On the contrary C++ SyncCompanion states that:
SyncYieldCycles does not cause HotSync Manager to check whether the user has clicked the Cancel button, nor does it send keep-alive messages to the handheld to maintain connection with it during long periods in which your conduit is not accessing it.
So maybe the jsync version also sends these keep-alive messages to the handheld in addition to yielding cycles to the progress display?
/Dan
Dan BjÃrkgren wrote:
Clark,
Could it be you need to call SyncYieldCycles? This based on the javadoc for SyncManager.yieldCycles in jsync:
/**
* Conduit uses to keep connection to hand-held device from timing out. Every 7 seconds
* HS must talk to device (the hand-held device must hear from PC
* or it will close the connection). This function breaks the conduits hold
* on the queue and allows PC to call the hand-held device, keeping the connection
* open.
*
* @param wMaxMiliSecs
* @exception palm.conduit.SyncException
*/
This behaviour is not documented in the C++ conduit reference, but I doubt it would only apply to jsync conduits?
/Dan
Clark Dorman wrote:
Thank you for the response. There is definitely bloat in the pdb but I believe that it is not the issue. The size of the database even when bloated should not be too large for the memory. It might be related though. The error I get is: Install synchronization failed HotSync Error: Connection is lost (6410).
This is the same as the errors reported in the links above.
-- For information on using the Palm Developer Forums, or to unsubscribe, please see http://www.palmos.com/dev/support/forums/
