Lustre 2.13.0 Changelog

Server support for kernels:

• RHEL 7.7 (3.10.0-1062.1.1.el7)

Client support for unpatched kernels:

• RHEL 7.7 (3.10.0-1062.1.1.el7)
• SLES12 SP4 (4.12.14-95.29.1)
• Ubuntu 18.04 (4.15.0-64)

Recommended e2fsprogs version:

• v1.45.2.wc1

Interoperability Support:

• Clients & Servers: Latest 2.12.X

Issues fixed between 2.12.0 and 2.13.0

New Feature

• LU-1365: Implement ldiskfs LARGEDIR support for e2fsprogs
• LU-9846: Overstriping - more than 1 stripe per OST per component
• LU-10070: PFL self-extending file layout
• LU-10092: PCC: Lustre Persistent Client Cache
• LU-11297: Align LNet routing with Multi-Rail and LNet health
• LU-11376: Special file/dir to represent DAOS Containers
• LU-12624: striped directory allocate stripes by QoS
• LU-12734: bash completion (for now: lctl set/get_param)

Improvement

• LU-1538: cleanup test scripts
• LU-2233: sanityn/40-46 takes ~800-900s
• LU-7004: fix "lctl set_param -P" to allow deprecation of "lctl conf_param"
• LU-8207: Add auto-stripe option to lfs_migrate
• LU-8434: Rewrite test framework using Python
• LU-8939: nodemap: remove deprecated lproc files
• LU-8955: Send SELinux policy info to server
• LU-9010: use static initialization for statically allocated objects when possible
• LU-9625: remove "lfs ls" and "lfs cp" and references
• LU-9629: lfs migrate does not work as a non-root user
• LU-10258: lfs mirror command to open fd on component
• LU-10885: enable flock by default
• LU-11010: Remove call to ‘return’ after call to ‘skip’ in Lustre test suites
• LU-11011: checksum type can not be selected permanently
• LU-11022: FLR1.5: "lfs mirror" usability for Burst Buffer
• LU-11041: enable build and test with kernel debug options
• LU-11078: customise run_ior()
• LU-11089: Performance improvements for lu_object locking
• LU-11213: DNE3: remote mkdir() in ROOT/ by default
• LU-11221: Do not hold pages locked for network IO on the server.
• LU-11233: gcc8 support
• LU-11285: don't stop on the first blocked lock in ldlm_reprocess_queue()
• LU-11296: page cache cleanup improvement on lock flush
• LU-11329: add get_maintainer.pl script from kernel
• LU-11355: Add Lustre trim support
• LU-11367: integrate LSOM with lfs find
• LU-11398: clean up ptlrpc completion callbacks
• LU-11408: propagate grant shrink interval immediately
• LU-11415: ksocklnd performance improvement on 40Gbps ethernet
• LU-11456: MDT can skip sync if the object is synced yet
• LU-11462: osd-zfs can skip object sync if it's not dirty
• LU-11467: FLR1.5: add "lfs mirror delete" command
• LU-11504: improve ha.sh to trigger lfsck after/during failover/failback
• LU-11526: Support 64MB I/O RPC
• LU-11555: warn when mounting ZFS pools without "multihost" enabled
• LU-11606: use llapi_error() for errors in liblustreapi files
• LU-11616: Optimize handling statahead delay
• LU-11673: Correct shell issues/mistakes in Lustre test suites
• LU-11717: t-f:exit_status () defect: test returns code 1 if log contains "FAIL"
• LU-11736: Do not apply bulk IO tuning on MDT or MGT devices
• LU-11739: Don't inherit default layout from root for new subdirectories
• LU-11810: allow "Fixes:" tag in commit message signoff block
• LU-11811: Provide an interface to set Lustre MDT into readonly mode
• LU-11820: reduce conf-sanity test duration
• LU-11830: Quiet compiler warnings on ARM
• LU-11834: config instance truncation due to buffer size mismatch
• LU-11838: Support linux kernel version 4.18
• LU-11843: customise sanity-quota to work with project quotas set before the suite
• LU-11845: Support encrypted ZFS volumes
• LU-11852: osc cache waiters wakeups and skipping queue
• LU-11876: Lustre dkms packages missing build dependency
• LU-11885: reset quota setting upon test failed
• LU-11902: Remove obsolete lustre-utils files and entries for DEB packages
• LU-11913: mq-deadline should be in the list of allowed schedulers
• LU-11914: Build error for l_getsepol.c due to missing openssl/evp.h
• LU-11923: meta_bg conflicts with resize_inode if explicitly specified for a small filesystem
• LU-11924: Combine cancellation of llog records at osp_sync_process_committed
• LU-11963: Add nonrotational flag to obd_statfs
• LU-11974: Improve llapi_layout_get_by_xattr()
• LU-11982: Incorrect text in lfs migrate help
• LU-11999: DNE performance improvement
• LU-12015: Update Changelog for ubuntu kernel version
• LU-12019: Recognize Debian Kernel in autoconf and unify kernel module package directory
• LU-12027: Allow "lfs find" atime/mtime/ctime with fine-grained values
• LU-12033: Git ignore generated *.install server and client files
• LU-12044: Translate HABEO_ macros to English
• LU-12045: Ability to split test script execution into parts
• LU-12090: Create an lfs command to remove files by FID
• LU-12093: CAP_SYS_RESOURCE capability check overhead
• LU-12103: Improve block allocation for large partitions
• LU-12133: sanityn test_35() defect
• LU-12137: update client to use iterate_shared
• LU-12157: lfs_migrate -v output format is not well-formed
• LU-12159: allow lfs getname to only print fsname
• LU-12202: remove obsolete files from contrib
• LU-12226: upgrade e2fsprogs to 1.45.2
• LU-12236: Support more than the default root network namespace
• LU-12270: Support linux kernel version 4.19
• LU-12302: 0@lo status shows as "down"
• LU-12313: Mark lustre_inode_cache as reclaimable
• LU-12326: test-framework to recognize Bochs as VM
• LU-12330: Support per-session JOBID without using the environment
• LU-12331: hardened usercopy whitelisting
• LU-12333: Add additional rpc flags to DEBUG_REQ_FLAGS
• LU-12341: Make test-framework aware of kmemleak
• LU-12349: print error to console when flock() is diabled
• LU-12368: concurrent statfs() calls on the client should be blocked
• LU-12430: remove misleading help message of "Consult build/README.kernel-source for details"
• LU-12439: Convert "Response timed out..." in lnet_finalize_expired_responses to CDEBUG
• LU-12440: Correct lnet_is_health_check when msg status is 0
• LU-12492: Deprecate live/dead router check interval parameters (instead of removing them)
• LU-12509: lockdep warning in ofd_precreate_objects()
• LU-12516: volatile file creation in .lustre/fid/
• LU-12520: reserve lru pages for read in batch
• LU-12527: lustreapi.h not c++ compatible
• LU-12533: Improve readahead RPC issuance for large window sizes
• LU-12535: lov_page_index surprisingly slow
• LU-12545: Cleanup LPROC_LL_*
• LU-12553: polling mode for changelog driver
• LU-12562: sanity 256 to use /tmp for temporary file
• LU-12578: obdfilter-survey w/echo-client does not gain from direct-io optimizations
• LU-12621: o2iblnd QP sizing -EINVAL retries
• LU-12635: gcc: -Werror=implicit-fallthrough
• LU-12776: update lustre/ChangeLog to reflect primary kernels
• LU-12789: Mellanox MOFED 4.7 support.
• LU-12802: racer test stucks after test finishes
• LU-12825: Build MOFED 4.7
• LU-12845: Fix typos in ‘lctl pcc’ help message

