - LNet Health: Peer net health not considered
*[https://jira.whamcloud.com/browse/LU-13026] - LNet Routing: gateway NI not selected properly
*[https://jira.whamcloud.com/browse/LU-13027] - LNet Routing: route consolidation
*[https://jira.whamcloud.com/browse/LU-13028] - LNet Discovery: toggling discovery on/off is not handled properly
*[https://jira.whamcloud.com/browse/LU-13029] - LNet Routing: asym routing is not working for multi-hop routing
*[https://jira.whamcloud.com/browse/LU-13035] - LNet MR: fix remote peer NI selection
*[https://jira.whamcloud.com/browse/LU-13039] - BUG: KASAN: stack-out-of-bounds in strchr+0x2d/0x50
*[https://jira.whamcloud.com/browse/LU-13040] - BUG: KASAN: slab-out-of-bounds in string_nocheck+0xd1/0x180
*[https://jira.whamcloud.com/browse/LU-13041] - LNet: bug in peer deletion
*[https://jira.whamcloud.com/browse/LU-13042] - sanity-selinux test_21b Error: 'touch (5)'
*[https://jira.whamcloud.com/browse/LU-13043] - Project quotas: force to ignore quota flags =8 on quota exceeded
*[https://jira.whamcloud.com/browse/LU-13049] - LNet: Handle shutdown properly
*[https://jira.whamcloud.com/browse/LU-13053] - conf-sanity test 87 call to umount_ldiskfs() fails
*[https://jira.whamcloud.com/browse/LU-13054] - MDS kernel BUG at ldiskfs/htree_lock.c:429!
*[https://jira.whamcloud.com/browse/LU-13057] - sanity-pcc test_1c: Failed to start copytool ost8
*[https://jira.whamcloud.com/browse/LU-13061] - osd_fid_lookup()) ASSERTION( fid_is_sane(fid) || fid_is_idif(fid) ) failed: [0x0:0x68:0x0]
*[https://jira.whamcloud.com/browse/LU-13064] - Check permissions when accessing changelogs
*[https://jira.whamcloud.com/browse/LU-13065] - runtests test 1 hangs when unmounting an OST
*[https://jira.whamcloud.com/browse/LU-13069] - osp_sync_thread()) ASSERTION( thread->t_flags != SVC_RUNNING ) failed: 308 changes, 111 in progress, 0 in flight
*[https://jira.whamcloud.com/browse/LU-13070] - mdd_orphan_destroy loop caused by compatibility issue on upgrades to 2.11 or later
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-13071] - LNet Health: reduce log severity
*[https://jira.whamcloud.com/browse/LU-13077] - Cleanup xattr checking
*[https://jira.whamcloud.com/browse/LU-13082] - Enable lgss_keyring debug traces
*[https://jira.whamcloud.com/browse/LU-13087] - The server highest transno can be lost, client reports 'went back in time'
*[https://jira.whamcloud.com/browse/LU-13088] - sleeping function in target_recovery_overseer
*[https://jira.whamcloud.com/browse/LU-13090] - fix lfs_migrate -p for file with an existing pool
*[https://jira.whamcloud.com/browse/LU-13092] - include lbuild-{fc,rhel,sles} to SIGNATURE
*[https://jira.whamcloud.com/browse/LU-13093] - osd_trans_exec_check()) lustre-OST0001: opcode 0: used 8, used now 8, reserved 1 LBUG
*[https://jira.whamcloud.com/browse/LU-13096] - incorrect used_mb in max_cached_mb
*[https://jira.whamcloud.com/browse/LU-13097] - sanityn test_47b: create must fail
*[https://jira.whamcloud.com/browse/LU-13099] - ll_set_inode()) Can not initialize inode warning
*[https://jira.whamcloud.com/browse/LU-13101] - Eviction during ll_open_cleanup()
*[https://jira.whamcloud.com/browse/LU-13102] - llog_cat_id2handle())  error opening log id  rc = -2 osp_sync_thread()) ASSERTION( thread->t_flags != SVC_RUNNING )
*[https://jira.whamcloud.com/browse/LU-13115] - always check object existence after mo_xattr_get()
*[https://jira.whamcloud.com/browse/LU-13116] - sanity-sec test_30b: touch skn
*[https://jira.whamcloud.com/browse/LU-13117] - Wrong job id parsing in cfs_get_environ()
*[https://jira.whamcloud.com/browse/LU-13119] - lustre-initialization crashed in common_file_perm() on SLES12
*[https://jira.whamcloud.com/browse/LU-13121] - ll_update_lsm_md() may deadlock
*[https://jira.whamcloud.com/browse/LU-13128] - a race between glimpse and lock cancel is not handled correctly
*[https://jira.whamcloud.com/browse/LU-13131] - Partial writes on multi-client strided files
*[https://jira.whamcloud.com/browse/LU-13133] - sanity-selinux test_21a and test_21b: sepol update in case of combined MGS/MDS
*[https://jira.whamcloud.com/browse/LU-13136] - (layout.c:2121:__req_capsule_get()) @@@ Wrong buffer for field 'niobuf_inline' (7 of 7) in format 'LDLM_INTENT_OPEN', 0 vs. 0 (server)
*[https://jira.whamcloud.com/browse/LU-13137] - User process segfaults since 2.13 client upgrade
*[https://jira.whamcloud.com/browse/LU-13142] - Cleanup layout checking
*[https://jira.whamcloud.com/browse/LU-13145] - LNet Health: increase transaction timeout
*[https://jira.whamcloud.com/browse/LU-13147] - sanity-lnet does not cleanup after itself on test failure
*[https://jira.whamcloud.com/browse/LU-13152] - llapi layout llc_pattern is not set properly on DoM component
*[https://jira.whamcloud.com/browse/LU-13153] - Function llapi_get_version_string fails
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-13154] - Interop: sanity-quota test 66 fails with 'nonroot user should fail to set projid'
*[https://jira.whamcloud.com/browse/LU-13156] - sanity-selinux test_21a test_21b: c0.sepol not updated on mds1 after 20 secs
*[https://jira.whamcloud.com/browse/LU-13157] - migrate symlink with target name length > 59 cause crash
*[https://jira.whamcloud.com/browse/LU-13159] - osd-zfs can leak index backup entries
*[https://jira.whamcloud.com/browse/LU-13160] - sanity-hsm test 70 timeout
*[https://jira.whamcloud.com/browse/LU-13163] - sanity test_65i hung: RIP: 0010:mdc_read_page+0x14f/0x9b0 [mdc]
*[https://jira.whamcloud.com/browse/LU-13165] - MSG_RESENT can be improperly cleared.
