Lustre 2.10.2 Changelog: Difference between revisions

From Lustre Wiki
Jump to navigation Jump to search
mNo edit summary
No edit summary
 
(2 intermediate revisions by one other user not shown)
Line 5: Line 5:
*RHEL 6.9 (2.6.32-696.16.1.el6)
*RHEL 6.9 (2.6.32-696.16.1.el6)
*RHEL 7.4 (3.10.0-693.5.2.el7)
*RHEL 7.4 (3.10.0-693.5.2.el7)
*SLES12 SP3 (4.4.92-6.18)
*SLES12 SP2 (4.4.92-6.18)
*Ubuntu 16.04 (4.4.0-71)
*Ubuntu 16.04 (4.4.0-98)


'''Recommended e2fsprogs version: '''<br />
'''Recommended e2fsprogs version: '''<br />
Line 15: Line 15:


== Issues fixed between 2.10.1 and 2.10.2 ==
== Issues fixed between 2.10.1 and 2.10.2 ==
'''New Features '''<br />
''' New Feature '''
*[https://jira.hpdd.intel.com/browse/LU-7990 LU-7990]: Large bulk IO support
*[https://jira.hpdd.intel.com/browse/LU-7990 LU-7990]: Large bulk IO support


'''Improvements '''<br />
''' Improvement '''
*[https://jira.hpdd.intel.com/browse/LU-7251 LU-7251]: reduce commit callbacks in OSP
*[https://jira.hpdd.intel.com/browse/LU-8276 LU-8276]: Make lru clear always discard read lock pages
*[https://jira.hpdd.intel.com/browse/LU-8276 LU-8276]: Make lru clear always discard read lock pages
*[https://jira.hpdd.intel.com/browse/LU-9417 LU-9417]: xattr cache memory usage can be reduced
*[https://jira.hpdd.intel.com/browse/LU-9417 LU-9417]: xattr cache memory usage can be reduced
*[https://jira.hpdd.intel.com/browse/LU-9462 LU-9462]: &quot;lfs setstripe&quot; doesn't support --component-add/del on a directory
*[https://jira.hpdd.intel.com/browse/LU-9462 LU-9462]: &quot;lfs setstripe&quot; doesn't support --component-add/del on a directory
*[https://jira.hpdd.intel.com/browse/LU-9946 LU-9946]: Have fileset on default nodemap
*[https://jira.hpdd.intel.com/browse/LU-9996 LU-9996]: include MOFED IB libs when build with lbuild
*[https://jira.hpdd.intel.com/browse/LU-9996 LU-9996]: include MOFED IB libs when build with lbuild
*[https://jira.hpdd.intel.com/browse/LU-10007 LU-10007]: Update health resource agents
*[https://jira.hpdd.intel.com/browse/LU-10007 LU-10007]: Update health resource agents
Line 29: Line 27:
*[https://jira.hpdd.intel.com/browse/LU-10141 LU-10141]: Integer overflow in llapi_changelog_start
*[https://jira.hpdd.intel.com/browse/LU-10141 LU-10141]: Integer overflow in llapi_changelog_start