Bug

• LU-1732: enable wide striping by default
• LU-1957: Test failure on test suite sanity, subtest test_180b
• LU-4322: sanity test_101a: too many discarded pages
• LU-4398: mdt_object_open_lock() may not flush conflicting handles
• LU-5604: Lots of FAIL_ID checking are lost
• LU-6504: Wrong indentation in socklnd
• LU-6836: sanity-quota test_4a: Passed grace time 12, 1436542665, 1436542679
• LU-6951: sanity test_27m: @@@@@@ FAIL: OST0 was full but new created file still use it
• LU-7159: sanity 224c always passes
• LU-7239: mdd_attr_set() synchronous when it need not be
• LU-7558: niobuf.c:721:ptl_send_rpc() LASSERT(AT_OFF || imp_state != LUSTRE_IMP_FULL || imp_msghdr_flags & MSGHDR_AT_SUPPORT ...)
• LU-7631: conf-sanity test_82a: getstripe -c wrong: found 2, expected 3
• LU-7912: Stale comment in osc_page_transfer_add
• LU-8047: optimize ll_md_blocking_ast() for a failed lock
• LU-8260: osd_ldiskfs fiemap code wrong address space usage
• LU-8299: (vvp_io.c:922:vvp_io_fault_start()) ASSERTION( cl_index(obj, offset) == fio->ft_index )
• LU-8760: sanity-lfsck test 31g hung
• LU-9193: Multiple hangs observed with many open/getattr
• LU-9341: PFL: append should not instantiate full layout
• LU-9706: conf-sanity test_53a: MDS soft lockup - CPU#0 stuck for 22s in osd_trans_commit_cb()
• LU-9792: "loop" devices testing with zfs is broken
• LU-9835: Potential buffer overrun in jt_opt_threads()
• LU-9862: The patch that landed for LU-7890 leads to static checker errors
• LU-9920: Use pagevec for marking pages dirty
• LU-9964: > 1 group lock on same file (group lock lifecycle/cbpending problem)
• LU-9971: MR: ABA problem in lnet_discover_peer_locked
• LU-10048: osd-ldiskfs to truncate outside of main transaction
• LU-10094: sanity test_17f: 'ls' fails with "ls: reading directory *: Input/output error"
• LU-10100: sanity test_27a: setstripe failed with "error on ioctl 0x8008669a for '*' (3): Invalid argument"
• LU-10153: route via two different networks not supported
• LU-10239: Lustre crash (client): The first extent to be fit in a RPC contains 17 chunks, which is over the limit 16.
• LU-10384: Replace nids doesn't add failover nid and add_conn string to config
• LU-10496: sanity-dom sanity test_39k: FAIL: mtime is lost on close: 1515705427, should be 1484169395
• LU-10717: several conf-sanity tests failed: FAIL: start mgs failed
• LU-10754: sanityn test 47b fails with 'create must fail'
• LU-10815: sanity-scrub test_16: @@@@@@ FAIL: test_16 failed with 2
• LU-10862: ldlm_pools_{count,shrink}() DLMTRACE it too noisy
• LU-10866: Add mount options for pacemaker script
• LU-10883: llog backup files processing is not skipped because of string сomparison error
• LU-10913: conf-sanity test_106: OST0000-osc-MDT0000: can't cancel record
• LU-10929: skip sanity/315 if IO accounting is not supported
• LU-10931: failed peer discovery still taking too long
• LU-10949: LBUG lu_object.h:688:lu_object_get()) ASSERTION( atomic_read(&o->lo_header->loh_ref) > 0 ) failed
• LU-10953: sanity/300k fails on local small setups
• LU-10966: `lfs check` documentation is misleading
• LU-11090: sanity-lfsck test_4 failed: Oops: 0000 [#1] IP: qsd_config 0x110/0x350 [lquota]
• LU-11111: crash doing LFSCK: orph_index_insert()) ASSERTION( !(obj->mod_flags & ORPHAN_OBJ)
• LU-11157: sanity test_42e: invalid arithmetic operator (error token is ".9")
• LU-11161: sanity test_160g fails for DNE with mds2: User cl9 still registered
• LU-11204: mdt_reint_unlink->lu_object_put() crash
• LU-11206: recovery-mds-scale test failover_ost fails with 'import is not in FULL state'
• LU-11208: Interop 2.10.4<->master sanity test_27G: 'testpool' is not empty
• LU-11239: sanity-lfsck test 36a fails with 'Fail to split mirror'
• LU-11243: Assertion and hang upon lod_add_device failure
• LU-11251: replay-single test_70b test: ASSERTION( req_transno < next_transno ) failed: LBUG
• LU-11292: LNet: Run discovery protocols on routers
• LU-11330: replay-single test_70d: Directory not empty rmdir fails
• LU-11359: racer test 1 times out with client hung in dir_create.sh, ls, … and MDS in ldlm_completion_ast()
• LU-11401: Interop: sanity-flr test 203 needs server version check
• LU-11402: Interop: sanity-quota test 60 needs a server version check
• LU-11403: lov_io_init_empty() Page fault on a file without stripes
• LU-11409: only first 100 OSC can shrink grants
• LU-11413: Large performance degradation in routing environment.
• LU-11418: hung threads on MDT and MDT won't umount
• LU-11426: 2/2 Olafs agree: changelog entries are emitted out of order
• LU-11434: Interop 2.10.5<->master conf-sanity test_109a: clear_conf failed