*[https://jira.whamcloud.com/browse/LU-13166] - sanity test_812b: FAIL: lfs quota should succeed
*[https://jira.whamcloud.com/browse/LU-13168] - Client panic "Freechain corrupt"/"Redzone Overwritten"
*[https://jira.whamcloud.com/browse/LU-13170] - sanity-lfsck/23b fails often with ZFS
*[https://jira.whamcloud.com/browse/LU-13174] - sanity-quota convert minutes to seconds to sleep
*[https://jira.whamcloud.com/browse/LU-13181] - kiblnd_fmr_pool_map error on the AARCH64 with 64k pages
*[https://jira.whamcloud.com/browse/LU-13182] - MAP_POPULATE hangs with Linux 5.4
*[https://jira.whamcloud.com/browse/LU-13183] - Linux 5.4 ldiskfs build (drop: ext4-remove-truncate-warning.patch)
*[https://jira.whamcloud.com/browse/LU-13184] - conf-sanity test_112: problem creating f112.conf-sanity.0 on OST0000
*[https://jira.whamcloud.com/browse/LU-13187] - sanity test_129: current dir size 4096,  previous limit 20480
*[https://jira.whamcloud.com/browse/LU-13190] - I/O can fail with -EBADF if fresh client has no connection to OST
*[https://jira.whamcloud.com/browse/LU-13191] - osp_sync_interpret() LBUG upon OST disk failure
*[https://jira.whamcloud.com/browse/LU-13194] - Interop: sanityn test 104 fails with '/mnt/lustre/f104.sanityn mtime (1580286855:1580286853) diff'
*[https://jira.whamcloud.com/browse/LU-13196] - Remove inode mutex for DIO reads
*[https://jira.whamcloud.com/browse/LU-13209] - SUSE 15 SP2 client build fix
*[https://jira.whamcloud.com/browse/LU-13210] - Build failure with gcc8 -Werror=implicit-fallthrough=
*[https://jira.whamcloud.com/browse/LU-13211] - ldiskfs for 5.4.x post 5.4.6
*[https://jira.whamcloud.com/browse/LU-13216] - Kernel NULL pointer dereference in lustre_msg_set_conn_cnt()
*[https://jira.whamcloud.com/browse/LU-13219] - parallel-scale-nfsv3: FAIL: setup nfs failed on RHEL 8.1
*[https://jira.whamcloud.com/browse/LU-13225] - add bash completion for lfs
*[https://jira.whamcloud.com/browse/LU-13227] - sanityn 16a  FAIL: fsx with O_DIRECT failed.
*[https://jira.whamcloud.com/browse/LU-13228] - write access to an mmapped file over soft quota limit causes sigbus
*[https://jira.whamcloud.com/browse/LU-13229] - Flock unlock request resend caused a memory leak.
*[https://jira.whamcloud.com/browse/LU-13232] - sanity test 160j fails with 'read changelog failed'
*[https://jira.whamcloud.com/browse/LU-13234] - ldiskfs/namei.c:3310 ldiskfs_orphan_add+0x11e/0x2a0 [ldiskfs]
*[https://jira.whamcloud.com/browse/LU-13235] - Fix wrong size in lnet_net_append_cpts
*[https://jira.whamcloud.com/browse/LU-13246] - osd_otable_it_next() can miss spin_unlock()
*[https://jira.whamcloud.com/browse/LU-13252] - Handle possible failure from name_create_mdt
*[https://jira.whamcloud.com/browse/LU-13253] - sanity/177 leaves threads blocked in libcfs_debug_dumplog()
*[https://jira.whamcloud.com/browse/LU-13254] - crash at lu_object_find() in mdt_lvbo_fill()
*[https://jira.whamcloud.com/browse/LU-13260] - Fix size checking issue in lov
*[https://jira.whamcloud.com/browse/LU-13261] -  Lustre : lfs1-MDT0000: small buffer size
*[https://jira.whamcloud.com/browse/LU-13262] - "(lu_ref.c:257:lu_ref_del()) ASSERTION( 0 ) failed" triggered by jbd2/ldlm tasks, with recent master configured with USE_LU_REF defined
*[https://jira.whamcloud.com/browse/LU-13263] - "(lu_ref.c:257:lu_ref_del()) ASSERTION( 0 ) failed" triggered by lu_ref_del() in osc_ldlm_glimpse_ast(), because no corresponding lu_ref_link posted, with recent master configured with USE_LU_REF defined
*[https://jira.whamcloud.com/browse/LU-13264] - Multiple "(cl_object.c:962:cl_env_percpu_put()) ASSERTION( cle == &cl_env_percpu[cpu] ) failed" cases triggered by cl_env_percpu_put() in osc_lock_upcall() context, with recent master configured with USE_LU_REF defined
*[https://jira.whamcloud.com/browse/LU-13265] - ln_push_target can be overwritten leading to incorrect lnet_peer objects.
*[https://jira.whamcloud.com/browse/LU-13267] - improve racer cleanup
*[https://jira.whamcloud.com/browse/LU-13270] - dom-performance 4 defects
*[https://jira.whamcloud.com/browse/LU-13271] - skip conf-sanity.sh 87, 115 tests if FILSET set
*[https://jira.whamcloud.com/browse/LU-13273] - run_ior() parameters order fix
*[https://jira.whamcloud.com/browse/LU-13276] - Potential infinite loop in push target resize code
*[https://jira.whamcloud.com/browse/LU-13277] - Potential deadlock in lnet_peer_discovery
*[https://jira.whamcloud.com/browse/LU-13278] - Asymmetry in discovery push vs reply handling
*[https://jira.whamcloud.com/browse/LU-13279] - sanity.sh:test_20() does not take DoM component into account
*[https://jira.whamcloud.com/browse/LU-13282] - ha.sh 1a7c352e97 fix: wait $LFSCK_BG_PID properly
*[https://jira.whamcloud.com/browse/LU-13288] - a performance regression on the single stream write
*[https://jira.whamcloud.com/browse/LU-13291] - mballoc should not skip uninitialized-on-disk groups
*[https://jira.whamcloud.com/browse/LU-13293] - Readahead doesn't work well for non-stride SSF
*[https://jira.whamcloud.com/browse/LU-13294] - wrong cpt malloc rotor handling leads to oops
*[https://jira.whamcloud.com/browse/LU-13296] - statfs isn't work properly with MDT statfs proxy
*[https://jira.whamcloud.com/browse/LU-13301] - lfs setstripe to deactivated OST does not return
*[https://jira.whamcloud.com/browse/LU-13314] - sanity test_56ob: lfs find /mnt/lustre/d56ob.sanity -mtime 1y' wrong: found 0, expected 1
*[https://jira.whamcloud.com/browse/LU-13324] - NULL pointer dereference at vvp_page_delete+0x14/0x140
*[https://jira.whamcloud.com/browse/LU-13345] - kernel update [SLES12 SP4 4.12.14-95.48.1]
*[https://jira.whamcloud.com/browse/LU-13349] - Statahead should skip the entry while fid change
*[https://jira.whamcloud.com/browse/LU-13352] - Test duration not reset when using ONLY_REPEAT test parameter
*[https://jira.whamcloud.com/browse/LU-13354] - update wirecheck for missed values
*[https://jira.whamcloud.com/browse/LU-13355] - adler32 wrapper in libcfs
*[https://jira.whamcloud.com/browse/LU-13356] - lctl conf_param hung on the MGS node
*[https://jira.whamcloud.com/browse/LU-13357] - striped directory doesn't support lookup
*[https://jira.whamcloud.com/browse/LU-13362] - Race between message finalize and discovery reply processing
*[https://jira.whamcloud.com/browse/LU-13365] - spin_lock in after_reply() eat up most of cpu
*[https://jira.whamcloud.com/browse/LU-13368] - lnet may be trying to use deleted routes leading to errors kiblnd_rejected(): 10.0.10.212@o2ib7 rejected: consumer defined fatal error
*[https://jira.whamcloud.com/browse/LU-13373] - Wrong verify flags for writing in ll_file_aio_write
*[https://jira.whamcloud.com/browse/LU-13375] - lfs mirror extend does not take file layout into account
*[https://jira.whamcloud.com/browse/LU-13377] - potential dead loops on short writing
*[https://jira.whamcloud.com/browse/LU-13379] - interop b2_12/master: lustre-initialization failed: Unrecognized mount option "localrecov"
*[https://jira.whamcloud.com/browse/U-13386] - allow over reservation to cover current read
*[https://jira.whamcloud.com/browse/LU-13389] - aborted file create may leave unattached inodes on MDS.