'''Bugs '''<br />
''' Bug '''
*[https://jira.hpdd.intel.com/browse/LU-3719 LU-3719]: divide error in ldiskfs_mb_normalize_request on MDT
*[https://jira.hpdd.intel.com/browse/LU-3719 LU-3719]: divide error in ldiskfs_mb_normalize_request on MDT
*[https://jira.hpdd.intel.com/browse/LU-4134 LU-4134]: obdfilter-suvery bugs and panics (ioctl API isn't protected over shutdown/setup property).
*[https://jira.hpdd.intel.com/browse/LU-4705 LU-4705]: LustreError: 89827:0:(mdc_locks.c:916:mdc_enqueue()) ldlm_cli_enqueue: -2
*[https://jira.hpdd.intel.com/browse/LU-4705 LU-4705]: LustreError: 89827:0:(mdc_locks.c:916:mdc_enqueue()) ldlm_cli_enqueue: -2
*[https://jira.hpdd.intel.com/browse/LU-5024 LU-5024]: (mdc_lib.c:163:mdc_pack_name()) ASSERTION( cpy_len == name_len &amp;&amp; lu_name_is_valid_2(buf, cpy_len) ) failed:
*[https://jira.hpdd.intel.com/browse/LU-7802 LU-7802]: set_param lru_size fails with 'error: set_param: setting /proc/fs/lustre/ldlm/namespaces/lustre-OST0000-osc-*/lru_size=clear: Invalid argument'
*[https://jira.hpdd.intel.com/browse/LU-7802 LU-7802]: set_param lru_size fails with 'error: set_param: setting /proc/fs/lustre/ldlm/namespaces/lustre-OST0000-osc-*/lru_size=clear: Invalid argument'
*[https://jira.hpdd.intel.com/browse/LU-8497 LU-8497]: GPF after page allocation failure in osp_send_update_thread()
*[https://jira.hpdd.intel.com/browse/LU-8497 LU-8497]: GPF after page allocation failure in osp_send_update_thread()
Line 39: Line 35:
*[https://jira.hpdd.intel.com/browse/LU-8578 LU-8578]: ptlrpcd threads hung waiting for cl_lock held by osc_extent_wait() process
*[https://jira.hpdd.intel.com/browse/LU-8578 LU-8578]: ptlrpcd threads hung waiting for cl_lock held by osc_extent_wait() process
*[https://jira.hpdd.intel.com/browse/LU-9140 LU-9140]: sanityn test_77j test_77k: The read / write rate exceeds x% of preset rate
*[https://jira.hpdd.intel.com/browse/LU-9140 LU-9140]: sanityn test_77j test_77k: The read / write rate exceeds x% of preset rate
*[https://jira.hpdd.intel.com/browse/LU-9158 LU-9158]: sanity-quota, test_12b: rebalancing failed
*[https://jira.hpdd.intel.com/browse/LU-9416 LU-9416]: lmv_hsm_ct_register() should add kkuc before sending registration RPCs
*[https://jira.hpdd.intel.com/browse/LU-9416 LU-9416]: lmv_hsm_ct_register() should add kkuc before sending registration RPCs
*[https://jira.hpdd.intel.com/browse/LU-9469 LU-9469]: conf-sanity test_61: test failed to respond and timed out
*[https://jira.hpdd.intel.com/browse/LU-9469 LU-9469]: conf-sanity test_61: test failed to respond and timed out
*[https://jira.hpdd.intel.com/browse/LU-9509 LU-9509]: parallel-scale-stress-hw_compilebench test stuck: timeout 2hours, usually takes &lt; 40 minutes.
*[https://jira.hpdd.intel.com/browse/LU-9530 LU-9530]: Interop 2.9&lt;-&gt;master sanity test_184a: swap of file layout failed
*[https://jira.hpdd.intel.com/browse/LU-9574 LU-9574]: Large file read performance degradation from multiple OST's
*[https://jira.hpdd.intel.com/browse/LU-9574 LU-9574]: Large file read performance degradation from multiple OST's
*[https://jira.hpdd.intel.com/browse/LU-9578 LU-9578]: having security_dentry_init_security doesn't mean it's enabled
*[https://jira.hpdd.intel.com/browse/LU-9578 LU-9578]: having security_dentry_init_security doesn't mean it's enabled
*[https://jira.hpdd.intel.com/browse/LU-9611 LU-9611]: fix default stripe count/offset proc tunables
*[https://jira.hpdd.intel.com/browse/LU-9660 LU-9660]: reduce ptlrpcd wakeups on idle system
*[https://jira.hpdd.intel.com/browse/LU-9663 LU-9663]: sanity test_103a: LU-2561 test failed
*[https://jira.hpdd.intel.com/browse/LU-9663 LU-9663]: sanity test_103a: LU-2561 test failed
*[https://jira.hpdd.intel.com/browse/LU-9672 LU-9672]: gss: fix expiration time of sunrpc cache