• LU-11444: RPC resend may corrupt the data
• LU-11453: sanity test 184a: Basic layout swap panics on Power8
• LU-11485: MDS allows "lfs setstripe" to mark last mirror as stale
• LU-11495: t-f defect: NAME=ncli llmount failed with FILESET set
• LU-11496: recovery scale tests with IOR load are not functional
• LU-11503: ha.sh defects: ha.sh: line 399: (1476438773 - start_time) / nr_loops: division by 0
• LU-11537: Fix assertion: osd_trans_exec_op()) ASSERTION( !ldiskfs_track_declares_assert )
• LU-11542: Race between imp_state and imp_invalid
• LU-11549: Unattached inodes after 3 min racer run.
• LU-11566: sanity test_60aa: llog_print_cb()) not enough space for print log records
• LU-11568: Get rid of SLAB_DESTROY_BY_RCU
• LU-11575: make debs fail on debian without systemd enabled
• LU-11579: cl_file_inode_init()) ASSERTION(inode->i_state & (1 << 3) ) failed:
• LU-11584: kernel BUG at ldiskfs.h:1907!
• LU-11585: recovery-double-scale test_pairwise_fail: Oops: BUG: unable to handle kernel NULL pointer dereference: qsd_lqe_read+0x284/0x570 [lquota]
• LU-11586: sanity-pfl test_10: FAIL: /mnt/lustre/d10.sanity-pfl/file2 does not inherite root layout (subdiretory mounts)
• LU-11591: MDS n003 crash, LustreError: 20558:0:(osp_sync.c:1258:osp_sync_thread()) ASSERTION( thread->t_flags != SVC_RUNNING ) failed: 4632 changes, 16403 in progress, 0 in flight
• LU-11594: sanity test_103a: FAIL: permissions failed
• LU-11601: IR doesn't handle EAGAIN after initial connect when pinger_recov is 0
• LU-11605: create_count stuck in 0 after changeing max_create_count to 0 and back 20 000
• LU-11617: lockdep exposed a possible deadlock in chlg_open()
• LU-11620: BUG: sleeping function called from invalid context at mm/slub.c:940
• LU-11625: ls -l returns "ls: cannot access Invalid argument"
• LU-11626: mdc: obd might go away while referenced by code in mdc_changelog
• LU-11634: sanityn/test_77 improvements
• LU-11636: t-f test_mkdir() does not support interop with non DNEII servers
• LU-11647: niobuf.c:330:ptlrpc_register_bulk()) ASSERTION( desc->bd_md_count == 0 ) failed:
• LU-11658: The cl_cache may be uninitialized while osc is activating
• LU-11661: sanityn: test_47g failed
• LU-11665: sanity test 101c: FAIL: Small 4k read IO 240 !
• LU-11669: add project in yml_test_group()
• LU-11670: Incorrect size when using lockahead
• LU-11672: improving lru_max_age policy when lru resize is disabled
• LU-11675: Don't allow new HSM requests during CDT_INIT
• LU-11678: sanity-quota test 1 fails with 'user write success, but expect EDQUOT'
• LU-11681: sanity test 65i fails with 'find /mnt/lustre failed'
• LU-11690: LBUG with very wide striping: lod_ea_store_resize()) ASSERTION( round <= lov_mds_md_size(2000, (0x0BD30000 | 0x0BD0))
• LU-11691: lfs getstripe buffer overflows with very large stripe counts
• LU-11696: "lfs getsom" returns "24" (sizeof lustre_som_attr) to userspace
• LU-11699: lod_device_free()) ASSERTION( atomic_read(&lu->ld_ref) == 0 ) failed: lu is ffff881a7e690000
• LU-11700: GPF in osp_reserved_mb_low_seq_write
• LU-11701: replay-single test_80a: FAIL: remote creation failed
• LU-11702: ibc_rxs is created and freed with different size
• LU-11712: Lustre build fails when CONFIG_BLK_DEV_INTEGRITY is unset
• LU-11716: sanity-scrub test_1b test: BUG: unable to handle kernel NULL pointer dereference IP: [<ffffffff816b68a9>] _raw_read_lock+0x9/0x20
• LU-11719: Refactor search_itree
• LU-11720: SRPM unnecessarily tied to kernel version
• LU-11721: "lfs df -i" prints used blocks percentage instead of inodes
• LU-11723: kernel update [RHEL7.6 3.10.0-957.1.3.el7]
• LU-11729: ARM: sanity test_810: BAD WRITE CHECKSUM with adler
• LU-11737: LustreError: 11060:0:(osd_handler.c:3985:osd_xattr_set()) ASSERTION( handle ) failed:
• LU-11742: ERROR: RPATH is not allowed
• LU-11743: lctl pool operations cannot be run on separate MGS system
• LU-11749: sanity-sec test 23b fails with 'Should return gid=60010 or 60010 on client2'
• LU-11750: lsvcgssd crash with latest krb5-libs
• LU-11752: sanity test 271f fails with '1 READ RPC occured'
• LU-11755: Remove redundant LASSERT() in ns_is_{client,server}()
• LU-11756: kib_conn leak
• LU-11757: lfs mkdir: Cannot allocate memory
• LU-11758: remove assertion from osp_statfs()
• LU-11759: racer cleanup
• LU-11760: formatted OST recognition change
• LU-11761: blocked MDT mount and high cpu usage from lodXXXX_recYYYY threads
• LU-11765: during failover test run, mdtest job fails, numerous stat failures 'No such file or directory'
• LU-11768: sanity-quota test 6 fails with ‘LNet: Service thread pid <pid> was inactive for …’
