Delayed ProvOutOfServiceEv and ProvInServiceEv

[derrickgunter]

The provider goes out of service then back in service but these events get queued for a long time. In the meantime some DNs go out of service and back in service causing CiscoAddrInServiceEv to be sent *before* ProvInServiceEv.  The result is that the application is never notified that some DNs are in service *after* the provider comes in service.  Is this a correct interpretation of the jtapi log?  If so, it looks like a jtapi problem.  See the attached jtapi log.

For example:

ProvOutOfServiceEv queued...

262975: Aug 22 02:02:42.968 ICT %JTAPI-JTAPI-7-UNK:(P1-jtapigenesys) ProvOutOfServiceEv [#35381] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12

262976: Aug 22 02:02:42.968 ICT %JTAPI-JTAPIIMPL-7-UNK:[com.genesyslab.ciscocm.JInterface@d9896e]ObserverProxy.queueEvents: queuing asynchronously

262977: Aug 22 02:02:42.968 ICT %JTAPI-MISC-7-UNK:ObserverThread(com.genesyslab.ciscocm.JInterface@d9896e): queuing com.cisco.jtapi.JtapiProviderEventSet

...

ProvInServiceEv queued...

264392: Aug 22 02:02:44.421 ICT %JTAPI-JTAPI-7-UNK:(P1-jtapigenesys) ProvInServiceEv [#35382] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12

264393: Aug 22 02:02:44.421 ICT %JTAPI-JTAPIIMPL-7-UNK:[com.genesyslab.ciscocm.JInterface@d9896e]ObserverProxy.queueEvents: queuing asynchronously

264394: Aug 22 02:02:44.421 ICT %JTAPI-MISC-7-UNK:ObserverThread(com.genesyslab.ciscocm.JInterface@d9896e): queuing com.cisco.jtapi.JtapiProviderEventSet

...

In the meantime CiscoAddrInServiceEv is queued and sent for 11371...

265115: Aug 22 02:02:44.531 ICT %JTAPI-JTAPI-7-UNK:(P1-jtapigenesys) [11371:P_Internal_HQ] CiscoAddrInServiceEv [#35426] Cause:100 CallCtlCause:0 CiscoCause:0 FeatReason:12

265116: Aug 22 02:02:44.531 ICT %JTAPI-JTAPIIMPL-7-UNK:[com.genesyslab.ciscocm.CcmAddrTermObserver@1a8d460]ObserverProxy.queueEvents: queuing asynchronously

...

268152: Aug 22 02:02:46.000 ICT %JTAPI-JTAPIIMPL-7-UNK:ObserverThread(com.genesyslab.ciscocm.CcmAddrTermObserver@1a8d460): delivering JAES[1]

...

ProvOutOfServiceEv and ProvInServiceEv sent over 3 seconds after being queued...

268887: Aug 22 02:02:47.562 ICT %JTAPI-JTAPIIMPL-7-UNK:ObserverThread(com.genesyslab.ciscocm.JInterface@d9896e): delivering JPES[1]

Could you confirm?  Thank you.

[derrickgunter]

Any ideas from the Cisco team? Thanks.

[mpotluri]

Can you check if providerChangedEvent has any blocking requests. If seems that the events are not processed though there are queued.

[ObserverThread(com.genesyslab.ciscocm.JInterface@d9896e)](P1-jtapigenesys) Request: registerFeature 1234

The above trace indicates that some blocking requests are called from within the call back which will cause application not to 'see' the subsequent events until the request returns.

[derrickgunter]

What do you mean by blocking requests?  How are these checked for?  Looks like jtapi is blocking, not the CTI application, right?  Thanks.

[derrickgunter]

Any updates? Thanks.