*[https://jira.hpdd.intel.com/browse/LU-9672 LU-9672]: gss: fix expiration time of sunrpc cache
*[https://jira.hpdd.intel.com/browse/LU-9682 LU-9682]: Nodemap ip range missing if delete another nodemap the same range
*[https://jira.hpdd.intel.com/browse/LU-9682 LU-9682]: Nodemap ip range missing if delete another nodemap the same range
*[https://jira.hpdd.intel.com/browse/LU-9733 LU-9733]: changelog_register return failure
*[https://jira.hpdd.intel.com/browse/LU-9741 LU-9741]: sanity test 27w failed with “/usr/bin/lfs getstripe -d /mnt/lustre/d27w.sanity failed”
*[https://jira.hpdd.intel.com/browse/LU-9741 LU-9741]: sanity test 27w failed with “/usr/bin/lfs getstripe -d /mnt/lustre/d27w.sanity failed”
*[https://jira.hpdd.intel.com/browse/LU-9752 LU-9752]: Unable to format zfs osts
*[https://jira.hpdd.intel.com/browse/LU-9752 LU-9752]: Unable to format zfs osts
Line 59: Line 49:
*[https://jira.hpdd.intel.com/browse/LU-9840 LU-9840]: LU-3529 causes 25% metadata performance regressions even without DNE
*[https://jira.hpdd.intel.com/browse/LU-9840 LU-9840]: LU-3529 causes 25% metadata performance regressions even without DNE
*[https://jira.hpdd.intel.com/browse/LU-9870 LU-9870]: rpms fail to build when SNMP is missing
*[https://jira.hpdd.intel.com/browse/LU-9870 LU-9870]: rpms fail to build when SNMP is missing
*[https://jira.hpdd.intel.com/browse/LU-9872 LU-9872]: parallel-scale-nfsv3 test_connectathon: connectathon failed: 1
*[https://jira.hpdd.intel.com/browse/LU-9887 LU-9887]: sanity-lfsck test_9a: FAIL: (4) Got speed 952, expected less than 144
*[https://jira.hpdd.intel.com/browse/LU-9888 LU-9888]: conf-sanity test_32b: test 32b failed with 1
*[https://jira.hpdd.intel.com/browse/LU-9888 LU-9888]: conf-sanity test_32b: test 32b failed with 1
*[https://jira.hpdd.intel.com/browse/LU-9894 LU-9894]: lfsck always prints that it started all types of lfsck
*[https://jira.hpdd.intel.com/browse/LU-9908 LU-9908]: conf-sanity test_41b: test failed to respond and timed out
*[https://jira.hpdd.intel.com/browse/LU-9908 LU-9908]: conf-sanity test_41b: test failed to respond and timed out
*[https://jira.hpdd.intel.com/browse/LU-9929 LU-9929]: Use &quot;setfacl&quot; to set &quot;default&quot; setting fail when nodemap enabled
*[https://jira.hpdd.intel.com/browse/LU-9929 LU-9929]: Use &quot;setfacl&quot; to set &quot;default&quot; setting fail when nodemap enabled
Line 68: Line 55:
*[https://jira.hpdd.intel.com/browse/LU-9941 LU-9941]: lsm_is_composite() isn't right
*[https://jira.hpdd.intel.com/browse/LU-9941 LU-9941]: lsm_is_composite() isn't right
*[https://jira.hpdd.intel.com/browse/LU-9951 LU-9951]: check whether posix_acl_update_mode exists
*[https://jira.hpdd.intel.com/browse/LU-9951 LU-9951]: check whether posix_acl_update_mode exists
*[https://jira.hpdd.intel.com/browse/LU-9966 LU-9966]: sanity test_411: fail to trigger a memory allocation error
*[https://jira.hpdd.intel.com/browse/LU-9956 LU-9956]: kernel upgrade [SLES12 SP3 4.4.82-6.3]
*[https://jira.hpdd.intel.com/browse/LU-9967 LU-9967]: sanity test_60a: Module llog_test is in use
*[https://jira.hpdd.intel.com/browse/LU-9978 LU-9978]: kernel update [RHEL7.4 3.10.0-693.2.2.el7]
*[https://jira.hpdd.intel.com/browse/LU-9978 LU-9978]: kernel update [RHEL7.4 3.10.0-693.2.2.el7]
*[https://jira.hpdd.intel.com/browse/LU-9980 LU-9980]: save_lustre_params() defect
*[https://jira.hpdd.intel.com/browse/LU-9980 LU-9980]: save_lustre_params() defect
*[https://jira.hpdd.intel.com/browse/LU-9997 LU-9997]: Suspicious assert check in ldlm_cli_cancel
*[https://jira.hpdd.intel.com/browse/LU-10000 LU-10000]: kernel update [SLES12 SP2 4.4.74-92.38]
*[https://jira.hpdd.intel.com/browse/LU-10011 LU-10011]: suppress annoying messages for lfs quota
*[https://jira.hpdd.intel.com/browse/LU-10015 LU-10015]: ASSERTION( cmid-&gt;device != ((void *)0) ) failed
