Configuring Optimization Features : Configuring CIFS prepopulation
  
Configuring CIFS prepopulation
You enable prepopulation and add, modify, and delete prepopulation shares in the Optimization > Protocols: CIFS Prepopulation page.
The prepopulation operation effectively performs the first SteelHead read of the data on the prepopulation share. Later, the SteelHead handles read and write requests as effectively as with a warm data store. With a warm data store, RiOS sends data references along with new or modified data, dramatically increasing the rate of data transfer over the WAN.
The first synchronization, or the initial copy, retrieves data from the origin file server and copies it to the RiOS data store on the SteelHead. Subsequent synchronizations are based on the synchronization interval.
The RiOS 8.5 and later Management Consoles include policies and rules to provide more control over which files the system transfers to warm the RiOS data store. A policy is a group of rules that select particular files to prepopulate. For example, you can create a policy that selects all PDF files larger than 300 MB created since January 1, 2017.
CIFS Prepopulation is disabled by default.
The AWS Cloud Accelerator doesn’t support CIFS Prepopulation. The Cloud Accelerator supports CIFS Prepopulation if it is deployed with WCCP or PBR (not with the Discovery Agent).
To enable CIFS prepopulation and add, modify, or delete a prepopulation share
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
CIFS Prepopulation page
2. Under Settings, complete the configuration as described in this table.
Control
Description
Enable CIFS Prepopulation
Prewarms the RiOS data store. In this setup, the primary interface of the SteelHead acts as a client and prerequests data from the share you want to use to warm the data store. This request goes through the LAN interface to the WAN interface out to the server-side SteelHead, causing the in-path interface to see the data as a normal client request.
When data is requested again by a client on the local LAN, RiOS sends only new or modified data over the WAN, dramatically increasing the rate of data transfers.
Enable Transparent Prepopulation Support using RCU
Opens port 8777 to allow manual warming of the RiOS data store using the Riverbed Copy Utility (RCU) to prepopulate your shares.
Most environments don’t need to enable RCU.
3. Click Apply to apply your settings.
4. When prepopulation is enabled, you can add shares and schedule automatic unattended synchronization as described in this table.
Control
Description
Add a Prepopulation Share
Displays the controls for adding a new prepopulation share.
Remote Path
Specify the path to the data on the origin server or the Universal Naming Convention (UNC) path of a share to which you want to make available for prepopulation.
Set up the prepopulation share on the remote box pointing to the actual share in the headend data center server. For example:
\\<origin file server>\<local name>
The share and the origin-server share names can’t contain any of these characters:
< > * ? | / + = ; : " , & []
Username
Specify the username of the local administrator account used to access the origin server.
The Windows account needs read-only permissions to the share location you specify in the Remote Path field.
Password
Specify the password for the local administrator account.
Comment
Optionally, include a comment to help you administer the share in the future. Comments can’t contain an ampersand (&).
Sync Time Limit
Specify a time limit that the synchronization job shouldn’t exceed.
Use this time format: H:M:S
Examples:
1 = 1 second
1:2 = 1 minute and 2 seconds
1:2:3 = 1 hour, 2 minutes, and 3 seconds
Sync Size Limit
Specify a limit on the amount of data in the synchronization job and select either MB or GB from the drop-down list. The default is MB.
Sync Using
Select either current files for syncing or use the latest share snapshot (if no snapshots are available, the system uses the current files).
Enable Scheduled Synchronization
Enables subsequent synchronization jobs after the initial synchronization. Select the type of synchronization the system performs after the initial synchronization.
Incremental Sync—The origin-file server retrieves only new data that was modified or created after the previous synchronization and sends it to the SteelHead.
Full Sync—The origin-file server retrieves all data and sends it to the SteelHead. Full synchronization is useful on SteelHeads that are frequently evicting the prepopulated data from the RiOS data store because of limited memory.
If the schedule for a full synchronization and an incremental synchronization coincide, the system performs a full synchronization.
Start Date/Time—Specify a base start date and time from which to schedule synchronizations.
Recurring Every—Specify how frequently scheduled synchronizations should occur relative to the start date and time. Leave blank to run the synchronization once.
Add Prepopulation Share
Adds the share to the Prepopulation Share list.
5. Click Save to Disk to save your settings permanently.
After you add a share, the CIFS prepopulation page includes the share in the Share table. The Share table provides an editable list of shares along with each share’s remote pathname, the date and time the next synchronizations will occur, the status, and any comments about the share.
When the status reports that the share has an error, mouse over the error to reveal details about the error.
Editing a prepopulation share
After adding a CIFS prepopulation share, you can edit it from the Configuration tab. You can create a policy (group of rules) to apply to the share, and you can schedule a date and time for share synchronization.
To edit a CIFS prepopulation share
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
2. Select the remote path for the share.
3. Select the Configuration tab.
4. Under Settings, complete the configuration as described in this table.
Control
Description
Remote Path
Specifies the path to the data on the origin server or the UNC path of a share available for prepopulation. This control is not editable under the share Configuration tab.
