Port of 3560 PoE switch gets reset automatically

[K_require]

We have PoE 3560 switch with IOS version c3560-ipbase-mz.122-25.SEE3. IP Phones connected to this switch get rebooted automatically. We found the following logs on switch:

Jul 25 22:23:58: %ILPOWER-7-DETECT: Interface Fa0/38: Power Device detected: IEEE PD

Jul 25 22:23:58: %ILPOWER-5-POWER_GRANTED: Interface Fa0/38: Power granted

Jul 25 22:23:59: %LINK-3-UPDOWN: Interface FastEthernet0/38, changed state to down

Jul 25 22:23:59: %SWITCH_QOS_TB-5-TRUST_DEVICE_LOST: cisco-phone no longer detected on port Fa0/38, port set to untrusted.

Jul 25 22:24:02: %LINK-3-UPDOWN: Interface FastEthernet0/38, changed state to up

Jul 25 22:24:03: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/38, changed state to up

Jul 25 22:24:21: %SWITCH_QOS_TB-5-TRUST_DEVICE_DETECTED: cisco-phone detected on port Fa0/38, port trust enabled.

On Cisco site I found following explaination:

SWITCH_QOS_TB Messages

The following are trusted boundary messages.

Error Message

%SWITCH_QOS_TB-5-TRUST_DEVICE_DETECTED : [chars] detected on port [chars], port trust enabled.

Explanation The trusted boundary facility has detected a device that matches the trusted device setting for the port, and the port trust state has been modified.

Recommended Action No action is required.

Error Message

%SWITCH_QOS_TB-5-TRUST_DEVICE_LOST : [chars] no longer detected on port [chars], port set to untrusted.

Explanation The trusted boundary facility has lost contact with a trusted device, and has set the port trust state to "untrusted".

Recommended Action No action is required.

How can we resolve this issue? Contact center agents are getting logged on to these IP Phones and lots of complaints of call drop are reported daily

[ivillegas]

Shortly after reload of the switch, do 'shut'/'no shut' on every port with phone, to force all the phones to exhibit a reboot due to 'power loss', and so avoiding any later unexpected and unscheduled reboots of Phones.

[ahochau_2]

Did you ever find the answer to this?

We are having the same issue and I am wondering if you found an answer.  Same scenario, contact center agent phones (7965) connected to a 3560 radomly have their ports get reset when on a call.  The phone goes hard down, and reboots causing a loss of network connectivity for the attached PC.  We are running version 12.2(35)SE5 on the switches and 9.1.1SR1 on the phones.

[Leo Laohoo]

Have you tried testing the cables using TDR of the switch?  (I'm not certain if TDR is supported on this model.)

1.  test cable tdr int ;
2.  Depending on your IOS, wait for about 7-10 seconds;
3.  sh cable tdr int

Please post the output here.

If everything looks "clean" then I'd recommend upgrading the IOS.  12.2(35) is pretty old.