*[https://jira.whamcloud.com/browse/LU-13399] - too many namespace LFSCK repair the shard messages
*[https://jira.whamcloud.com/browse/LU-13402] - sanity test_252: Invalid number of mdtlov clients returned by /usr/sbin/lr_reader
*[https://jira.whamcloud.com/browse/LU-13404] - ‘lfs mirror read’ does not write to output file
*[https://jira.whamcloud.com/browse/LU-13408] - tgt_cancel_slc_locks()) ASSERTION( lock->l_client_cookie != 0 )
*[https://jira.whamcloud.com/browse/LU-13411] - processing of update logs couldn't delete empty plain llogs
*[https://jira.whamcloud.com/browse/LU-13416] - Data corruption during IOR testing with DoM files and hard failover
*[https://jira.whamcloud.com/browse/LU-13418] - Some minor typos crept into gnilnd
*[https://jira.whamcloud.com/browse/LU-13423] - Fix cleanup_netns in sanity-lnet.sh
*[https://jira.whamcloud.com/browse/LU-13424] - unable to migrate mirrored files
*[https://jira.whamcloud.com/browse/LU-13425] - "run 'lfs migrate -m 1 -c 1 -H 3 dir1' to finish migration" is broken
*[https://jira.whamcloud.com/browse/LU-13426] - "lfs migrate" on DoM component clobbers LOV EA FID
*[https://jira.whamcloud.com/browse/LU-13427] - "lfs df -v" prints too many MDTs
*[https://jira.whamcloud.com/browse/LU-13429] - client login node crash: kernel:LustreError: 190:0:(lov_io.c:121:lov_io_sub_init()) ASSERTION(is_index_within_mirror(lov, index, lio->lis_mirror_index) ) failed: [0x20004f178:0x2f6c:0x0]iot  = 6, index = 0, mirror = 0
*[https://jira.whamcloud.com/browse/LU-13430] - Rework __vfs_getxattr/__vfs_setxattr/__vfs_removexattr support
*[https://jira.whamcloud.com/browse/LU-13431] - Rework preferred IO size
*[https://jira.whamcloud.com/browse/LU-13432] - test-frameworkto preserve MODOPTS_LIBCFS
*[https://jira.whamcloud.com/browse/LU-13433] - lfs mirror delete "usage" and manpage do not match
*[https://jira.whamcloud.com/browse/LU-13437] - rename may miss revoking LOOKUP lock to cause stale dentry on client
*[https://jira.whamcloud.com/browse/LU-13441] - ASSERTION( i < 1000 ) failed
*[https://jira.whamcloud.com/browse/LU-13449] - recovery-small test_140b: FAIL: recovery took too long 4362 20
*[https://jira.whamcloud.com/browse/LU-13454] - NULL dereference in lnet_health_check lnet_incr_hstats
*[https://jira.whamcloud.com/browse/LU-13455] - HSM client disconnected and fail to reconnect to server
*[https://jira.whamcloud.com/browse/LU-13460] - use correct obd index in print_failed_tgt()
*[https://jira.whamcloud.com/browse/LU-13461] - LNet routing: wrong gw ni may be selected to reach undiscovered peer
*[https://jira.whamcloud.com/browse/LU-13464] - MDT0000 remount in recovery 40 hours
*[https://jira.whamcloud.com/browse/LU-13465] - Interop 2.12 <-> master sanity test_27A: lfs setstripe: setstripe error for '/mnt/lustre': Unknown error 524
*[https://jira.whamcloud.com/browse/LU-13466] - BUG: unable to handle kernel NULL pointer dereference in class_exp2cliimp
*[https://jira.whamcloud.com/browse/LU-13467] -  truncate deadlock with DoM files
*[https://jira.whamcloud.com/browse/LU-13470] - sysfs ping write creates a flood-ping situation that could not be normally stopped
*[https://jira.whamcloud.com/browse/LU-13471] - LNet: discovery can glitch
*[https://jira.whamcloud.com/browse/LU-13472] - lnet: routing aliveness might be stale
*[https://jira.whamcloud.com/browse/LU-13473] - page discard of non-composite file does not set mirror info
*[https://jira.whamcloud.com/browse/LU-13474] - Lustre failover fails when SRPC enabled
*[https://jira.whamcloud.com/browse/LU-13476] - vvp_set_pagevec_dirty / vvp_page_completion_write lock ordering appears to trigger RCU stalls
*[https://jira.whamcloud.com/browse/LU-13477] - LNet: Force full discovery
*[https://jira.whamcloud.com/browse/LU-13478] - LNet: peer update adjustment on discovery toggle
*[https://jira.whamcloud.com/browse/LU-13481] - sanity test_33h: MDT index mismatch 5 times
*[https://jira.whamcloud.com/browse/LU-13486] - tests sanity/32* fail
*[https://jira.whamcloud.com/browse/LU-13490] - readahead thread breaks read stats in jobstats
*[https://jira.whamcloud.com/browse/LU-13493] - vvp_set_pagevec_dirty()) ASSERTION( page->mapping ) failed: mapping must be set
*[https://jira.whamcloud.com/browse/LU-13498] - sanity test 56w fails with '/usr/bin/lfs_migrate -y -c 7 /mnt/lustre/d56w.sanity/file1 failed '
*[https://jira.whamcloud.com/browse/LU-13499] - client UUID is truncated
*[https://jira.whamcloud.com/browse/LU-13508] - crash in sanity test 160j
*[https://jira.whamcloud.com/browse/LU-13511] - MDS 2.12.4 ASSERTION( top->loh_hash.next == ((void *)0) && top->loh_hash.pprev == ((void *)0) ) failed
*[https://jira.whamcloud.com/browse/LU-13519] - expand inode if possible for project quota
*[https://jira.whamcloud.com/browse/LU-13520] - a performance regression on directory creation
*[https://jira.whamcloud.com/browse/LU-13522] - mdt.mdt_mds_mds_conns is not counted upon MDT stop and start
*[https://jira.whamcloud.com/browse/LU-13525] - struct sepol_downcall_data is badly formed
*[https://jira.whamcloud.com/browse/LU-13528] - sanity test_69: read succeeded, expect -ENOENT
*[https://jira.whamcloud.com/browse/LU-13535] - Files truncated/corruption due to lfsck
*[https://jira.whamcloud.com/browse/LU-13539] - Some LASSERTF are missing terminating newlines
*[https://jira.whamcloud.com/browse/LU-13541] - Application hang and kernel "divide error" in ll_readpage
*[https://jira.whamcloud.com/browse/LU-13543] - hsm.actions file is broken on RHEL 8.2
*[https://jira.whamcloud.com/browse/LU-13549] - 'make rpms' fails on centos8.1 client
*[https://jira.whamcloud.com/browse/LU-13553] - o2iblnd: convert LASSERT (tx->tx_sending > 0); to proper error handling
*[https://jira.whamcloud.com/browse/LU-13555] - Do not treat repo ml kernels as 'special' just map them to RHEL
