The ‘Pull’ API implements caching to protect the server from many clients making concurrent ‘pull’ checks.
- Every 5 seconds the API makes one database call to get a list of all the ‘transport target endpoints’ which have pending files.
Note : The same ‘organization endpoint’ can participate in multiple transports.
- When the first file is given the cache entry is removed, so in effect the client can get one file every 5 seconds from a single ‘‘transport target endpoint’.
- The default logic is to pull ‘all pending files’ and then exit (to override the default and pull a single file use -n argument – it is useful if the client script wants to do special processing with each file).
- In order to make sure all pending files are given in one execution the client waits up to 6 seconds for one ‘cache refresh’ before determining that are no longer pending files.