• LU-11770: preserve kernel API when T10-PI patches are applied
• LU-11771: bad output in target_handle_reconnect: Recovery already passed deadline 71578:57
• LU-11773: setstripe YAML API does not support lcme_flags
• LU-11777: interop: sanityn test 102 fails with 'check_fhandle_syscalls failed'
• LU-11778: interop: sanity test 317 fails with 'Expected Block 4096 got 10240 for f317.sanity'
• LU-11779: interop: sanity-hsm test 255 fails with 'request on 0x200007935:0x268:0x0 is not SUCCEED on mds1'
• LU-11780: interop: replay-single test 132a fails with Component #1 instantiation was not replayed
• LU-11790: ext4: include terminating u32 in size of xattr entries when expanding inodes
• LU-11796: Remove LASSERT(r0->lo_nr <= lov_targets_nr(dev)) in maintenance branches
• LU-11798: cur_grant goes to 0 and never increases with 2.8 client and 2.10 server
• LU-11816: LNet Health: Correct timeout defaults
• LU-11821: sizeof() needs correct casting to compare with negatives
• LU-11825: Remove LU-8964/pio feature & supporting framework
• LU-11827: Race between llog_cat_declare_add_rec and llog_cat_current_log
• LU-11828: PFL crashes when invariant checking is enabled
• LU-11835: DOM open resend doesn't return size
• LU-11836: DOM read-open resend vs getattr deadlock
• LU-11849: lfs find ! --projid doesn't work as expected
• LU-11851: kernel BUG at ...ldiskfs/htree_lock.c:665
• LU-11855: fix out-of-block vars
• LU-11860: lnetctl fails when configuring numeric addresses
• LU-11861: master can't build with lock debug enabled
• LU-11862: lfsck avoid unexpected repairing
• LU-11867: OST precreate object FID in LMA mismatch causes OST unusable
• LU-11868: ZFS ea size limited to 32K
• LU-11871: Missing man page lctl-nodemap-set-fileset
• LU-11873: sanity test_801a: FAIL: (2) unexpected barrier status 'expired'
• LU-11878: sanity test 103b: OOM because of too many bash processes: page allocation stalls for 18420ms
• LU-11880: Allow Lustre build on Raspbian and other Debian derivatives
• LU-11881: lctl pool_list: Value too large for defined data type (75)
• LU-11883: Make deny_unknown visible on default nodemap
• LU-11891: make --mdt-index usage consistent
• LU-11892: Memory leak in MDT Coordinator
• LU-11894: Check for asymmetrical route messages in LNet
• LU-11897: ineffective memory allocation in ptlrpc
• LU-11898: libcfs module parameter libcfs_debug_mb has no effect
• LU-11899: Possible resource/memory leaks
• LU-11905: Mismatched ENTRY/RETURN in mdc_intent_open_pack
• LU-11907: sanity fails to cleanup d60g.sanity
• LU-11911: lov_iocontrol: BUG: unable to handle kernel NULL pointer dereference at 0000000000000100
• LU-11917: sanity test_258a: error, failed to take i_mutex, rc=0
• LU-11919: cl_max_dirty_pages not zeroed on startup
• LU-11920: memory leak in lod_lds_buf_get()
• LU-11922: mkfs.lustre in 1.44.3.wc1 causes corruption if 'metadata_csum' option enabled
• LU-11925: Attributes not updated after open+append and write to archived,released file
• LU-11926: Lost lease lock on migrate error
• LU-11927: support for SLES12 SP4
• LU-11930: sbi_flags printing broke
• LU-11931: RDMA packets sent from client to MGS are timing out
• LU-11933: sanity-hsm: test_406 sanity: test_230b osd_declare_dir_delete() ASSERTION( dt_object_exists(dt) )
• LU-11934: replay-single test_70c: Oom on client
• LU-11943: many input/output error after soak running for couple of hours
• LU-11944: WARNING: at ../fs/attr.c:212 notify_change doing tiny_writes
• LU-11946: configure --enable-dist fails if libyaml-devel or zlib-devel are not installed
• LU-11947: Regression caused by LU-11461
• LU-11951: sanity: test_231a failure, idle disconnect & import generation disagreement
• LU-11956: conf-sanity test_32a failed with 1
• LU-11960: Add missing libssl-dev DEB package
• LU-11964: Heavy load and soft lockups on MDS with DOM
• LU-11965: interop: sanity test 311 fails with 'create_count is zero'
• LU-11967: MDS LBUG ASSERTION( o->opo_reserved == 0 ) failed
• LU-11975: sanity test 60a hangs in llog_test_10
• LU-11997: Crash in lustre_swab_fiemap
• LU-12013: Crashes in sanity-lfsck test 13
• LU-12014: check correct size in ll_dom_finish_open()
• LU-12017: Truncate vs setxattr deadlock with DoM
• LU-12018: deadlock on OSS: quota reintegration vs memory release
• LU-12020: LustreError: 32657:0:(mdc_lib.c:137:mdc_pack_name()) ASSERTION( lu_name_is_valid_2(buf, cpy_len) ) failed:
• LU-12021: Error message of mdt_attr_valid_xlate() when 2.12 client to 2.10 server
• LU-12025: Adding OST may cause EIO - delay activation of new OSTs on existing filesystem
• LU-12030: rhel7.6 removed kmalloc-128 slab