*[https://jira.hpdd.intel.com/browse/LU-10015 LU-10015]: ASSERTION( cmid-&gt;device != ((void *)0) ) failed
*[https://jira.hpdd.intel.com/browse/LU-10023 LU-10023]: kernel update [SLES12 SP3 4.4.82-6.9]
*[https://jira.hpdd.intel.com/browse/LU-10023 LU-10023]: kernel update [SLES12 SP3 4.4.82-6.9]
Line 80: Line 65:
*[https://jira.hpdd.intel.com/browse/LU-10047 LU-10047]: sanity test_102d: mkdir -i 2 -c2 /mnt/lustre/d102d.sanity failed
*[https://jira.hpdd.intel.com/browse/LU-10047 LU-10047]: sanity test_102d: mkdir -i 2 -c2 /mnt/lustre/d102d.sanity failed
*[https://jira.hpdd.intel.com/browse/LU-10086 LU-10086]: LNET_MINOR conflicts with USERIO_MINOR
*[https://jira.hpdd.intel.com/browse/LU-10086 LU-10086]: LNET_MINOR conflicts with USERIO_MINOR
*[https://jira.hpdd.intel.com/browse/LU-10089 LU-10089]: kiblnd_fmr_pool_map() Failed to map mr 10/11 elements
*[https://jira.hpdd.intel.com/browse/LU-10088 LU-10088]: kernel update [RHEL6.9 2.6.32-696.13.2.el6]
*[https://jira.hpdd.intel.com/browse/LU-10098 LU-10098]: Lustre resource agent can mis-identify resource mounts
*[https://jira.hpdd.intel.com/browse/LU-10119 LU-10119]: systemd Failed at step EXEC spawning /usr/sbin/lustre_routes_config: Exec format error
*[https://jira.hpdd.intel.com/browse/LU-10119 LU-10119]: systemd Failed at step EXEC spawning /usr/sbin/lustre_routes_config: Exec format error
*[https://jira.hpdd.intel.com/browse/LU-10131 LU-10131]: Update inode attributes on unlink
*[https://jira.hpdd.intel.com/browse/LU-10131 LU-10131]: Update inode attributes on unlink
*[https://jira.hpdd.intel.com/browse/LU-10132 LU-10132]: IO execvp errors 2.10 client/EE3.1.1 server
*[https://jira.hpdd.intel.com/browse/LU-10132 LU-10132]: IO execvp errors 2.10 client/EE3.1.1 server
*[https://jira.hpdd.intel.com/browse/LU-10142 LU-10142]: kernel update [RHEL7.4 3.10.0-693.5.2.el7]
*[https://jira.hpdd.intel.com/browse/LU-10142 LU-10142]: kernel update [RHEL7.4 3.10.0-693.5.2.el7]
*[https://jira.hpdd.intel.com/browse/LU-10161 LU-10161]: lbuild: Failed to determine the default route interface
*[https://jira.hpdd.intel.com/browse/LU-10164 LU-10164]: kernel update [SLES12 SP3 4.4.92-6.18]
*[https://jira.hpdd.intel.com/browse/LU-10164 LU-10164]: kernel update [SLES12 SP3 4.4.92-6.18]
*[https://jira.hpdd.intel.com/browse/LU-10166 LU-10166]: Changelog reader crashes when closing the changelog
*[https://jira.hpdd.intel.com/browse/LU-10166 LU-10166]: Changelog reader crashes when closing the changelog
*[https://jira.hpdd.intel.com/browse/LU-10168 LU-10168]: kernel update [SLES12 SP2 4.4.90-92.45]
*[https://jira.hpdd.intel.com/browse/LU-10204 LU-10204]: ladvise_willread_performance() function misses line continuation slash
*[https://jira.hpdd.intel.com/browse/LU-10204 LU-10204]: ladvise_willread_performance() function misses line continuation slash
*[https://jira.hpdd.intel.com/browse/LU-10217 LU-10217]: Sanity 248 fails with newer dd output
*[https://jira.hpdd.intel.com/browse/LU-10220 LU-10220]: Incorrect record length allocated in mdd_changelog_data_store_by_fid()
*[https://jira.hpdd.intel.com/browse/LU-10220 LU-10220]: Incorrect record length allocated in mdd_changelog_data_store_by_fid()
*[https://jira.hpdd.intel.com/browse/LU-10234 LU-10234]: tests: sanity-hsm: tests 34,35,36 are not using wait correctly
*[https://jira.hpdd.intel.com/browse/LU-10234 LU-10234]: tests: sanity-hsm: tests 34,35,36 are not using wait correctly
*[https://jira.hpdd.intel.com/browse/LU-10241 LU-10241]: kernel update [RHEL6.9 2.6.32-696.16.1.el6]
*[https://jira.hpdd.intel.com/browse/LU-10241 LU-10241]: kernel update [RHEL6.9 2.6.32-696.16.1.el6]
*[https://jira.hpdd.intel.com/browse/LU-10330 LU-10330]: Interop sanity test_101g: unable to set max_pages_per_rpc=4M


