Known Issues for Release S-Cz8.1.0M1
The Known Issues below apply to release S-Cz8.1.0M1 of the Subscriber Aware Load Balancer (SLB). For Known Issues on the following products, when they are Subscriber-Aware Load Balancer (SLB) cluster members, refer to the Oracle Communications Session Border Controllers (OCSBC) product's Release Notes.
TLS
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
|
26779685 |
There is an
issue when the OCSLB has a
cluster-config
element with a
service-port
sub-element configured with:
When configured as above, and when supporting multiple services on the applicable tunnel, the OCSLB drops all TLS traffic on that tunnel. |
S-Cz7.3.10 | S-Cz8.1.0M1 | |
Balance/Rebalance Feature
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
|
20577310 |
When approaching the maximum supported registrations per second, the user may find that only a small percentage of endpoint registrations correctly expire and must re-register. | S-Cz7.2.10 | ||
High Availability (HA)
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
|
22066052 |
During a HA failover, IPT core miss errors are incremented, which results in retransmissions. | S-Cz7.2.10p3 | ||
|
26050436 25814014 |
When an SLB is operating in HA mode, the standby periodically sends SYNC messages to the Active. If, for any reason, the Active's response is delayed, the standby resends these SYNCs. If this "resend" happens after the active has responded, the active writes a "Stray response" log message to the log.lbp file. This issue may confuse the user, but does not impact service. | S-Cz7.3.10 | ||
|
26006697 |
When the SLB is operating in HA mode and the user removes an SBC from a cluster that has a large number (~100k) of endpoints, both the active and secondary SLB may crash. | S-Cz7.3.10 | S-Cz7.3.10p1 | |
|
25969936 |
When the SLB is operating in HA mode, it writes an inordinate number of SYNC messages to the log.lbp file. This makes it difficult for the user to parse through this file when troubleshooting a related issue, and causes the file to wrap sooner than expected. | S-Cz7.3.10 | S-Cz7.3.10p1 | |
| When populated
with 10 million registrations or more, the SLB may not synchronize all
registrations before the default
redundancy-config's
becoming-standby-time
timeout of 180000.
Workaround: Set the SLB's becoming-standby-time timeout to 1800000 when handling 10 million registrations or more. |
S-Cz7.3.10 | |||
IMS-AKA
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
| This release does not support IMS-AKA endpoints. | OCSBC Release S-CZ739p1 | OCSBC Release S-CZ800 | ||
Handling Fragmentation
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
|
25933886 |
When operating on the Acme Packet 6100, the OCSLB may crash when receiving 200 messages per second that are fragmented at 12K bytes or more. | S-Cz7.3.10 | ||
Duplicate IPs on Applicable SBC Interfaces
| ID | Description | Found In | Fixed In | |
|---|---|---|---|---|
| The Oracle Communications Session Border Controller does not support overlapping IP addresses on Media interfaces between end points and SBCs in deployments that use an Oracle Communications Subscriber Aware Load Balancer for load balancing. | S-Cz7.3.10 | |||