• LU-12034: env allocation in ptlrpc_set_wait() causes slowdown on the client
• LU-12037: Possible DNE issue leading to hung filesystem
• LU-12039: Fix 'lfs mkdir' in sanity-selinux
• LU-12040: File lost during recovery
• LU-12042: sanity test_103a: FAIL: LU-974 umask test failed under remote dir
• LU-12053: intermittent ping failures with MR/DD
• LU-12055: lod: ignore root layout if parent's one includes pool
• LU-12059: Update ZFS Version to 0.7.13
• LU-12065: Client got evicted when lock callback timer expired on OSS
• LU-12068: sanity-lfsck test_6b: (7.2) 0x0 is not larger than 0x0
• LU-12069: fix styling in mdt_reint_migrate_internal()
• LU-12072: Simplify the code for lov operating
• LU-12075: mdt_reint_migrate_internal() can break CoS logic
• LU-12080: LNet Health: references left on eq
• LU-12081: rename shouldn't PDO lock target parent if it's remote
• LU-12094: NULL pointer dereference in lprocfs_import_seq_write
• LU-12095: ocd_connect_flags are wrong during reconnect
• LU-12098: changelog_deregister appears not to reliably clear all changelog entries
• LU-12101: lctl push causes a crash
• LU-12104: SEGV during "lfs find [--mdt-count=[+,-]<count>, --mdt-hash=<hashtype>]"
• LU-12120: LustreError: 15069:0:(tgt_grant.c:561:tgt_grant_incoming()) LBUG
• LU-12127: Improve LFSCK orphan handling
• LU-12131: Improve GSS/SSK handling in Lustre test suite
• LU-12138: kernel update [SLES12 SP4 4.12.14-95.13.1]
• LU-12139: kernel update [SLES12 SP3 4.4.176-94.88.1]
• LU-12140: DOM: limitaion of data size
• LU-12141: Wrong requires kernel version detected of lustre kmod built against target rhel7.6
• LU-12147: statone utility doesn't form correct string with gcc8
• LU-12151: metadata performance difference on root and non-root user
• LU-12152: lnetctl export corrupts memory on routers
• LU-12160: use-after-free in osd_object_delete()
• LU-12165: Enable quota enforcement crashed kernel
• LU-12166: Tests if ZFS backend project supported is broken
• LU-12168: obdfilter-survey SHORT msg
• LU-12169: Assertion on mdc_pack_name
• LU-12172: kernel update [RHEL7.6 3.10.0-957.10.1.el7]
• LU-12178: MDS deadlock with 2.12.0 (quotas?)
• LU-12195: sanity/43 and sanityn/14 fail on local setups
• LU-12199: md's are not detached from uncommitted messages that have health check performed on them
• LU-12203: sanity-lfsck test_36[b,c] defect: tests are to be skipped for FILESET
• LU-12206: error handling on mdt_init0()
• LU-12208: formatall() defect: LU-8323 breaks interop with mixed old clients installed
• LU-12212: Often requests timeouts during dbench run
• LU-12218: ASSERTION( mbits == (req->rq_mbits & PTLRPC_BULK_OPS_MASK) ) failes if OBD_CONNECT_BULK_MBITS isn't used.
• LU-12219: obdfilter-survey: line 509: size * 1024 / (actual_rsz * thr): division by 0 (error token is ")")
• LU-12221: LBUG: (statahead.c:477:ll_sai_put()) ASSERTION( !sa_has_callback(sai) )
• LU-12225: jobid_get_from_cache stalled on spin_lock(&pidmap->jp_lock)
• LU-12227: Lustre init script does not check if ZFS devices are already mounted
• LU-12229: ost-pools test 1j fails with 'test_1j returned 1'
• LU-12242: kernel update [RHEL7.6 3.10.0-957.12.1.el7]
• LU-12247: sanityn test 34 fails with 'test_34 returned 1'
• LU-12248: sanity test_418: df and lfs df inodes output mismatch
• LU-12253: lfs: double free or corruption (!prev)
• LU-12261: sanity test 43b fails with 'expected error, got success'
• LU-12263: build breakage with 2.12.53 tag
• LU-12264: The lnet peer discovery queue (lnet_peer.lp_dc_pendq) is susceptible to concurrent manipulation
• LU-12266: Fix up non-directory creation in SGID directories
• LU-12267: sanity test_103a: some ACL tests fail with SELinux enforced
• LU-12276: Lustre client compilation failing on CentOS 7.6 + MOFED 4.6
• LU-12279: client got evicted due to network issue.
• LU-12282: Pass IB_OPTIONS to debian rules doesn't work
• LU-12296: ll_dom_lock_cancel() should zero kms attribute
• LU-12299: avoid panic for too large cpu partitions
• LU-12306: kernel update [RHEL7.6 3.10.0-957.12.2.el7]
• LU-12309: write fails with -EOPNOTSUPP with zfs 0.8.0-rc5 using 0.7.x pool
• LU-12314: Add Missing Description to sanity test 258a
• LU-12323: save_stack_trace_tsk is not exported on arm architectures
• LU-12324: zero transno on transaction commit
• LU-12328: FLR mirroring on 2.12.1-1 not usable if OST is down
• LU-12335: mb_prealloc_table table read/write code is racy
• LU-12342: lsvcgss config file overwritten on update
• LU-12343: osc_lock.c:687:osc_ldlm_weigh_ast()) LBUG
• LU-12345: backport - ext4: optimize ext4_find_delalloc_range() in nodelalloc mode
• LU-12350: sanity-flr test_33: file content error: expected: ost1, actual: ost2