*[https://jira.whamcloud.com/browse/LU-13557] - LU-11023 (quota pools) breaks building with gcc9
*[https://jira.whamcloud.com/browse/U-13559] - lfs mirror delete wrong error message
*[https://jira.whamcloud.com/browse/LU-13566] - socklnd: wrong NID to interface mapping
*[https://jira.whamcloud.com/browse/LU-13578] - sanity test_39r: atime on client != ost
*[https://jira.whamcloud.com/browse/LU-13580] - sanity-selinux test_5: sec context seen from 1st mount point is not correct
*[https://jira.whamcloud.com/browse/LU-13581] - LU-9859 libcfs: add support for Xarray breaks build on 3.10.0-1062.9.1.el7.x86_64.debug
*[https://jira.whamcloud.com/browse/LU-13593] - Fix growing message buffer
*[https://jira.whamcloud.com/browse/U-13599] - LustreError: 30166:0:(service.c:189:ptlrpc_save_lock()) ASSERTION( rs->rs_nlocks < 8 ) failed
*[https://jira.whamcloud.com/browse/LU-13605] - Intermediate routers modify the final destination NID
*[https://jira.whamcloud.com/browse/LU-13606] - Poor performance with routed clients to multi-rail servers when single server interface fails
*[https://jira.whamcloud.com/browse/LU-13608] - MDT stuck in WAITING, abort_recov stuck too
*[https://jira.whamcloud.com/browse/LU-13609] - lctl --device MGS llog_catlist doesn't list all config files.
*[https://jira.whamcloud.com/browse/LU-13614] - replay-single test_117: LBUG: ASSERTION( atomic_read(&obd->obd_req_replay_clients) == 0 ) failed
*[https://jira.whamcloud.com/browse/LU-13617] - Client dead lock leads to eviction from MDS (selinux is enabled)
*[https://jira.whamcloud.com/browse/LU-13628] - sanityn test_106a test_106c: supported stx_attributes: got ''?'', expected ''30''
*[https://jira.whamcloud.com/browse/LU-13635] - Add -D option back to lfs_migrate
*[https://jira.whamcloud.com/browse/LU-13645] - Various data corruptions possible in lustre.
*[https://jira.whamcloud.com/browse/LU-13648] - Route status can be set incorrectly via lnet_notify()
*[https://jira.whamcloud.com/browse/LU-13649] - osd_destroy()) ASSERTION( !lu_object_is_dying(dt->do_lu.lo_header) ) failed
*[https://jira.whamcloud.com/browse/LU-13653] - Ignore quota on slave MDTs during creating striped directory
*[https://jira.whamcloud.com/browse/LU-13662] - LNet: crash on bad user space parameters
*[https://jira.whamcloud.com/browse/LU-13666] - sanityn test_80b: ASSERTION( lo->ldo_dir_stripe_count > 0 ) failed
*[https://jira.whamcloud.com/browse/LU-13667] - ptlrpc_pinger_main is stuck in endless loop
*[https://jira.whamcloud.com/browse/LU-13673] - quiet compiler warnings on arm7l
*[https://jira.whamcloud.com/browse/LU-13675] - LNetError: 14769:0:(o2iblnd.h:1003:kiblnd_queue2str()) LBUG
*[https://jira.whamcloud.com/browse/LU-13677] - quota pools qunit sorting doesn't work
*[https://jira.whamcloud.com/browse/LU-13680] - large allocations in osd_bufs_get() failing
*[https://jira.whamcloud.com/browse/LU-13683] - lfs migrate on non-existing file returns exit code 0
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-13686] - lctl pool_add returns error randomly
*[https://jira.whamcloud.com/browse/LU-13687] - interop: "lfs find" prints spurious "No such file or directory" error messages
*[https://jira.whamcloud.com/browse/LU-13688] - Handle in-tree executed HSM copytools correctly
*[https://jira.whamcloud.com/browse/LU-13690] - lfs mirror ... /mnt/lustre/.lustre/fid/$FID causes CWARN() about invalid FID formats
*[https://jira.whamcloud.com/browse/LU-13692] - MDS slow/hung threads at mdt_object_local_lock
*[https://jira.whamcloud.com/browse/LU-13693] - lfs getstripe should avoid opening regular files
*[https://jira.whamcloud.com/browse/LU-13697] - short io for AIO
*[https://jira.whamcloud.com/browse/LU-13700] - sanity test_230o: dir split not finished
*[https://jira.whamcloud.com/browse/LU-13701] - sanity-lnet test_205: @@@@@@ FAIL: Should have failed
*[https://jira.whamcloud.com/browse/LU-13709] - 'lfs mkdir -i -1' may not initialize target MDTs correctly
*[https://jira.whamcloud.com/browse/LU-13711] - Typo in libssl dependency
*[https://jira.whamcloud.com/browse/LU-13712] - Flaw in MR Routing Algorithm
*[https://jira.whamcloud.com/browse/LU-13713] - LNet Router: bug in the predetermined gateway case
*[https://jira.whamcloud.com/browse/LU-13719] - lov tgt 36 not cleaned! deathrow=0, lovrc=1
*[https://jira.whamcloud.com/browse/LU-13721] - "lfs find --pool" does not work for PFL files
*[https://jira.whamcloud.com/browse/LU-13728] - lnetctl: inconsistent global show
*[https://jira.whamcloud.com/browse/LU-13729] - list_add corruption messages seen in a stack trace seen while trying to mount OST
*[https://jira.whamcloud.com/browse/LU-13731] - RHEL 8.2: VM_FAULT_RETRY is defined in mm_types.h instead of mm.h
*[https://jira.whamcloud.com/browse/LU-13732] - "lfs fid2path" can't match the rootpath correctly
*[https://jira.whamcloud.com/browse/LU-13733] - report client stats sumsq
*[https://jira.whamcloud.com/browse/LU-13736] - Do not set preferred NI for MR peer
*[https://jira.whamcloud.com/browse/LU-13742] - Provide an selinux_is_enabled for newer kernels
*[https://jira.whamcloud.com/browse/LU-13746] - lctl llog_print segfault if no log name supplied
*[https://jira.whamcloud.com/browse/LU-13750] - LNet lnetctl: peer add doesn't work as specified in the manual
*[https://jira.whamcloud.com/browse/LU-13751] - sanity test_160j: FAIL: read changelog failed
*[https://jira.whamcloud.com/browse/LU-13754] - GSS-based authentication fails on CentOS/RHEL 7.8
*[https://jira.whamcloud.com/browse/LU-13759] - sanity-dom sanityn_test_20 fails with '1 page left in cache after lock cancel'
*[https://jira.whamcloud.com/browse/LU-13763] - ptlrpc_invalidate_import()) lsrza-OST0000_UUID: Unregistering RPCs found (0). Network is sluggish? Waiting them to error out.