Username
Specify the username of the local administrator account used to access the origin server.
Change Password
Select the check box and then specify the password for the local administrator account.
Comment
Optionally, include a comment to help you administer the share in the future. Comments can’t contain an ampersand (&).
Sync Time Limit
Specify a time limit that the synchronization job shouldn’t exceed.
Use this time format: H:M:S
Examples:
1 = 1 second
1:2 = 1 minute and 2 seconds
1:2:3 = 1 hour, 2 minutes, and 3 seconds
Sync Size Limit
Specify a limit on the amount of data in the synchronization job.
Sync Using
Select to synchronize the current files or select the latest share snapshot (if no snapshots are available, the system uses the current files).
5. Click Apply to apply your configuration.
6. Click Save to Disk to save your settings permanently.
To add a CIFS prepopulation policy
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
2. Select the remote path for the share.
3. Select the Configuration tab.
4. Click Add a Policy.
5. Complete the configuration as described in this table.
Control
Description
Add a Policy
Displays the controls to add a policy. A policy is a group of rules applied to a share. There are no limits on the number of policies or the number of rules within a policy.
A file needs to pass every rule in only one policy to be selected for synchronization.
An empty policy with no rules selects everything.
RiOS doesn’t validate rules or policies; use caution to avoid including or excluding everything.
Policy Name
Specify a name for the policy.
Description
Describe the policy.
Add Rule
Click to add a new rule to a policy. You can add rules that prepopulate the RiOS data store according to filename, file size, or the time of the last file access, creation, or modification.
Synchronize files that match all of the following rules
Select a filter from the drop-down list and type or select a value for the rule from the drop-down list. The control changes dynamically according to the rule type.
Examples:
Select all TXT and PDF files:
File extension or name matches *.txt; *.PDF
Select all files that have been modified within the last two hours:
Modify time is within, when syncing 02:00:00
Select all TXT files larger than 300 MB and created since Jan 1st, 2013:
File size is greater than 300 MB
File extension/name matches *.txt
Creation Time is newer than 2013/01/01 00:00:00
Use the mouse to hover over the information icon for a tool tip about the filter.
To delete a rule, click the red x.
Add Policy
Adds the policy to the policy list.
6. Click Apply to apply your configuration.
7. Click Save to Disk to save your settings permanently.
To schedule a synchronization
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
2. Select the remote path for the share.
3. Select the Configuration tab.
4. Complete the configuration as described in this table.
Control
Description
Enable Scheduled Synchronization
Enables subsequent synchronization jobs after the initial synchronization. Select the type of synchronization the system performs after the initial synchronization.
Incremental Sync—The origin file server retrieves only new data that was modified or created after the previous synchronization and sends it to the SteelHead.
Full Sync—The origin file server retrieves all data and sends it to the SteelHead. Full synchronization is useful on SteelHeads that are frequently evicting the prepopulated data from the RiOS data store because of limited memory.
If the schedule for a full synchronization and an incremental synchronization coincide, the system performs a full synchronization.
Start Date/Time—Specify a base start date and time from which to schedule synchronizations.
Recurring Every—Specify how frequently scheduled synchronizations should occur relative to the start date and time. Leave blank to synchronize once.
5. Click Apply to apply your configuration.
6. Click Save to Disk to save your settings permanently.
Performing CIFS prepopulation share operations
After adding a CIFS prepopulation share, you can synchronize the share or perform a dry run of what would be synchronized.
To perform an operation on a CIFS prepopulation share
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
2. Select the remote path for the share.
3. Select the Operations tab.
4. Click a button to perform an operation on a share as described in this table. You can perform only one operation at a time.
Operation
Description
Sync Now
Synchronizes the share using the current settings.
Perform Dry Run
Creates a log of what would be synchronized using the current settings, without actually synchronizing anything.
Cancel Operation
Cancels the operation.
Viewing CIFS prepopulation share logs
After adding a CIFS prepopulation share, you can view CIFS prepopulation share logs to see more detail regarding recent synchronizations, the initial copy of the share, or the last share synchronization.
To view CIFS prepopulation share logs
1. Choose Optimization > Protocols: CIFS Prepopulation to display the CIFS Prepopulation page.
2. Select the remote path for the share.
3. Select the Operations tab.
4. Click one of these links to view a log file.
Log File
Description
Recent syncs
Contains logs for the last few share synchronizations. The log includes how many directories, files, and bytes were received and how long it took to receive them. The log also lists any errors or deletions.
Initial sync
Includes how many directories, files, and bytes were received initially and how long it took to receive them. The log also lists any errors or deletions.
Last dry run
Includes a log of what would have been synchronized with the current share configuration, without actually synchronizing anything.
To print the report, choose File > Print in your web browser to open the Print dialog box.
Related topics
Configuring CIFS optimization