Petals ESB Container

Service Assembly lifecycle should mirror as much as possible SU lifecycle

Details

  • Type: Improvement Request Improvement Request
  • Status: Resolved Resolved
  • Priority: Minor Minor
  • Resolution: Fixed
  • Affects Version/s: 5.0.0
  • Fix Version/s: 5.0.1
  • Security Level: Public
  • Description:
    Hide

    The JBI specification clearly specifies that if all of the SUs of an SA fail to deploy, then the SA fails to deploy, but if at least one of the SUs does not fail, then the SA is properly deployed.

    The JBI specification is not clear for the other operations so we make the following choices:

    • We mirror the same behaviour for start and init.
    • We simply log errors for stop and shutdown.

    The idea is to be able to report enough informations to the administrators when possible while not preventing to undeploy an SA when there is problems.

    For example, a common use case such as starting an SA with only one SU that fails simply fail (while before it simply silently failed...)

    Show
    The JBI specification clearly specifies that if all of the SUs of an SA fail to deploy, then the SA fails to deploy, but if at least one of the SUs does not fail, then the SA is properly deployed. The JBI specification is not clear for the other operations so we make the following choices:
    • We mirror the same behaviour for start and init.
    • We simply log errors for stop and shutdown.
    The idea is to be able to report enough informations to the administrators when possible while not preventing to undeploy an SA when there is problems. For example, a common use case such as starting an SA with only one SU that fails simply fail (while before it simply silently failed...)
  • Environment:
    -

Issue Links

Activity

Victor NOËL made changes - Tue, 22 Mar 2016 - 15:50:48 +0100
Field Original Value New Value
Status New [ 10000 ] Open [ 10002 ]
Priority Minor [ 4 ]
Assignee Christophe DENEUX [ cdeneux ] Victor NOËL [ vnoel ]
Victor NOËL made changes - Tue, 22 Mar 2016 - 15:50:51 +0100
Status Open [ 10002 ] In Progress [ 10003 ]
Victor NOËL made changes - Tue, 22 Mar 2016 - 15:51:03 +0100
Link This issue blocks PETALSDISTRIB-240 [ PETALSDISTRIB-240 ]
Victor NOËL made changes - Tue, 22 Mar 2016 - 15:52:56 +0100
Description The JBI specification clearly specifies that if all of the SUs of an SA fail to deploy, then the SA fails to deploy, but if at least one of the SUs does not fail, then the SA is properly deployed.

The JBI specification is not clear for the other operations so we make the following choices:
 * We mirror the same behaviour for start and init.
 * We simply log errors for stop and shutdown.

The idea is to be able to report enough informations to the administrators when possible while not preventing to undeploy an SA when there is problems.
The JBI specification clearly specifies that if all of the SUs of an SA fail to deploy, then the SA fails to deploy, but if at least one of the SUs does not fail, then the SA is properly deployed.

The JBI specification is not clear for the other operations so we make the following choices:
 * We mirror the same behaviour for start and init.
 * We simply log errors for stop and shutdown.

The idea is to be able to report enough informations to the administrators when possible while not preventing to undeploy an SA when there is problems.

For example, a common use case such as starting an SA with only one SU that fails simply fail (while before it simply silently failed...)
Victor NOËL made changes - Tue, 22 Mar 2016 - 17:08:52 +0100
Status In Progress [ 10003 ] Resolved [ 10004 ]
Fix Version/s 5.0.1 [ 10579 ]
Resolution Fixed [ 1 ]

People

Dates

  • Created:
    Tue, 22 Mar 2016 - 15:50:41 +0100
    Updated:
    Tue, 22 Mar 2016 - 17:08:52 +0100
    Resolved:
    Tue, 22 Mar 2016 - 17:08:52 +0100