[HBASE-22567] [HBCK2] Add new methods for dealing with missing regions in META while Master is online

• Type: New Feature
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.16.0
• Fix Version/s: hbase-operator-tools-1.0.0
• Component/s: hbck2
• Labels: hbck2

Release Note

Adds new, lightweight option for recovering missing regions in meta, "addFsRegionsMissingInMeta" command, as well "readonly" "reportMissingRegionsInMeta" to show the list of regions to be readded in meta. This is a less intrusive alternative to OfflineMetaRepair, can be used when master and meta are still capable of coming online.

Detailed description as provided by the command usage help:

{noformat}
 addFsRegionsMissingInMeta <NAMESPACE|NAMESPACE:TABLENAME>...
   Options:
    -d,--force_disable aborts fix for table if disable fails.
   To be used in scenarios where some regions may be missing in META,
   but there's still a valid 'regioninfo' metadata file on HDFS.
   This is a lighter version of 'OfflineMetaRepair tool commonly used for
   similar issues on 1.x release line.
   This command needs META to be online. For each table name passed as
   parameter, it performs a diff between regions available in META,
   against existing regions dirs on HDFS. Then, for region dirs with
   no matches in META, it reads regioninfo metadata file and
   re-creates given region in META. Regions are re-created in 'CLOSED'
   state at META table only, but not in Masters' cache, and are not
   assigned either. To get these regions online, run HBCK2
'assigns'command
   printed at the end of this command results for convenience.

   NOTE: If using hbase releases older than 2.3.0, a rolling restart of
   HMasters is needed prior to executing the provided 'assigns' command.

   An example adding missing regions for tables 'tbl_1' on default
   namespace, 'tbl_2' on namespace 'n1' and for all tables from
   namespace 'n2':
     $ HBCK2 addFsRegionsMissingInMeta default:tbl_1 n1:tbl_2 n2
   Returns HBCK2 'assigns' command with all re-inserted regions.
   SEE ALSO: reportMissingRegionsInMeta
...
 reportMissingRegionsInMeta <NAMESPACE|NAMESPACE:TABLENAME>...
   To be used in scenarios where some regions may be missing in META,
   but there's still a valid 'regioninfo metadata file on HDFS.
   This is a checking only method, designed for reporting purposes and
   doesn't perform any fixes, providing a view of which regions (if any)
   would get re-added to meta, grouped by respective table/namespace.
   To effectively re-add regions in meta, addFsRegionsMissingInMeta should be executed.
   This command needs META to be online. For each namespace/table passed
   as parameter, it performs a diff between regions available in META,
   against existing regions dirs on HDFS. Region dirs with no matches
   are printed grouped under its related table name. Tables with no
   missing regions will show a 'no missing regions' message. If no
   namespace or table is specified, it will verify all existing regions.
   It accepts a combination of multiple namespace and tables. Table names
   should include the namespace portion, even for tables in the default
   namespace, otherwise it will assume as a namespace value.
   An example triggering missing regions report for tables 'table_1'
   and 'table_2', under default namespace:
     $ HBCK2 reportMissingRegionsInMeta default:table_1 default:table_2
   An example triggering missing regions report for table 'table_1'
   under default namespace, and for all tables from namespace 'ns1':
     $ HBCK2 reportMissingRegionsInMeta default:table_1 ns1
   Returns list of missing regions for each table passed as parameter, or
   for each table on namespaces specified as parameter.
...
{noformat}

• Assignee: Wellington Chevreuil
• Reporter: Wellington Chevreuil
• Created: 11/Jun/19 16:59
• Updated: 13/Sep/19 21:04
• Resolved: 06/Sep/19 17:09

Description

Following latest discussion on HBASE-21745, this proposes an hbck2 command that allows for inserting back regions missing in META that still have regioninfo available in HDFS. Although this is still an interactive and simpler version than the old OfflineMetaRepair, it still relies on hdfs state as the source of truth, and performs META updates mostly independently from Master (apart from requiring Meta table been online).

For a more detailed explanation on this command behaviour, pasting command usage text:

To be used for scenarios where some regions may be missing in META,
but there's still a valid 'regioninfo' metadata file on HDFS.
This is a lighter version of 'OfflineMetaRepair' tool commonly used for
similar issues on 1.x release line.
This command needs META to be online. For each table name passed as
parameter, it performs a diff between regions available in META,
against existing regions dirs on HDFS. Then, for region dirs with
no matches in META, it reads regioninfo metadata file and
re-creates given region in META. Regions are re-created in 'CLOSED'
state at META table only, but not in Masters' cache, and are not
assigned either. A rolling Masters restart, followed by a
hbck2 'assigns' command with all re-inserted regions is required.
This hbck2 'assigns' command is printed for user convenience.
WARNING: To avoid potential region overlapping problems due to ongoing
splits, this command disables given tables while re-inserting regions.
An example adding missing regions for tables 'table_1' and 'table_2':
$ HBCK2 addMissingRegionsInMeta table_1 table_2
Returns hbck2 'assigns' command with all re-inserted regions.

Issue Links

• is related to: HBASE-22690 Deprecate / Remove OfflineMetaRepair in hbase-2+
• is related to: HBASE-22707 [HBCK2] MasterRpcServices assigns method should try to reload regions from meta if the passed regions isn't found under AssignmentManager RegionsStateStore
• relates to: HBASE-21966 Fix region holes, overlaps, and other region related errors
• relates to: HBASE-21745 Make HBCK2 be able to fix issues other than region assignment
• links to: GitHub Pull Request #18
• links to: HBCK2 addMissingRegionsToMeta PR1