*[https://jira.whamcloud.com/browse/LU-13764] - lp_dc_error is not cleared when new discovery ping is sent as response to incoming push
*[https://jira.whamcloud.com/browse/LU-13765] - ldiskfs_mb_mark_diskspace_used:3472: aborting transaction: error 28 in __ldiskfs_handle_dirty_metadata
*[https://jira.whamcloud.com/browse/LU-13775] - Wrong memory copy in tgt_pages2shortio
*[https://jira.whamcloud.com/browse/LU-13776] - pjdfstest script should not write outside of /tmp
*[https://jira.whamcloud.com/browse/LU-13784] - llmount.sh cannot start Lustre with project quota enabled
*[https://jira.whamcloud.com/browse/LU-13787] - Build failure libcfs needed by snmp
*[https://jira.whamcloud.com/browse/LU-13790] - socklnd: NID to interface mapping issues
*[https://jira.whamcloud.com/browse/LU-13791] - Capabilities are not effective
*[https://jira.whamcloud.com/browse/LU-13809] - PFL file lost during recovery
*[https://jira.whamcloud.com/browse/LU-13818] - can not install lustre-client-utils on ubuntu 20.04
*[https://jira.whamcloud.com/browse/LU-13822] - ptlrpc_expired_set() needs to reschedule from time to time
*[https://jira.whamcloud.com/browse/LU-13824] - sanity test 230q fails with 'stripe hash crush,migrating != crush'
*[https://jira.whamcloud.com/browse/LU-13826] - fix compatibility for LL_IOC_MDC_GETINFO
*[https://jira.whamcloud.com/browse/LU-13835] - reuse same cl_dio_aio for one IO
*[https://jira.whamcloud.com/browse/LU-13836] - lctl show_route prints incorrect route aliveness
*[https://jira.whamcloud.com/browse/LU-13840] - Remove use of "-o" for lfs setquota
*[https://jira.whamcloud.com/browse/LU-13845] - Kernel crash on: lfs quota -u $(( (1<<32) -1))
*[https://jira.whamcloud.com/browse/LU-13846] - move iov iter forward by ourself
*[https://jira.whamcloud.com/browse/LU-13855] - build: missing dependency check on binutils-devel
*[https://jira.whamcloud.com/browse/LU-13891] - ladvise: buffer allocated for ioctl is never released
*[https://jira.whamcloud.com/browse/LU-13896] - Reference leak in check for mr_forwarding
*[https://jira.whamcloud.com/browse/LU-13899] - IOR data corruption detected during automated lnet fofb testing
*[https://jira.whamcloud.com/browse/LU-13900] - don't call aio_complete() in lustre upon errors
*[https://jira.whamcloud.com/browse/LU-13909] - release invalid dentries proactively on client
*[https://jira.whamcloud.com/browse/LU-13910] - return 0 if no error is found in mdt_path_current()
*[https://jira.whamcloud.com/browse/LU-13915] - Avoid atomic operation while bitmap prefetch
*[https://jira.whamcloud.com/browse/LU-13918] - lfs mirror extend command ignores flags
*[https://jira.whamcloud.com/browse/LU-13920] - HSM: hsm_actions are not processed after MDS failover
*[https://jira.whamcloud.com/browse/LU-13922] - Client blocked in lstat()
*[https://jira.whamcloud.com/browse/LU-13931] - lhsm_posix with lt- prefix when build from source tree with --disable-shared
*[https://jira.whamcloud.com/browse/LU-13935] - ofd_attr_set and ofd_destroy do not properly handle object removal
*[https://jira.whamcloud.com/browse/LU-13940] - it_lock_bits should be bit-wise tested
*[https://jira.whamcloud.com/browse/LU-13943] - lnet-selftest fails due to clients and server lists are empty
*[https://jira.whamcloud.com/browse/LU-13945] - copy_file_range() compat definition lacks proper headers
*[https://jira.whamcloud.com/browse/LU-13948] - t-f defect: MODOPTS_<module> ignored on FAILOVER=HARD mode
*[https://jira.whamcloud.com/browse/LU-13949] - autogen.sh is missing from the tarball generated by "make dist"
*[https://jira.whamcloud.com/browse/LU-13960] - conf-sanity test 53a and 53b fail with ‘/usr/lib64/lustre/tests/functions.sh: line 152: var: invalid indirect expansion’
*[https://jira.whamcloud.com/browse/LU-13967] - Tests fail/hang with “can't notify - lge_glbl_data is not set”
*[https://jira.whamcloud.com/browse/LU-13972] - kiblnd can continue attempting to reconnect indefinitely.
*[https://jira.whamcloud.com/browse/LU-13975] - 'ls -R' fails on encrypted dir without key
*[https://jira.whamcloud.com/browse/LU-13981] - tests should be consistent in use of "truncate"
*[https://jira.whamcloud.com/browse/LU-13982] - test_184c has possible infinite loop.
*[https://jira.whamcloud.com/browse/LU-13983] - rmdir should release inode on Lustre client
*[https://jira.whamcloud.com/browse/LU-13984] - Failing to send lock callbacks keeps a CPU busy
*[https://jira.whamcloud.com/browse/LU-13985] - seq_file next function must change *pos
*[https://jira.whamcloud.com/browse/LU-13986] - livelock is possible in distribute_txn_commit_thread()
*[https://jira.whamcloud.com/browse/LU-13987] - Don't re-enqueue glimpse lock on read
*[https://jira.whamcloud.com/browse/LU-13988] - mdt_lock_handle_fini()) ASSERTION( !lustre_handle_is_used(&lh->mlh_reg_lh) ) failed
*[https://jira.whamcloud.com/browse/LU-13989] - BL AST vs failed lock enqueue race
*[https://jira.whamcloud.com/browse/LU-13990] - ldlm_flock_deadlock() ASSERTION(req != lock) failed
*[https://jira.whamcloud.com/browse/LU-13992] -  osc_extent_make_ready()) ASSERTION( last_oap_count > 0 ) failed
*[https://jira.whamcloud.com/browse/LU-13994] - osc_object_ast_clear()) ASSERTION( lvb != ((void *)0) ) failed
*[https://jira.whamcloud.com/browse/LU-13998] - tests: sanityn test_104 incorrect MDT times printed
*[https://jira.whamcloud.com/browse/LU-14006] - raise CAP_NET_BIND_SERVICE before calling rdma_resolve_addr()
*[https://jira.whamcloud.com/browse/LU-14009] - BUG at ldiskfs/htree_lock.c:892
*[https://jira.whamcloud.com/browse/*[https://jira.whamcloud.com/browse/LU-14010] - dkms-deb client package missing dependant packages
*[https://jira.whamcloud.com/browse/LU-14012] - new layout incompletely initialized in lod_layout_convert()
*[https://jira.whamcloud.com/browse/LU-14016] - libcfs: use atomic64_t for libcfs_kmemory
*[https://jira.whamcloud.com/browse/LU-14024] - ofd_inconsistency_verification_main use after free on shutdown.