• LU-12352: libcfs crashes with certain cpu_npartitions values
• LU-12361: cp --sparse doesn't work correctly in
• LU-12367: readahead page count calculation is incorrect for unaligned reads
• LU-12374: client went down w/ panic during lustre_rmmod
• LU-12375: lnet should start after opa service
• LU-12381: o2iblnd uses wrong IB interface
• LU-12382: Deadloop on tiny write
• LU-12383: lfs project inhert difference between ldiskfs and zfs
• LU-12385: 2.12.2: IOR read testing reliably inducing LNetError: Timed out RDMA with ...@o2ib1 (6): c: 0, oc: 0, rc: 63
• LU-12387: l_tunedisk does not properly handle multipath devices
• LU-12392: gcc8 can't build lfs.c
• LU-12394: lctl get_param llite.lfs*.extents_stats only displays first two buckets
• LU-12395: Failed dependencies while installing lustre-client-tests and lustre-iokit rpms on RHEL 8.0 client
• LU-12396: lfs df output is treated as binary data on RHEL 8.0 client
• LU-12399: sanity test_420: FAIL: incorrect perms on /mnt/lustre/d420.sanity/testdir/testfile
• LU-12401: Various issues with checksum calculation for Kerberos and SSK
• LU-12402: LNet Health: lnet_finalize() recurssion
• LU-12411: Hang on lnetctl route del
• LU-12412: mdt: task umount:19248 blocked for more than 120 seconds
• LU-12420: llog_reader crashes when reading llog that is not inited
• LU-12422: LNet MR routing: update route state in case discovery is off
• LU-12423: LNet MR routing: only push updates if discovery is enabled
• LU-12424: LNet MR routing: possible loop when discovery is off
• LU-12427: LNet MR routing: output a warning if discovery is off
• LU-12428: sanity-sec: test_13 nodemap_del failed with 1
• LU-12431: Close race on cp_state changes
• LU-12436: Memory allocation failure error dropped
• LU-12438: Use kernel_read/write instead of vfs_read/write for PCC
• LU-12441: Response tracker is not detached on router ping reply
• LU-12443: sanity test_271f: @@@@@@ FAIL: expect 1 READ RPC, occured
• LU-12446: Variable STRIPE_BYTES Undefined under sanityn.sh/16b
• LU-12447: Fedora28 Lustre client build failure
• LU-12455: readcache max file size show/store inconsistent
• LU-12456: kernel update [RHEL8.0 4.18.0-80.4.2.el8_0]
• LU-12457: support for RHEL 7.7
• LU-12458: kernel update [RHEL7.6 3.10.0-957.21.3.el7]
• LU-12462: osc_cache_writeback_range(): ASSERTION( ext->oe_start >= start && ext->oe_end <= end ) failed
• LU-12472: Kerberos: sanity-krb5.sh needs some fixes
• LU-12473: pool_list by path incorrectly prints path as pool prefix
• LU-12474: racer can't be found on Ubuntu 18.04
• LU-12478: RHEL8 packaging
• LU-12479: gcc8 warnings
• LU-12481: conf-sanity test_61: setfattr: /mnt/lustre/f61.conf-sanity: Argument list too long
• LU-12482: sanity-hsm uses nonexisting variable
• LU-12483: sanity test 60h improper mds version check.
• LU-12484: typo in is_project_quota_supported for zfs case
• LU-12485: (osd_handler.c:2146:osd_object_release()) ASSERTION( !(o->oo_destroyed == 0 && o->oo_inode && o->oo_inode->i_nlink == 0) ) faile
• LU-12489: Tests marked slow that are not slow (anymore?)
• LU-12491: large-scale: crash in lu_env_remove() RIP: memcmp+0x9/0x50
• LU-12494: kernel update [SLES12 SP4 4.12.14-95.19.1]
• LU-12495: sanity test_413b: subdirs shouldn't be evenly distributed
• LU-12501: lfs df can never return
• LU-12510: mds server hangs cv_wait_common
• LU-12513: Address YAML limitations
• LU-12523: Add jobid to rpctrace debug messages
• LU-12524: mdc_close() matched open debug msg causes memory fragmentation
• LU-12536: Processes stuck in unkillable sleep waiting on IO during Lustre re-export of NFS testing
• LU-12537: recovery-mds-scale failover_mds: BUG: unable to handle kernel NULL pointer dereference at (null) IP: lnet_attach_rsp_tracker.isra.32+0xad/0x200 [lnet]
• LU-12538: bad spinlock magic in sanity test 413b
• LU-12539: Pass --with-o2ib option along when building deb packages
• LU-12540: overstripe landing broke a MKFS options handling
• LU-12558: kernel update [SLES12 SP3 4.4.180-94.100.1]
• LU-12559: Reconnecting from idle exposes import in IMP_NEW state, resulting in EIO
• LU-12560: acceptance small doesn't use full path for finding test-groups
• LU-12565: Use bit locking in obd_echo
• LU-12566: GPF when umounting client
• LU-12568: LNetError: 28086:0:(lib-move.c:2862:lnet_detach_rsp_tracker()) ASSERTION( rspt->rspt_cpt == cpt ) failed
• LU-12569: IBLND_CREDITS_HIGHWATER does not check connection queue depth
• LU-12570: sanity test 134a crash with SSK in use
• LU-12575: lbuild: build MOFED on el8
• LU-12579: runtests: test_1: @@@@@@ FAIL: Space not all freed: now 309264kB, was 307632kB.
