Configuring the PAWN / SRB Connector

Before configuring PAWN to connect to the SRB, you should know the following:

• Connection settings for the SRB
• Resource that you will use in the SRB
• Destination collection in the SRB

Loading the driver

The SRB connector consists of two driver files, the jargon software from SDSC and the srb gateway itself. You will need to load these files before you can configure the SRB connector. You will only need to load these files once. Download these two files from below:

• pawn-srb.jar
• jargon.jar

Now use the PAWN Scheduler interface to load these driver files.

Click on 'Drivers' at the top of the scheduler interface then click the 'Add' button. This will prompt you for the name of the SRB driver. Enter 'edu.umiacs.pawn.resource.srb.SRBFactory' and click 'Add'. Now, select new SRB entry under 'Resource Drivers'.

For both of the files you downloaded click 'Add Jar' at the bottom of the screen, browse to where you selected the file and click open. You should see the two jar file listed above when you are finished.

Configuring an SRB Resource

Resources are assigned to domains that exist on the various management servers. Resource configuration is handled under the 'Producers' configuration. From the Producer's window, select the management server that will use the resource. You may see a list of other resources that have already been configured. To create a new SRB resource, click 'Add' at the bottom and type in the name of the resource along with the domain that will use this resource. The domain can be changed later if necesary. For driver, select the 'edu.umiacs.pawn.resource.srb.SRBFactory' driver if it isn't already selected. Click 'Next' when finished.

You will now see the Modify Resource window. This is where you enter any required connection information for the SRB. The fields are described below.

MCAT (Required)

The name of the MCAT server for your srb.

Zone (Required)

The zone of your srb.

Port (Required)

The port that your mcat runs on. The default of 5544 is the default that the SRB uses.

Low data port (Optional)

If your SRB is behind a firewall, it may have been configured to use a certin port range for data transfers. You should enter the low and high port that was configured. In most cases, leaving these blank is OK.

High data port (Optional)

See above

Version (Required)

The version of the SRB server that you are running. You should be able to leave this alone and Jargon will figure out the appropriate version. If you have problems connecting, you may need to set this.

There are two times the driver needs to connect to the SRB, first is to do the physical data transfer to the SRB and second is used by the client to browse the SRB and select a destination collection. For security reasons, you may want to specify two different accounts for these actions. The 'Publish account' and 'Client Browsing Account' allow you to do this. The Publish Account is required.

Publish Account Username (Required)

Account name in the SRB

Publish Account Domain (Required)

Domain for the account listed above

Publish Account Password (Required)

Password for the account. At this time, GSI logins are not supported by the driver.

Use Publish Settings

When the client connects, should it use the username and password listed above, or should it use a different account. If you uncheck this, you will need to supply an additional account below.

After you have entered all the required items, click the Transfer tab to configure the destination in the SRB where data will be ingested.

Resource (Required)

Select the resource that will be used to store data pushed into the SRB.

Base Directory (Required)

The directory that will be used as a starting location for clients when they choose where to save files in the SRB. You can click the folder icon on the right to browse the SRB.

Duplicate Files (Required)

Should duplicate files in the SRB be overwritten, or the new files renamed. If renamed is selected, then the new files will be loaded in to the SRB with a number appended depending on the pre-existing files (ie, file.1, file.2, etc..)

Record GUIDS

If this is checked the driver will record the global identifier for objects from the SRB rather than the srb path name. This is useful if you will be replicating files across multiple zones or referencing files by guid rather than path.

Manifest Destination

How should the driver handle the METS files that PAWN generates. By default, they are ignored and not pushed in to the SRB. If 'Inline' or 'Seperate' is selected the METS files will be pushed in addition to selected data. The METS files will also have the location of data and metadata objects updated to reflect the new SRB location.

• Ignore - do nothing with the METS files
• Inline - store the METS files in the directory they describe, ie, one file per directory.
• Seperate - store the METS files in one directory seperate from ingested data and metadata.
  • Base Directory - The directory that will be used as a starting location a clients browsing, or final destination METS files in the SRB. You can click the folder icon on the right to browse the SRB.
  • Prompt for destination - If selected the client will ask for the final destination in the SRB, otherwise the directory listed above is used.

When you are finished, click save. Now, anyone in the specified domain and with appropriate priveledge can load data from PAWN into the SRB.

-- MikeSmorul - 20 Jul 2006

to top