*[https://jira.whamcloud.com/browse/LU-14027] - Client recovery statemachine hangs in recovery disconnected during lock reply
*[https://jira.whamcloud.com/browse/LU-14031] - long time between reconnects
*[https://jira.whamcloud.com/browse/LU-14036] - lbuild fail to build MOFED 5.1
    * [LU-14037] - lu_device_fini()) ASSERTION( atomic_read(&d->ld_ref) == 0 )
    * [LU-14039] - Set LA_TYPE while working on osp-mdt device
    * [LU-14042] - LBUG LustreError: 7775:0:(lov_io.c:475:lov_io_slice_init()) ASSERTION( obj->lo_lsm != ((void *)0) ) failed:
    * [LU-14045] - Fix O_DIRECT and encrypted files
    * [LU-14049] - ofd_access_log_reader thread leak
    * [LU-14050] - ofd_access_log_reader off by one error in alr_sort_and_print_thread()
    * [LU-14051] - ofd_access_log_reader should flush batch file in the sort and print thread
    * [LU-14053] - ofd_access_log_reader may fail to notice that consumer exited
    * [LU-14057] - Quiet innacurate log message on server side
    * [LU-14061] - LU-12514 (lustre_tgt mount type) breaks scripts that used 'mount -t lustre' and expected targets
    * [LU-14069] - OBD_FAIL_LDLM_CANCEL_BL_CB_RACE is buggy in ldlm_handle_cp_callback
    * [LU-14070] - consider recovering case in tgt_brw_unlock()) assertions
    * [LU-14072] - client evicitions with DIO
    * [LU-14080] - gnilnd updates for 5.3+ kernel changes
    * [LU-14082] - Use l_req_mode while lock cleanup
    * [LU-14083] - Don't overwrite KBUILD_EXTRA_SYMBOLS
    * [LU-14095] - Multiple tests crash with “ASSERTION( rsi->h.cache_list.next == ((void *)0) ) failed “
    * [LU-14097] - conf-sanity/91 fails on local setup
    * [LU-14100] - gnilnd minor build fix
    * [LU-14102] - sanity/160l FAIL: MTIME changelog not recorded
    * [LU-14105] - lfsck shouldn't LBUG() on disk data
    * [LU-14109] - export --backup option is missing from lnetctl man page
    * [LU-14115] - Project Quota broken on Stripe directory
    * [LU-14116] - gss: undefined reference to 'krb5int_enc_des3'
    * [LU-14121] - EACESS Permission denied for slurm log files when using nodemap with admin=0
    * [LU-14125] - client starved for grant but OST has plenty of free space
    * [LU-14128] - truncate does not set file size correctly in corner case
    * [LU-14130] - Skip 0 row in rpc_stats procfs
    * [LU-14131] - reduce credits for overwritting
    * [LU-14136] - lfs mirror extend prints confusing error messages
    * [LU-14143] - SEEK_HOLE returns -ENXIO if file ends at components boundary
    * [LU-14149] - FIEMAP should set FIEMAP_EXTENT_DATA_ENCRYPTED for fscrypt files
    * [LU-14151] - GSS context initialization fails on RHEL8.2 client
    * [LU-14155] - Avoid easize set to 0
    * [LU-14158] - lfs changelog do not display old changelog after changelog_catalog  wrapped arround
    * [LU-14159] - gcc8 warning on kthread_run calls
    * [LU-14161] - conf-sanity/91 may fail due to rhashtable restart
    * [LU-14172] - DIR Stat performance regression in striped dir
    * [LU-14181] - sanity test_64e fails with SHARED_KEY on RHEL/CentOS 8.2
    * [LU-14185] - alloc fails for OSP obj xattr cache
    * [LU-14186] - fallocate does not extend uninitialized PFL components
    * [LU-14187] - LU-14134 causes many racer timeouts
    * [LU-14191] - setstripe: cannot create composite file: Numerical result out of range
    * [LU-14192] - overstriping is not checked for existence in the 1st component
    * [LU-14198] - Several dist-* make-targets broken when building with uid > 2^21
    * [LU-14199] - sanity-selinux test 21a  fails with 'client mount without sending sepol should be refused'
    * [LU-14205] - RHEL8.3: sanity test 398c fails with 'fio write error'
    * [LU-14208] - sanity-hsm test 55 fails with 'request on 0xM:0xN:0x0 is not FAILED on mds1'
    * [LU-14214] - incorrect locking in ofd_object_fallocate()
    * [LU-14227] - sanity-lnet test 21 missing space leading to invalid command
    * [LU-14263] - Kerberos ticket renewal not processed by lustre
    * [LU-14267] - osd_ldiskfs_write_record(): do not update in-bh inode every time
    * [LU-14281] - sanity-sec test 56 fails with 'filefrag /mnt/lustre/d56.sanity-sec/f56.sanity-sec failed'
    * [LU-14283] - sanity-sec test 52 crashes due to ‘divide error: 0000 [#1] SMP PTI’
    * [LU-14286] - fallocate() does not zero on-disk data
    * [LU-14296] - really raise CAP_NET_BIND_SERVICE before calling rdma_resolve_addr()
    * [LU-14298] - sanity-sec test_44 fails on aarch64 client
    * [LU-14299] - sanity-quota test 61 fails with 'write failed, expect succeed'
    * [LU-14301] - EOPNOTSUPP, ENOTSUPP, ENOTSUP confusion
    * [LU-14303] - parallel-scale test rr_alloc fails with 'failed while setting qos_threshold_rr & creat_count'
    * [LU-14306] - sanity-sec test_52: BUG: Bad rss-counter state
    * [LU-14307] - sanity test 60g crashes with ‘unable to handle kernel NULL pointer dereference at 0000000000000020’ in lquota_lqe_debug
    * [LU-14312] - Interop: sanity test 272b fails with 'failed to migrate to the new composite layout'
    * [LU-14317] - building lustre on arm64 hit warning: format '%llu' expects argument of type 'long long unsigned int'
    * [LU-14324] - fsx fallocate detection is buggy
    * [LU-14326] - sanity-dom test_fsx: crash in osc_extent_make_ready()
    * [LU-14337] - return stripe_count=1 instead of 0 for DoM files
    * [LU-14339] - Add a option %H to avoid the jobid too long
    * [LU-14355] - LustreError: 3782:0:(sec.c:411:import_sec_validate_get()) import ffff8dff56545800 (CONNECTING) with no sec
    * [LU-14356] - mkfs.lustre exited with 22 on Ubuntu 20.04.1 aarch64
    * [LU-14389] - crash in lov_delete_composite() with racer+migrate
    * [LU-14424] - write performance regression in Lustre-2.14.0-RC1
** Test
    * [LU-13586] - check quota pools with PFL and SEL
    * [LU-13810] - Check OST pool quota hard limit at file with wide striping
    * [LU-13817] - Make lfs quota -uv --pool more verbose
