The compare option looks at a snapshot of all remote sites and compares the names, addresses and balances at that point in time with the main server. All differences are noted, you can then use the Bounce or Extract options to make the necessary corrections.

This is a three step process,

1. Extract the data

   • Manually - use xmenu options:

     -> a) Utilities
     -> j) Compare data with remote sites
     -> a) Run xud program
     

     This must be run on all machines, it will startup a program that quickly extracts the client name and current balance to a file.

     Note: If you connected via the modem to the remote site you do NOT need to stay on-line till the program has finished.

   • Automatically - if you want to generate this list on a regular basis add an entry to vet's crontab such as

        10 4 25 * * /usr/local/bin/vet -p xud
     

     (This example will run on the 25th of every month at 4:10 am). You can use the Automation Manager to set this up for you.

2. Let the machines talk

   The file generated by the above option will be transferred over to the main machine the next time the send utility ( x_send) is run. It will become visible when the checking program ( x_check) runs on the main system.

   You can check if all the files have transferred by using xmenu option:

   -> a) Utilities
   -> j) Compare data with remote sites
   -> b) View current xud files.
   

   This will list the available files, called xud.site-number e.g. branch 1 will be xud.1, branch 3 will be xud.3. Included in the list will be the date the file was received on the local machine and the date the run took place.

   You should be able to tell from that listing if all the files are current and available for the next step. If not, either wait a bit longer or force a transfer of data.

3. Compare the data.

   Again via xmenu option:

   -> a) Utilities
   -> j) Compare data with remote sites
   

   -> c) Check for balances or
   -> d) Check addresses
   

   Each option should be self explanatory. You will be asked if you want to send the results to the screen or printer (Unix default printer only).

   Note: The program has to store all the data in memory while it does the comparison, this will not be a problem for most of you, you should be aware, on low memory machines it may take a while.

   The system will only show records where there is a difference. Where differences are found an option within the vet system can be used to make corrections.

4. Correcting Mistakes or Errors

   Note: Before you jump in and start correcting mistakes check that there are no logfiles waiting to be processed, or there are lock files and other genuine reasons for the differences. Use xshow to check.

   Depending on the errors reported you have a few options:

   • Manually edit
     On the site that is wrong, switch off logging, make the adjustments and switch logging back on.

   • Extract Work
     If it looks like a whole logfile has disappeared then you should look at the extract option in Premvet 5.

   • Bounce Work
     If it is an odd animal here and there then you should look at the bounce option in Premvet 5.

   • Every thing
     If it appears that a large number of records have problems, you should consider transferring the whole database over.