Added isValid connection health check

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed

[he-is-harry]

• Added initial isValid(timeout) check which will return true if the client can send a request to the server and receive a reply within the timeout
  • timeout is given in seconds
  • The API's implementation can be updated once the ping protocol support is complete on the server side
• Updated Connection.send so that it can accept a communicationTimeout option to timeout a request to the server and return a timeout error
• Added unit and integration tests for isValid

[he-is-harry]

Here is a rough draft for the reconnect on timeout change: he-is-harry@bcdd7e0.

TODOs:

• Testing, the change is a rough draft so little cases were tested, so several more tests should be run to check that the queue always runs the newly enqueue tasks
  • Things to try: Multiple reconnects, affect of reconnect on transaction, pending tasks should currently receive an error in callback after timeout in current task
• Investigate the connection properties that should be used for reconnect (maybe _redirectHost / _redirectPort is better?)
• Adding a connection readyState for the 'ready for reconnect' state after timeouts would be helpful
• Currently reconnect does not support redirection, perhaps there should be an integration with ConnectionManager
  • The Client and ConnectionManager connect code could be moved into Connection as a refactoring change
• The added Connection.reset function can be refactored to reduce duplicate code in Connection constructor
• Packet count check can be removed