'''Technical Tasks''' <br />
''' Task '''
*[https://jira.hpdd.intel.com/browse/LU-10150 LU-10150]: Update ZFS Version to 0.7.3
*[https://jira.hpdd.intel.com/browse/LU-10150 LU-10150]: Update ZFS Version to 0.7.3


'''Question/Request''' <br />
''' Question/Request '''
*[https://jira.hpdd.intel.com/browse/LU-9963 LU-9963]: add parallel-scale test_parallel_grouplock to ALWAYS_EXCEPT list
*[https://jira.hpdd.intel.com/browse/LU-9963 LU-9963]: add parallel-scale test_parallel_grouplock to ALWAYS_EXCEPT list


[[Category:Changelog]]
[[Category:Changelog]]

Latest revision as of 08:20, 25 April 2018

Server support for kernels:

  • RHEL 7.4 (3.10.0-693.5.2.el7)

Client support for unpatched kernels:

  • RHEL 6.9 (2.6.32-696.16.1.el6)
  • RHEL 7.4 (3.10.0-693.5.2.el7)
  • SLES12 SP2 (4.4.92-6.18)
  • Ubuntu 16.04 (4.4.0-98)

Recommended e2fsprogs version:

  • v1.42.13.wc6

Interoperability Support:

  • Clients & Servers: Latest 2.9.X

Issues fixed between 2.10.1 and 2.10.2

New Feature

Improvement

  • LU-8276: Make lru clear always discard read lock pages
  • LU-9417: xattr cache memory usage can be reduced
  • LU-9462: "lfs setstripe" doesn't support --component-add/del on a directory
  • LU-9996: include MOFED IB libs when build with lbuild
  • LU-10007: Update health resource agents
  • LU-10041: osd-zfs to choose dnode size
  • LU-10141: Integer overflow in llapi_changelog_start