• LU-12581: Slab object "Freechain corrupt" on client
• LU-12586: truncate to start of uninstantiated PFL component silently fails
• LU-12587: DIO fallback to Buffer IO unexpectedly
• LU-12589: sanity test_102a: setfattr: /mnt/lustre/f102a.sanity: Numerical result out of range
• LU-12590: Lustre lustre_msg_hdr_size_v2() bug
• LU-12593: update_log corruption
• LU-12595: Attempt to add route using non-local gateway should return EHOSTUNREACH, not EINVAL
• LU-12600: Lustre tgt_brw_write() bug
• LU-12602: Lustre mdt_getxattr_pack_reply() bug
• LU-12603: Lustre ldlm_request_cancel() bug
• LU-12604: Lustre mdt_file_secctx_unpack() bug
• LU-12605: Lustre target_handle_connect() bug
• LU-12608: kernel update [RHEL7.6 3.10.0-957.27.2.el7]
• LU-12611: LNet Routing: don't stop adding routes for any error
• LU-12612: Lustre osd_bufs_get() bug
• LU-12613: Lustre lustre_msg_string() bug
• LU-12614: Lustre ldlm_cancel_hpreq_check() bug
• LU-12615: Lustre mdt_object_remote() bug
• LU-12616: MDS node crashed LustreError: 23042:0:(mdt_handler.c:5135:mdt_init0()) ASSERTION( info != ((void *)0) )
• LU-12622: Another batch of test fixes for SSK
• LU-12625: sometimes configure can't detect timespec64 with gcc8
• LU-12626: Attempt to create existing net returns EINVAL
• LU-12627: ASSERTION( info->fti_attr.la_valid == 0 )
• LU-12639: sanity test 317 uses uninitialized variable $facet
• LU-12650: get_root_path() filesystem name compare error that leads to fid2path fail
• LU-12657: sanity/103 and sanityn/25 fail with 4.18 kernel
• LU-12660: kernel update [SLES12 SP4 4.12.14-95.29.1]
• LU-12671: replay-vbr accesses non-existent mdd.sync_permission setting
• LU-12672: recovery-small test 66 does not work correctly
• LU-12674: osp should handle -EINPROGRESS on llog objects
• LU-12675: LBUG "(lu_object.c:1196:lu_device_fini()) ASSERTION( atomic_read(&d->ld_ref) == 0 ) failed: Refcount is 1" running sanity.sh on latest master
• LU-12681: Data corruption - due incorrect KMS with SEL files
• LU-12685: ll_statahead_threda has incorrect error handling code
• LU-12686: ln_mt_waitq use by lnet monitor thread is completely ineffective
• LU-12690: error handling of ll_och_fill()
• LU-12691: obd_max_recoverable_clients is not atomic
• LU-12694: make "lfs quota" display correct group quota information
• LU-12703: sanity test_421a: FAIL: rmfid with fsname failed
• LU-12705: build fails on power9 client
• LU-12707: mds-survey test 1 crashes with a NULL pointer dereference in echo_object_free()
• LU-12712: sanity-pfl tests triggering “not SEL magic on SEL file”
• LU-12719: crash in lustre_find_lwp_by_index()
• LU-12724: kernel update [RHEL7.7 3.10.0-1062.1.1.el7]
• LU-12739: Race with discovery thread completion and message queueing
• LU-12745: Lustre fails to compile against zfs dkms 0.8+
• LU-12751: error() is missing in few places in sanity.sh
• LU-12753: sanity test_300a: FAIL: 1:stripe_count is 1, expect 2
• LU-12755: CPU soft lockup on mkfs.lustre
• LU-12758: sanity-quota test 65 fails with 'usr quota not match'
• LU-12760: sanity test_161d incorrect use of stack_trap
• LU-12763: Use alternate ping processing for non-mr peers
• LU-12764: ubuntu16 client build failure
• LU-12769: replay-dual test 0b hangs in client mount
• LU-12773: sanity test 805 uses do_node instead of do_facet
• LU-12777: conf-sanity test 103 fails with ‘set mdt quota type failed’
• LU-12783: interop: sanity test 27I fails with “multiop failed”
• LU-12784: interop: sanity test 102ha fails with “saving trusted.big on /mnt/lustre/f102ha.sanity failed”
• LU-12790: LustreError: 96800:0:(jobid.c:103:jobid_get_from_environ()) jobid: get 'PBS_JOBID' failed: rc = -75
• LU-12795: lp_utils is duplicated and not used.
• LU-12799: NULL pointer dereference IP: ptlrpc_disconnect_and_idle_import
• LU-12800: mdd_is_parent() goes into infinite loop
• LU-12803: lustre in staging interferes with 2.12.2 on Ubuntu 18.04
• LU-12824: Unable to add single Infiniband interface to multiple o2ib LNets
• LU-12838: ptlrpc watchdog ratelimiting is broken
• LU-12842: lctl llog_print doesn't accept snapshot names with leading number
• LU-12844: error: ‘strncpy’ specified bound depends on the length of the source argument
• LU-12859: Both "flock" and "localflock" appear in /proc/mounts
• LU-12893: lnet: peer ni selection broken
• LU-12903: lctl-pcc man page does not exist
• LU-12932: parallel-scale test rr_alloc fails with ‘failed while setting qos_threshold_rr & creat_count’
• LU-12946: Multipath path flapping issue
• LU-13023: PCC: The size of the PCC copy is not truncated to 0 when attach
• LU-13030: pcc: auto_attach doesn't work after client cache cleared

