Job status is shown failed even Router is rebooted

[er_yogendrasingh]

am using following cli

cwcli config reload -u $CWUser -p $CWPass -ipaddress $deviceIP

Though my router is properly reloaded but still m getting Job status as failed.. could NEbody let me know the possible reasons..

[Joe Clarke]

This is typical. RME cannot verify the job actually completed because the device was reloaded. In order to get around this, specify a larger timeout on the command line using the -timeout option.

[er_yogendrasingh]

Hi Joe,

I tried with the timeout option but now am getting the following error.. looks like this error in not related to timeout this seems to be some other issue occurring, this time Router is not restarted

C:\Documents and Settings\cworks>cwcli config reload -u admin -p admin -ipaddres

s 10.10.10.102 -timeout 10

* Warning * The -p option is highly insecure and *not* recommended.

* Warning * See -u option for more details.

INFO - Devices to be attempted in the job:

10.10.10.102

INFO - The job 1170 is created

Waiting for the job results ...

..............!

- Job Status: Job Failed

Failed Devices:

10.10.10.102

ERROR - CM0102 Reload Device failed Cause: java.net.SocketTimeoutExcepti

on: Read timed out

SUMMARY

========

Failed: reload

[Joe Clarke · Accepted solution]

You need to use a much larger timeout. The default is 30, so you'll need to go at least as large as that. How long does it take for the router to completely reload, and become responsive again? Set the timeout to that plus 10 seconds.

[er_yogendrasingh]

my router generally takes not more than 90 sec to completely reboot..

but looks like now with this java exception even reload event is not going to router.. earlier at least the router was restarting but but the error is return from Cisco works itself and not even rebooting the router

ERROR - CM0102 Reload Device failed Cause: java.net.SocketTimeoutExcepti

on: Read timed out

SUMMARY

========

Failed: reload

C:\Documents and Settings\cworks>cwcli config reload -u admin -p admin -ipaddres

s 10.10.10.102 -timeout 100

* Warning * The -p option is highly insecure and *not* recommended.

* Warning * See -u option for more details.

INFO - Devices to be attempted in the job:

10.10.10.102

INFO - The job 1171 is created

Waiting for the job results ...

..!

- Job Status: Job Failed

Failed Devices:

10.10.10.102

ERROR - CM0102 Reload Device failed Cause: java.net.SocketTimeoutExcepti

on: Read timed out

SUMMARY

========

Failed: reload

[Joe Clarke]

You will need to enable debugging, and post the log. Without it, I cannot say where the timeout is occurring.

[er_yogendrasingh]

attached is the log file... :-)

following cli was executed

C:\Documents and Settings\cworks>cwcli config reload -u admin -p admin -ipaddres

s 10.10.10.102 -timeout 100 -d 5 -l relaodRouter.log

[Joe Clarke]

Please post a binary sniffer trace of all traffic to this device while the reload job is running.

[er_yogendrasingh]

Hi joe,

wouldnt it be possible to debug the issue without binary sniffer trace, i would not be able to provide the same

[Joe Clarke]

If you cannot post it to the forum, please open a TAC services request with the trace and the previously posted log.

[er_yogendrasingh]

Hi Joe,

please find the attached pcap output captured on my Ciscoworks server using ethereal while sending reload Command.

[Joe Clarke]

The config needs to be saved before this device can be reloaded. Push a write mem job to the device, then a reload.

[er_yogendrasingh]

Hi Joe,

tried saving the config first also.. but no success :-((

attached are the logs for cli sent by me on CW server also including Console logs of my router with "show clock" command executed before and after the reload which shows that it take no more than 3 mins to reload the router.. Please note that Router was reloaded by Ciscoworks job but still return Job status as FAILED

[Joe Clarke]

I looked into this, and it appears they broke config reload in RME 4.1 when using telnet or SSH as the transport. As a workaround, make TFTP your first config deployment protocol, and make sure "snmp-server system-shutdown" is configured on the device, and that DCR has a valid read-write SNMP credential, and the device will be reloaded via SNMP. The status for an SNMP reload job will be successful if the device was properly reloaded.

[er_yogendrasingh]

Thnx Joe,

Which mean we would not likely to see the same issue on some other version of RME :-)

I highly appreciate your time spent on this:-)