Bug

  • LU-3719: divide error in ldiskfs_mb_normalize_request on MDT
  • LU-4705: LustreError: 89827:0:(mdc_locks.c:916:mdc_enqueue()) ldlm_cli_enqueue: -2
  • LU-7802: set_param lru_size fails with 'error: set_param: setting /proc/fs/lustre/ldlm/namespaces/lustre-OST0000-osc-*/lru_size=clear: Invalid argument'
  • LU-8497: GPF after page allocation failure in osp_send_update_thread()
  • LU-8541: change lru_max_age to units of ms
  • LU-8578: ptlrpcd threads hung waiting for cl_lock held by osc_extent_wait() process
  • LU-9140: sanityn test_77j test_77k: The read / write rate exceeds x% of preset rate
  • LU-9416: lmv_hsm_ct_register() should add kkuc before sending registration RPCs
  • LU-9469: conf-sanity test_61: test failed to respond and timed out
  • LU-9574: Large file read performance degradation from multiple OST's
  • LU-9578: having security_dentry_init_security doesn't mean it's enabled
  • LU-9663: sanity test_103a: LU-2561 test failed
  • LU-9672: gss: fix expiration time of sunrpc cache
  • LU-9682: Nodemap ip range missing if delete another nodemap the same range
  • LU-9741: sanity test 27w failed with “/usr/bin/lfs getstripe -d /mnt/lustre/d27w.sanity failed”
  • LU-9752: Unable to format zfs osts
  • LU-9782: High CPU usage with random IO test.
  • LU-9802: Endianness swap issue for field lcm_entry_count
  • LU-9814: prealloc table change don't work after 2.10 upgrade
  • LU-9840: LU-3529 causes 25% metadata performance regressions even without DNE
  • LU-9870: rpms fail to build when SNMP is missing
  • LU-9888: conf-sanity test_32b: test 32b failed with 1
  • LU-9908: conf-sanity test_41b: test failed to respond and timed out
  • LU-9929: Use "setfacl" to set "default" setting fail when nodemap enabled
  • LU-9930: "BUG: unable to handle kernel NULL pointer dereference at 0000000000000088" in ll_statahead_thread()
  • LU-9941: lsm_is_composite() isn't right
  • LU-9951: check whether posix_acl_update_mode exists
  • LU-9956: kernel upgrade [SLES12 SP3 4.4.82-6.3]
  • LU-9978: kernel update [RHEL7.4 3.10.0-693.2.2.el7]
  • LU-9980: save_lustre_params() defect
  • LU-10000: kernel update [SLES12 SP2 4.4.74-92.38]
  • LU-10015: ASSERTION( cmid->device != ((void *)0) ) failed
  • LU-10023: kernel update [SLES12 SP3 4.4.82-6.9]
  • LU-10029: project inherit attribute should be removeable
  • LU-10040: nodemap and quota issues (ineffective GID mapping)
  • LU-10047: sanity test_102d: mkdir -i 2 -c2 /mnt/lustre/d102d.sanity failed
  • LU-10086: LNET_MINOR conflicts with USERIO_MINOR
  • LU-10088: kernel update [RHEL6.9 2.6.32-696.13.2.el6]
  • LU-10119: systemd Failed at step EXEC spawning /usr/sbin/lustre_routes_config: Exec format error
  • LU-10131: Update inode attributes on unlink
  • LU-10132: IO execvp errors 2.10 client/EE3.1.1 server
  • LU-10142: kernel update [RHEL7.4 3.10.0-693.5.2.el7]
  • LU-10164: kernel update [SLES12 SP3 4.4.92-6.18]
  • LU-10166: Changelog reader crashes when closing the changelog
  • LU-10168: kernel update [SLES12 SP2 4.4.90-92.45]
  • LU-10204: ladvise_willread_performance() function misses line continuation slash
  • LU-10220: Incorrect record length allocated in mdd_changelog_data_store_by_fid()
  • LU-10234: tests: sanity-hsm: tests 34,35,36 are not using wait correctly
  • LU-10241: kernel update [RHEL6.9 2.6.32-696.16.1.el6]

Task

Question/Request

  • LU-9963: add parallel-scale test_parallel_grouplock to ALWAYS_EXCEPT list