Task

• LU-10171: define pct(a,b) once
• LU-11394: Updates for ZFS 0.8.0, no separate SPL
• LU-11683: Update ZFS Version to 0.7.12
• LU-11689: lfs project should proceed all dirs if hit errors
• LU-12026: verify that MDS stores atime/mtime/ctime during LSOM update
• LU-12561: Remove RHEL6/SLES11 support from master for 2.13

Technical task

• LU-10777: DoM performance is bad with FIO write
• LU-10894: Compilebench results are lower than expected
• LU-11298: LNet: Router peer instead of Router Peer NI
• LU-11299: LNet: Router pinger
• LU-11300: LNet: Router Aliveness and Health
• LU-11378: LNet: modify router selection to be MR aware
• LU-11421: DoM: manual migration OST-MDT, MDT-MDT
• LU-11470: LNet Health: Add drop all rule
• LU-11475: LNet discovery: allow deleting primary_nid in router case
• LU-11477: handle health for both incoming and outgoing messages
• LU-11478: LNet: discovery sequence numbers could be misleading
• LU-11551: LNet: Do not allow deletion fo rouer interface
• LU-11641: LNet Router: handle discovery off case
• LU-11649: lnet router: handle health off
• LU-11664: LNet Router: push updates when router interfaces change
• LU-12074: Update lustre tests
• LU-12163: Fix cpt confusion
• LU-12164: LNet: code cleanup
• LU-12200: LNet: MR Routing use peer_timeout
• LU-12201: LNet: detach response tracker
• LU-12237: LNet Health: Initialize recovery ping mdh
• LU-12249: LNet Health: list corruption
• LU-12254: lnet: correct discovery event queue release
• LU-12339: LNet Health: selecting loopback interface for sending
• LU-12344: LNet Health: tweak health errors
• LU-12526: Auto attach for PCC during IO

Question/Request

• LU-9581: recovery-small tests 52 60 remove from ALWAYS_EXCEPT
• LU-12298: Cannot Start Lustre 2.12.1 Server