| Feature-ID |
Desired Improvement |
Priority |
Comments |
Issue Link |
Engg Response |
| HA-1 |
HA for MDB and EJB endpoints |
P1 |
At present the MDB and EJB endpoints cannot be easily used easily with SailFin in a clustered architecture. Provide a way for handling this. |
|
|
| HA-2 |
Geographic redundancy within and across the cluster |
P3 |
|
|
|
| HA-3 |
Support network segmentation |
P2 |
|
|
|
| HA-4 |
Support synchronous replication |
P2 |
|
|
|
| HA-5 |
Rolling upgrade |
P2 |
|
|
|
| HA-6 |
Cluster aware Overload Protection |
P2 |
|
|
|
| HA-7 |
Two-tier deployment topology |
P1 |
Support is built in; however need's to be classified as production quaility by categorizing it as a feature for release driver |
|
|
|
| HA-8 |
SNMP support |
P2 |
Provide an end-to-end implementation supporting one of widely accepted H/w IP sprayer; wherein cluster reconfiguration events can be notified and acted upon by the fronting IP sprayer |
|
|
|
| HA-9 |
Support LVS fronting solutions and ability to communicate the cluster reconfiguration events like disabling, enabling of instances. This is specific to the self load balancing topology |
P1 |
|
|
| HA-10 |
Monitoring data on CLB |
P1 |
|
|
|
|
| HA-11 |
CLB layer can be dynamically added-to/deleted-from layer stack |
P1 |
CLB creation and deletion should not require a instance restart. This has dependency on layer handler. |
Issue 806 |
|
|
| HA-12 |
Clb http proxy does not support request pipelining |
P3 |
|
https://sailfin.java.net/issues/show_bug.cgi-421 |
|
| HA-13 |
ENHANCEMENT: Time-to-live on all SIP and HTTP sockets |
P3 |
|
https://sailfin.java.net/issues/show_bug.cgi-1104 |
|