** New Feature
    * [LU-3606] - Implement fallocate() support for ldiskfs Lustre
    * [LU-10997] - Ubuntu 18 support
    * [LU-11023] - OST Pool Quotas
    * [LU-11025] - DNE3: directory restripe
    * [LU-11200] - Centos 8 arm64 server support
    * [LU-12036] - Add option to create new OSTs inactive
    * [LU-12275] - Client-side file data encryption
    * [LU-12461] - Contribute epython scripts to aid crash dump analysis
    * [LU-13180] - Reserve bit for RDMA-only memory RPC
    * [LU-13385] - lnetctl: calculate Lustre traffic service-id
    * [LU-13946] - OpenZFS 2.0 Compatibility
** Task
    * [LU-12861] - Add and use scnprintf() where using snprintf() is not handling return correctly
    * [LU-12904] - Support for linux kernel version 5.4
    * [LU-13006] - need test for /sys/fs/lustre/jobid_this_session
    * [LU-13238] - add OFD layer object access logs
    * [LU-13255] - Abstract code for wait-with-periodic-warning
    * [LU-13359] - change lfs quota --pool to print used based only on OSTs from the pool
    * [LU-13376] - add batching to ofd_access_log_reader
    * [LU-13479] - "could not allocate data integrity bioset" on centos7
    * [LU-13904] - Support non-modular builds
    * [LU-14034] - test-framework and sanity-hsm use libtool nonsensically
** Improvement
    * [LU-1538] - cleanup test scripts
    * [LU-2225] - sanity/27m to poll for specific state
    * [LU-4198] - Improve IO performance when using DIRECT IO using libaio
    * [LU-8465] - parallel e2fsck performance at scale
    * [LU-9019] - Migrate lustre to standard 64 bit time kernel API
    * [LU-9091] - Replace lprocfs_str_with_units_to_s64 with string_get_size()
    * [LU-9114] - Make MDS (And other server threads?) hog CPU less
    * [LU-9441] - Use kernel threads in predictable fashion to confine OS noise
    * [LU-9555] - "df /path" and "lfs df /path" should return projid-specific values
    * [LU-10235] - mkdir should check for directory existence on client before taking write lock
    * [LU-10428] - LNet events should generated without resource lock held
    * [LU-10447] - Replace use of $SETSTRIPE and $GETSTRIPE with '$LFS setstripe' and '$LFS getstripe' in test suites
    * [LU-10467] - use standard linux wait_event macros
    * [LU-10810] - SEEK_HOLE and SEEK_DATA support for lseek
    * [LU-10934] - integrate statx() API with Lustre
    * [LU-11310] - support for SLES 15
    * [LU-11365] - test-framework::add() should load ldiskfs
    * [LU-11505] - improve functions() run mpi tests with ability to set the composite layout
    * [LU-11607] - Reduce repeated function calls in Lustre test suites
    * [LU-11623] - Allow caching of open-created dentries
    * [LU-11631] - symlink migration should not create remote inodes
    * [LU-11668] - mdd_parent_fid()) ASSERTION( (((mdd_object_type(obj)) & 00170000) == 0040000) ) failed
    * [LU-11963] - Add nonrotational flag to obd_statfs
    * [LU-11971] - Send file creation time to clients
    * [LU-12043] - improve Lustre single thread read performances
    * [LU-12071] - bypass pagecache for large files
    * [LU-12133] - sanityn test_35() defect
    * [LU-12137] - update client to use iterate_shared
    * [LU-12222] - LNet should select loopback NI when possible
    * [LU-12280] - Provide option to disable soft quota expiration
    * [LU-12303] - Use lnet_health_sensitivity for restoring health for each lnet_recovery_internal
    * [LU-12332] - Add a liblustreapi call for IOC_MDC_GETFILEINFO
    * [LU-12353] - optimizations for ldiskfs quota updates
    * [LU-12355] - Support for linux kernel version 5.0
    * [LU-12400] - Support for linux kernel version 5.2
    * [LU-12410] - Refactor lnetctl peer nid handling to utilize nidstrings library
    * [LU-12477] - Remove obsolete config checks
    * [LU-12518] - improve Lustre unaligned IO read performances
    * [LU-12521] - print_instance() incorrect if fsname contains a dash
    * [LU-12542] - LDLM improvements form linux lustre client work
    * [LU-12546] - add option to abort recovery between MDTs but not between client/MDT
    * [LU-12631] - Report latency of client operations
    * [LU-12634] - Support for linux kernel version 5.3
    * [LU-12662] - Run pjdfstest POSIX test in review and full testing
    * [LU-12722] - exclude local client mounted on MDS/OSS from recovery
    * [LU-12785] - DOM2: dynamic DoM component size as MDT becomes full
    * [LU-12806] - llapi: have llapi_fd2fid rely on name_to_handle_at() rather than ioctl()
    * [LU-12820] - obd_dirty_transit_pages is always zero and can be removed
    * [LU-12871] - enable changelog garbage collection by default
    * [LU-12899] - Make Lustre build report the required package: kernel-rpm-macros
    * [LU-12905] - sanity/51e can take less with debugging disabled
    * [LU-12910] - osc.*.short_io_bytes parameter cannot be set to the maximum
    * [LU-12929] - Field net_state in struct lnet_net is never used, only set.  It should be discarded
    * [LU-12930] - use schedule_timeout_{un,}interruptible where possible.
    * [LU-12931] - Fix minor errors with cfs_time_seconds usage
    * [LU-12936] - Discard pt_number from lnet_peer_table
    * [LU-12941] - Track aliveness at the peer level
    * [LU-12942] - Optimize check for routing feature flag
    * [LU-12950] - Сhange force_over_512tb lustre mount option to force_over_1024tb for ldiskfs
    * [LU-12966] - print logs when lfsck detected inconsistent_owner
    * [LU-12973] - lustre/doc/.gitignore ignores relevant files
    * [LU-12977] - fix i_mutex for ldiskfs_truncate() in osd_execute_truncate()
    * [LU-12984] - Add --newerXY support for lfs find
    * [LU-12988] - improve mount time on huge ldiskfs filesystem
    * [LU-13004] - Use a single style of IO buf descriptor  - iovec
    * [LU-13005] - Remove LNet event queuing subsystem (lib-eq)
    * [LU-13013] - osd can save credits on mapped blocks
    * [LU-13036] - Reduce memory consumption by lnet.
