show ccd stats
The show ccd stats command displays endpoint statistics for the SBC members of the cluster.
ORACLE# show ccd stats
17:10:09-54
-------- Period ------- ---- LifeTime ----
SD Active Rate High Total Total PerMax High
bass I285714 0.0 285714 0 285.71K 13.76K 285.71K
guinness I285714 0.0 285714 0 285.71K 13.76K 285.71K
magichat I285714 0.0 285714 0 285.71K 13.76K 285.71K
newcastel I285714 0.0 285714 0 285.71K 13.76K 285.71K
samadams I285714 0.0 285714 0 285.71K 13.76K 285.71K
sixtus I285714 0.0 285714 0 285.71K 13.76K 285.71K
westy I285714 0.0 285714 0 285.71K 13.76K 285.71K
Total endpoints: 153908
Total rate : 0.0
Total SDs : 9
ORACLE#The Period stats provided represent an accumulation of data for the amount of time specified after the dash separator in the timestamp printed in the first line of output (in this example, the period represents 54 seconds).
The single ASCII character between the SD column and the Active column is the state of that SBC; the letter I represents InService.
The Rate column displays the transmission rate of new endpoint associations to that particular SBC. (In the sample, no new endpoints are arriving in the cluster, so all of the SBCs show a rate of 0.0.) The High field indicates the highest number of active endpoint associations for the current period.
When issued with an optional hostname argument, the show ccd stats command provides a detailed report for the target hostname.
ORACLE# show ccd stats bass
15:09:25-59
SD bass [InService]
State -- Period -- -------- Lifetime --------
Active High Total Total Permax High
Tunnels 1 1 0 2 1 1
Service Ports 2 2 0 2 1 2
endpoints 53571 53571 0 53571 14399 53571
Contacts 53571 53571 0 53571 14399 53571
Sessions 0 0 0 0 0 0
---- Lifetime ----
Recent Total PerMax
Heartbeats rcvd 30 27426 15
Heartbeats Missed 0 1 1
Tunnel Adds 0 2 1
Tunnel Removes 0 1 1
Service Adds 0 2 1
Service Removes 0 0 0
endpoint Removes 0 0 0
endpoint Promotes 0 53571 13561
endpoints Skipped 0 0 0
Rebalance Source 0 0 0
Rebalance Targe 0 0 0
Rebalance Request 0 0 0
Rebalance Replies 0 0 0
CPU Above Limit 0 0 0
CPU Above Threshold 0 0 0
Online Transitions 0 0 0
Offline transitions 0 0 0
Tunnel Add Fails 0 0 0
CCD Tunnel Add Fails 0 0 0
Tunnel Remove Fails 0 0 0
CCD Tunnel Remove Fails 0 0 0
Service Add Fails 0 0 0
CCD Svc Add Fails 0 0 0
Service Remove Fails 0 0 0
CCD Svc Remove Fails 0 0 0
Service Adds No Cfg 0 0 0
Bad Service Handle 0 0 0
endpoint Remove Fails 0 0 0
endpoint Prom Fail 0 0 0
ORACLE#
The Period stats provided represent an accumulation of data for the amount of time specified after the dash separator in the timestamp printed in the first line of output (in this example, the period represents 59 seconds).
Tunnels contains the number of tunnels between the Subscriber-Aware Load Balancer and the target SBC, in this case, bass.
Service Ports contains the number of Service Ports advertised by the target SBC when it joined the cluster.
endpoints and Contacts contain the number of endpoint associations the Subscriber-Aware Load Balancer has assigned to the target SBC. If there is only one registering device at a given endpoint, a one-to-one correlation between endpoints and contacts is expected. However, if the atom-limit-divisor parameter has been set to a non-default value, the number of contacts exceeds the number of endpoints.
Sessions contains the number of active calls.
The Lifetime stats provided represent an accumulation of data since the last reboot.
HeartBeats rcvd contains the number of heartbeat/keepalive messages received from the target SBC. Heartbeats are sent every two seconds by the SBC.
HeartBeats Missed contains the number of scheduled heartbeat/keepalive messages not received from the target SBC.
The Tunnel Adds and Tunnel Removes counters are incremented when an SBC joins the cluster and leaves the cluster, respectively.
The Service Adds and Service Removes counters are incremented when an SBC advertises support for a service and withdraws support for a service, respectively. This generally happens only when an SBC first joins the cluster, or if the configuration on a clustered SBC is changed, saved, and activated.
The endpoint Removes counter tracks the number of SBC-originated Cluster Control messages that request the Subscriber-Aware Load Balancer to delete a forwarding rule. Such a request can be the result of (1) a rebalance operation (when the Subscriber-Aware Load Balancer asks for the SBC to nominate candidates for rebalancing), (2) an endpoint de-registration with the SBC, or (3) an endpoint is power down. Generally, whenever a registration cache entry on a clustered endpoint is removed by the SBC, it notifies the Subscriber-Aware Load Balancer to remove that binding.
The endpoint Promotes counter tracks the number of promotion messages the SBC sends to the Subscriber-Aware Load Balancer to validate an untrusted forwarding rule. When the SLB first creates a forwarding rule for a new endpoint, it treats it as untrusted. When the SBC receives a 200 OK for a REGISTER message from that endpoint’s registrar, the SBC sends a Promote Cluster Control message to the Subscriber-Aware Load Balancer. At this point, the Subscriber-Aware Load Balancer modifies the particular forwarding rule and assigns it trusted status. If this Promote message is not received within the time configured as the untrusted-grace-period in the lbp-config, the Subscriber-Aware Load Balancer deletes the untrusted entry.
endpoints Skipped contains the number of endpoints in its registration cache that the SBC has skipped over during a rebalance request. Skipping may be done for one of two reasons: either the most appropriate user for rebalancing was in an active phone call (and rebalance-skip-calls was enabled in cluster-config), or the rebalance-skip-ahead value in cluster-config was set to a nonzero value. In this case, when the SBC is asked to nominate users for rebalance, it will skip over any users whose registration cache entry is due to expire within the number of milliseconds set as the rebalance-skip-ahead value.
Rebalance Source contains the number of times the target SBC was used as a source of endpoints during a rebalance operation (that is, it supplied endpoints to a cluster member that was added to the cluster after itself).
Rebalance Target contains the opposite: the number of times that SBC was the recipient of endpoints from other sources during a rebalance operation.
The Rebalance Requests and Rebalance Replies counters increment upon receipt of a Cluster Control message from the Subscriber-Aware Load Balancer to the SBC asking it to divest itself of endpoints, and the responsive Cluster Control message from the SBC that indicates the endpoints the SBC has chosen.
The CPU Above Limit and CPU Above Threshold counters increment whenever an SBC has reported a high CPU value, and has been taken out of consideration for new endpoint assignments. Generally, the CPU limit and threshold are the same value (90%). However, it is possible to configure the threshold to be lower using the sip-config option load-limit.