Details
-
Sub-task
-
Status: Closed
-
Normal
-
Resolution: Fixed
-
None
-
None
-
None
Description
When deleting and renaming nodes to the deleted node, we cannot find out on repo cluster nodes whether some node was missing due to jackrabbit journal updates during loading the model, or whether the node is really missing. If we avoid the renaming, we can better identify a glitch in repository clustering
Currently, we have the following model for publishing a configuration:
if we start with :
+hst:configurations
+hippogogreen
+hippogogreen-preview
When we publish, we remove the hippogogreen and rename the hippogogreen-preview to hippogogreen.
I now want to go to this model:
We start with
+hst:configurations
+hippogogreen
When clicking edit, we get
+hst:configurations
+hippogogreen
+hippogogreen-v0
After publish we get
+hst:configurations
+hippogogreen-v0
After edit, we get
+hst:configurations
+hippogogreen-v0
+hippogogreen-v1
Now, I need to add one more thing:
On the hst:sites, I need to keep track of the version only. So for example, we start with
+hst:sites
+ hippogogreen (no version)
+ hippogogreen-preview (no version)
After edit, we get:
+hst:sites
+ hippogogreen (no version)
+ hippogogreen-preview (version=0)
After publish, we get
+hst:sites
+ hippogogreen (version=0)
+ hippogogreen-preview (version=0)
After edit, we get
+hst:sites
+ hippogogreen (version=0)
+ hippogogreen-preview (version=1)