CME Phones resetting

[Nelson Serrao]

I am using CME Version 7.0(1) and have a few 7960 phones configured. The phones keep resetting and it comes up as registration rejected. After a few seconds the phones re register without doing anything. I am using the loopback IP address as option 150 and for the CME IP address. The phones are using load 8.0(10.0) and all the files for it are in the flash. The phone type is configured under ephone. The max ephones and dn's is configure to the max. Below is a screen shot of what I am seeing:

DeviceName CurrentPhoneload PreviousPhoneload LastReset

=====================================================================

SEP00070E8B4725 8.0(10.0) 8.0(10.0) TCP-timeout

*Jan 21 21:04:16.457: %IPPHONE-6-UNREGISTER_ABNORMAL: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:1 DeviceType:Phone has unregistered abnormally.

*Jan 21 21:04:16.457: %IPPHONE-6-REGISTER: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:2 DeviceType:Phone has registered.

*Jan 21 22:06:54.357: %IPPHONE-6-REG_ALARM: 10: Name=SEP00070E8B4725 Load=8.0(10.0) Last=TCP-timeout

*Jan 21 22:06:54.365: %IPPHONE-6-UNREGISTER_ABNORMAL: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:2 DeviceType:Phone has unregistered abnormally.

*Jan 21 22:06:54.365: %IPPHONE-6-REGISTER: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:1 DeviceType:Phone has registered.

*Jan 21 22:08:01.853: %IPPHONE-6-REG_ALARM: 10: Name=SEP00070E8B4725 Load=8.0(10.0) Last=TCP-timeout

*Jan 21 22:08:01.857: %IPPHONE-6-UNREGISTER_ABNORMAL: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:1 DeviceType:Phone has unregistered abnormally.

*Jan 21 22:08:01.861: %IPPHONE-6-REGISTER: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:2 DeviceType:Phone has registered.

*Jan 21 22:13:37.441: %IPPHONE-6-REG_ALARM: 10: Name=SEP00070E8B4725 Load=8.0(10.0) Last=TCP-timeout

*Jan 21 22:13:37.445: %IPPHONE-6-UNREGISTER_ABNORMAL: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:2 DeviceType:Phone has unregistered abnormally.

*Jan 21 22:13:37.445: %IPPHONE-6-REGISTER: ephone-1:SEP00070E8B4725 IP:192.168.2.165 Socket:1 DeviceType:Phone has registered.

[Nicholas Matthews]

Can you send 'show proc cpu hist', ideally within an hour of the phones resetting?

hth,

nick

[Nelson Serrao]

333333333444443333333333333333333344444333333333333333333333

100

90

80

70

60

50

40

30

20

10

0....5....1....1....2....2....3....3....4....4....5....5....6

0 5 0 5 0 5 0 5 0 5 0

CPU% per second (last 60 seconds)

444445544444444444444444444444444444444444444444444455444444

100

90

80

70

60

50

40

30

20

10 ** **

0....5....1....1....2....2....3....3....4....4....5....5....6

0 5 0 5 0 5 0 5 0 5 0

CPU% per minute (last 60 minutes)

* = maximum CPU% # = average CPU%

1 2 5 24 1 11 1 2

55754344333434457638533934937333443344333664863

100

90

80

70

60

50 *

40 * *

30 * * **

20 * * * ** * ** * *

10 **** *** *** ** * * ** ***

0....5....1....1....2....2....3....3....4....4....5....5....6....6....7..

0 5 0 5 0 5 0 5 0 5 0 5 0

CPU% per hour (last 72 hours)

* = maximum CPU% # = average CPU%

[Nicholas Matthews]

Well,

I can tell you why your phones are unregistering - they are having TCP timeouts. The phones will send keepalives every 30 seconds, and if they miss 3 or 4 (don't remember offhand) they unregister.

The tricky part is finding out why there is a TCP timeout. Often it is a busy CPU, which isn't the case here.

Other TCP timeout possible causes:

Firewalls

ACLs

Packet Loss

Phone/CME bugs

Routing issues

What's the L3/L2 topology between the phones and CME? What's the platform?

You can use this command on CME to allow the phones to SPAN their PC ports so you can sniff:

telephony-service

service phone spanToPCPort 0

You could run a sniffer and see what's going on with the wire. You could also try switching IOS on the CME. Check interfaces on CME with 'show int' and on the switches for errors, output drops, etc. If you have QOS on the LAN, turn it off.

Very hard to tell what's going on with speculation.

hth,

nick

[Nelson Serrao]

I did as instructed and sniffed the packets. I was able to see the TFTP requesting for some files in packets 15606 to 15614. Am attaching the capture file for reference. However, I am struggling to find the cause. Please can someone assist.

[toormehdi]

I had same issue. please upgrade to 7.1 with 12.4.22XB