EV index self-healing?

Hello,

 

I have performed several DR tests with indexes. Especially I tested 2 things

- index volume loss (just root path of an index volume was kept, everything below I deleted)

- index volume restores from an older backup

It was index for an Exchange mailbox archive.

In both cases, when an end user performed a search in MS Outlook, he received an error. But it seems that this search attempt launched some automatic "self-heal" on the EV server, and after a few seconds, index volume was ok and up-to-date. (And there was nothing reported about this in EV Event Viewer.)

 

I think this feature is ok, but I did not find anything in documentation. In Admin Guide, there is something about automatic index metadata synchro, but I think it is something else.

Has anybody some comments to it?

 

Regards

Michal

 

 

 

 

1 Solution

Accepted Solutions
Highlighted
Accepted Solution!

I did find

I did find this, https://support.symantec.com/en_US/article.HOWTO56266.html. It talks about the synch queue
2.    Synchronization queue count
Synchronization queue count will contain the number of index volumes requiring any pending work discovered by storage via indexing admin service, Search or StoragePendingWorkDiscovery. 
 
        Synchronization Queue count is also updated by following processes:
          i.    Storage via the Index Admin Service when items are added updated or deleted in an EV archive.
          ii.    Search, if an index volume is searched for which the metadata in the directory database indicates that there is work pending.
 
        StoragePendingWorkDiscovery checks for any pending indexing work, EVIndexVolumesProcessor is then notified about any pending work by updating synchronization queue count. 
 
        StoragePendingWorkDiscovery runs as part of synchronization process, by default synchronization runs are configured as following:
          i.    Startup of indexing service
          ii.    1 hourly synchronization check
          iii.    6 hourly synchronization check
          iv.    10 hourly synchronization check

and this:

What are the different types of internal 64bit indexing synchronisation? http://www.symantec.com/docs/HOWTO59165
 

In Enteprise Vault (EV) 10, three major system synchronisation types run at regular intervals in the background:

  •  Hourly
  •  Failed volumes
  •  Full checks


The synchronisation schedule is configurable through the Vault Administration Console (VAC) as follows:

  1. Launch the VAC and navigate to the Enterprise Vault servers -> <EVServer> node.
  2. Right click and select the properties option.
  3. In the properties dialog, select the Advanced Tab and check for the following settings:
      • Frequency of checks for index volumes to process, default 1 hour
      • Frequency of checks for failed volumes, default 6 hours
      • Frequency of full checks for index volumes to process, default 10 hours

To modify the value:

  1. Select the setting and click on the Modify button.
  2. In the pop up dialog, enter the new values and click the OK button.
  3. Click on the OK button on the properties dialog and again click on the OK button to dismiss the warning dialog.
  4. Restart the services for the changes to be effective.

To reset back to the default value: 

  1. Select the setting and click on Modify button.
  2. In the pop up dialog,  click the Reset button followed by the OK button.
  3. Click on the OK button on the properties dialog and again click on the OK button to the dismiss warning dialog.
  4. Restart the services for the changes to be effective.

For more information on verifying the event viewer logs, see “How to verify the information reported in event views logs on internal indexing synchronisation?"

 
 
For the best O365 Reporting Solution click here Radar-Reporting

View solution in original post

2 Replies
Highlighted

Yes, that is expected.  When

Yes, that is expected.  When EV detects the index is out of date it will update it.  I know this is discussed somewhere but I don't have the reference at hand.

For the best O365 Reporting Solution click here Radar-Reporting
Highlighted
Accepted Solution!

I did find

I did find this, https://support.symantec.com/en_US/article.HOWTO56266.html. It talks about the synch queue
2.    Synchronization queue count
Synchronization queue count will contain the number of index volumes requiring any pending work discovered by storage via indexing admin service, Search or StoragePendingWorkDiscovery. 
 
        Synchronization Queue count is also updated by following processes:
          i.    Storage via the Index Admin Service when items are added updated or deleted in an EV archive.
          ii.    Search, if an index volume is searched for which the metadata in the directory database indicates that there is work pending.
 
        StoragePendingWorkDiscovery checks for any pending indexing work, EVIndexVolumesProcessor is then notified about any pending work by updating synchronization queue count. 
 
        StoragePendingWorkDiscovery runs as part of synchronization process, by default synchronization runs are configured as following:
          i.    Startup of indexing service
          ii.    1 hourly synchronization check
          iii.    6 hourly synchronization check
          iv.    10 hourly synchronization check

and this:

What are the different types of internal 64bit indexing synchronisation? http://www.symantec.com/docs/HOWTO59165
 

In Enteprise Vault (EV) 10, three major system synchronisation types run at regular intervals in the background:

  •  Hourly
  •  Failed volumes
  •  Full checks


The synchronisation schedule is configurable through the Vault Administration Console (VAC) as follows:

  1. Launch the VAC and navigate to the Enterprise Vault servers -> <EVServer> node.
  2. Right click and select the properties option.
  3. In the properties dialog, select the Advanced Tab and check for the following settings:
      • Frequency of checks for index volumes to process, default 1 hour
      • Frequency of checks for failed volumes, default 6 hours
      • Frequency of full checks for index volumes to process, default 10 hours

To modify the value:

  1. Select the setting and click on the Modify button.
  2. In the pop up dialog, enter the new values and click the OK button.
  3. Click on the OK button on the properties dialog and again click on the OK button to dismiss the warning dialog.
  4. Restart the services for the changes to be effective.

To reset back to the default value: 

  1. Select the setting and click on Modify button.
  2. In the pop up dialog,  click the Reset button followed by the OK button.
  3. Click on the OK button on the properties dialog and again click on the OK button to the dismiss warning dialog.
  4. Restart the services for the changes to be effective.

For more information on verifying the event viewer logs, see “How to verify the information reported in event views logs on internal indexing synchronisation?"

 
 
For the best O365 Reporting Solution click here Radar-Reporting

View solution in original post