Complete the following procedure when you want to restore
a Search index in a multi-node environment where restarting individual
Search nodes must be avoided.
Before you begin
For information about how to create a backup
copy of the Search index, see Backing up the Search index.Note: IBM® Connections applications maintain
delete and access-control update information for a maximum of 30 days.
Indexes that are more than 30 days old are not considered suitable
for restoration because they might contain obsolete or orphan content.
About this task
You can use a backup copy of the Search index to restore
the index in the event of loss or corruption. You must restore the
backup consistently for all the nodes in your deployment.Procedure
To restore the Search index in the event of loss or corruption,
complete the following steps:
- Disable any regular indexing tasks that you
have configured.
- To list the indexing tasks, enter the following command:
SearchService.listIndexingTasks()
- To disable tasks, enter the following command:
SearchService.disableAllTasks()For
example:
SearchService.disableAllTasks()
There
is only one indexing task by default.
- To prepare the HOMEPAGE database to successfully load restored
indexes on each node, enter the following command:
SearchService.notifyRestore(Boolean
isNewIndex)where the isNewIndex parameter specifies
whether all entries are removed from the database table that is used
by the file content extraction process to track the status of individual
files. Set the parameter to false when you are restoring an index
backup.
For example:
SearchService.notifyRestore("false")
For
more information about this command, see Backing up the Search
index using wsadmin commands.
- On each Search node, delete the contents of the index directory
and all its subdirectories from the IBM Connections
Search data directory.
- On each Search node, copy the backup index and all its
subdirectories into the Search directory.
- On each Search node, reload the index using the following
command:
SearchService.reloadIndex()
- Re-enable your indexing task or tasks using the
SearchService.enableAllTasks() command.
For example:
SearchService.enableAllTasks()
Note: If you don't want to
enable all tasks (for example, if some tasks were disabled before you started these steps and you
want to keep them disabled), use the SearchService.enableTask(String taskName)
command instead to enable one task at a time.
The next indexing task to run resumes
indexing at the point at which the restored index was last successfully indexed.