    * [LU-13056] - lctl llog_catlist, llog_print, etc. should default to MGS device
    * [LU-13059] - kernel update [RHEL7.7 3.10.0-1062.9.1.el7]
    * [LU-13098] - supress connection restore message on idling connections
    * [LU-13110] - kernel update [SLES12 SP4 4.12.14-95.45.1]
    * [LU-13111] - support for SLES12 SP5
    * [LU-13120] - lustre-osd-zfs-mount depends on zfs, kmod-lustre-osd-zfs
    * [LU-13122] - osd-zfs to use 8K blocksize for llog objects
    * [LU-13127] - make Lustre error when hash algo is not available
    * [LU-13130] - sanity-scrub to use total device size with ZFS
    * [LU-13134] - try to use slab allocation for cl_page
    * [LU-13141] - Include block allocator improvements to additional distributions
    * [LU-13151] - Add parent FID to MTIME changelogs
    * [LU-13164] - remove unused LUSTRE_DIRECTIO_FL
    * [LU-13169] - Allow test-framework to repeat subtests
    * [LU-13178] - Update ZFS Version to 0.8.3
    * [LU-13224] - Expose lctl get/set/list param functionality in lustreapi
    * [LU-13226] - SUSE 15 SP2 server ldiskfs
    * [LU-13241] - mount.lustre for large filesystem runs slow debugfs commands
    * [LU-13258] - Bind linux workqueues to specific core
    * [LU-13268] - customize lnet-selftest for performance
    * [LU-13269] - make lnet-selftest.sh more flexible
    * [LU-13272] - customise dom-performance
    * [LU-13274] - Building againt lustreapi using -std=c99
    * [LU-13275] - enhance racer with various FOFB scenarios
    * [LU-13280] - customize run_mdtest() for several mount points
    * [LU-13281] - ha.sh improvements: different layouts  for clients directories
    * [LU-13283] - ha.sh improvement: add racer to nonmpi load
    * [LU-13297] - parallel-scale enhancement
    * [LU-13312] - Optimized RA for stride read under memory pressure
    * [LU-13347] - kernel update RHEL 7.8
    * [LU-13366] - SEL addons
    * [LU-13369] - kernel update [RHEL7.7 3.10.0-1062.18.1.el7]
    * [LU-13383] - lazy atime update on OST
    * [LU-13388] - racer crash: general protection fault: 0000 RIP: lod_obj_for_each_stripe+0xf7/0x2d0 [lod]
    * [LU-13394] - avoid sending OST RPC for project setting
    * [LU-13405] - kernel update [SLES15 SP1 4.12.14-197.37.1]
    * [LU-13412] - fix read if readahead window smaller than rpc size
    * [LU-13421] - kernel update [RHEL8.1 4.18.0-147.8.1.el8_1]
    * [LU-13445] - enhance ha.sh to support different users
    * [LU-13488] - Support RHEL 8.2
    * [LU-13501] - Disable health on single-rail deployments
    * [LU-13502] - Provide ability to fine-tune response tracking
    * [LU-13503] - allow setting larger max_mod_rpcs_in_flight dynamically
    * [LU-13509] - Improve ptlrpc_register_bulk() behavior
    * [LU-13510] - Allow control over LND timeouts independent of lnet_transaction_timeout and retry_count
    * [LU-13533] - Disable lazy_itable_init
    * [LU-13556] - kernel update [RHEL7.8 3.10.0-1127.8.2.el7]
    * [LU-13571] - Refine which network errors result in LNet Health activity
    * [LU-13585] - add mustfail check to ha.sh
    * [LU-13590] - support for RHEL 7.9
    * [LU-13595] - add option to lustre_rmmod to dump debug logs
    * [LU-13597] - add processing time/latency, IO sizes, stats to job_stats
    * [LU-13600] - limit number of RPCs in flight during recovery
    * [LU-13651] - Conditionally skip finding compatible HSM requests
    * [LU-13657] - kernel update [RHEL8.2 4.18.0-193.6.3.el8_2]
    * [LU-13676] - script to show unique backtraces from crash utility
    * [LU-13699] - Improve: mdt_identity.c Error handling and Error message reporting
    * [LU-13723] - Use ERR_CAST() instead of ERR_PTR(PTR_ERR())
    * [LU-13734] - Allow duplicate nets in kernel ip2nets syntax
    * [LU-13735] - Loosen restrictions on LNet Health parameter settings
    * [LU-13740] - Ubuntu 20.04 LTS release
    * [LU-13761] - MOFED 5.1 support
    * [LU-13782] - LNet Routers should monitor the ni_fatal flag to inform peers of changes to route status
    * [LU-13819] - Update ZFS Version to 0.8.4
    * [LU-13827] - ofd_access_log_reader to print top hot files
    * [LU-13837] - Introduce constant for LNET_NIDNET(LNET_NID_ANY)
    * [LU-13839] - support for RHEL 8.3
    * [LU-13919] - kernel update [RHEL7.8 3.10.0-1127.19.1.el7]
    * [LU-13921] - change output format for ofd_access_log_reader
    * [LU-13944] - t-f: add ability to set clients parameters
    * [LU-13961] - kernel update [RHEL8.2 4.18.0-193.19.1.el8_2]
    * [LU-13969] - Updates to lustre-release yaml.sh
    * [LU-13977] - sanity test_255a uses integer comparison for non-integers
    * [LU-14020] - Update SUSE SP2 ldiskfs
    * [LU-14029] - kernel update [SLES15 SP2 5.3.18-24.24.1]
    * [LU-14043] - lfs mirror extend need not use O_DIRECT on source
    * [LU-14052] - ofd_access_log_reader support for multiple MDTs
    * [LU-14054] - add option to ofd_access_log_reader to exit on close of last log
    * [LU-14071] - add OST pool quota options for lfs-quota.1 and lfs-setquota.1
    * [LU-14075] - kernel update [RHEL8.2 4.18.0-193.28.1.el8_2]
    * [LU-14077] - kernel update [SLES15 SP1 4.12.14-197.64.1]
    * [LU-14129] - kernel update [RHEL7.9 3.10.0-1160.6.1.el7]
    * [LU-14134] - reduce credits for new writing potentially
    * [LU-14140] - osd_idc_find_and_init() should dump details
    * [LU-14154] - ldiskfs doesn't builld with gcc10
    * [LU-14170] - Update ZFS version to 2.0.0
    * [LU-14194] - add mdc/mdd related checksum parameter files
    * [LU-14218] - kernel update [SLES15 SP1 4.12.14-197.75.1]
    * [LU-14265] - Allow "lfs mkdir" without arguments
    * [LU-14334] - update lnet/ChangeLog with current information
** Technical task
    * [LU-4423] - Tracking of patches from upstream kernel to Lustre client
    * [LU-10664] - DoM: make DoM lock enqueue non-blocking
    * [LU-11300] - LNet: Router Aliveness and Health
    * [LU-12923] - replace CLASSERT() with BUILD_BUG_ON()
    * [LU-13415] - DoM: keep optimizations even when OST stripe is initialized

Revision as of 21:39, 17 February 2021

Server support for kernels:

  • RHEL 8.3 (4.18.0-240.1.1.el8_3)

Client support for unpatched kernels:

  • RHEL 8.3 (4.18.0-240.1.1.el8_3)
  • SLES15 SP2 (5.3.18-24.24.1)
  • Ubuntu 20.04 (5.4.0-40)

Recommended e2fsprogs version:

  • v1.45.6.wc5

Interoperability Support:

  • Clients & Servers: Latest 2.12.X, 2.13

Issues fixed between 2.13.0 and 2.14.0