04-18-2012 06:17 AM
Hi Experts!
I can't get Callhome to work on our Nexus switches. We use a Transport Gateway with e-mail between devices and the gateway.
The problem is that the Nexus won't send any email, though it always default to the CiscoTAC-1 profile for which I have not specified an e-mail address. On my custom destination-profile there of course are a e-mail address specified but it's never used.
Here my config:
callhome
email-contact xxx@xx.xx
phone-contact +xxxxx
streetaddress xxxxx
destination-profile XXX-XXX
destination-profile XXX-XXXformat XML
destination-profile XXX-XXX message-level 4
destination-profile XXX-XXX email-addr xxx@xx.xx
destination-profile XXX-XXX alert-group Diagnostic
destination-profile XXX-XXX alert-group configuration
destination-profile XXX-XXX alert-group environmental
transport email smtp-server xx.xx.xx.xx port 25 use-vrf default
transport email from xxx@xx.xx
enable
When i try to register by using callhome send diagnostic this is what shows up in the debug:
2012 Apr 18 14:35:42.949270 callhome: Destination profile CiscoTAC-1
2012 Apr 18 14:35:42.949329 callhome: IN send_callhome_email_mesg with msg_id :1159126505
2012 Apr 18 14:35:42.949442 callhome: executing command:mv /dev/shm/callhome.d/workspace//workspaceJFtz4L /callhome/spool/7FTPEQ
2012 Apr 18 14:35:42.976959 callhome: no email address configured for destination profile:CiscoTAC-1
2012 Apr 18 14:35:42.976987 callhome: problem in creating the email script for destination profile:CiscoTAC-1
2012 Apr 18 14:35:42.977001 callhome: In cleanup_spool:
2012 Apr 18 14:35:42.977015 callhome: executing command:rm -rf /callhome/spool/7FTPEQ
2012 Apr 18 14:35:43.031563 callhome: retcode[0] 23 : retcode[1] 1077608463
2012 Apr 18 14:35:43.031603 callhome: problem in transporting the message Could send message; transport methods failedCould send message; transport methods failedCould send message; transport methods failed
2012 Apr 18 14:35:43.031618 callhome: No messages sent out
2012 Apr 18 14:35:43.031632 callhome: Exiting send_callhome_mesg
2012 Apr 18 14:35:43.031648 callhome: could not format or send message for callhome alert:(null)
2012 Apr 18 14:35:43.031685 callhome: Error is : warning:no callhome message sent
2012 Apr 18 14:35:43.031705 callhome: Response for test callhome
2012 Apr 18 14:35:43.031823 callhome: setting error code for last callhome message to 1077608465
2012 Apr 18 14:35:43.031922 callhome: No of elements in the process list is 0
Can someone please help me getting this Callhome configuration work?
Regards,
Johan
Solved! Go to Solution.
04-18-2012 11:47 AM
First, the version helps identify if you are hitting a bug. To get the entire call home configuration, use the command show run callhome all.
Now on to your specific issue. Check the output to
show callhome destination-profile
There is an issue in Nexus 7000 when you upgrade, the email configuration shows in the running configuration but not in the destination profile. But it is an old bug. You can check if your email server shows up there in the command output.
But the answer might be easier than that. Some user generated Call Home messages are sent at severity level 2. Since you have
destination-profile XXX-XXX message-level 4
in the configuration, these severity 2 messages will *NOT* be sent - only messages above severity 4 will ever be sent.
Please try again after configuring the following command
destination-profile XXX-XXX message-level 0
In case you are wondering, Call Home severities are opposite of syslog severities. Here is the mapping
Call Home Level | Keyword | Syslog Level | Description |
9 | catastrophic | N/A | Network-wide catastrophic failure |
8 | disaster | N/A | Significant network impact |
7 | fatal | Emergency (0) | System is unusable |
6 | critical | Alert (1) | Critical conditions, immediate attention needed |
5 | major | Critical (2) | Major conditions
|
4 | minor | Error (3) | Minor conditions |
3 | warning | Warning (4) | Warning conditions |
2 | notification | Notice (5) | Basic notification and informational messages. Possibly independently insignificant. |
1 | normal | Information (6) | Normal event signifying return to normal state. |
0 | debugging | Debug (7) | Debugging messages. |
The last suggestion if that does not, is look at the quickstart configuration guide for what you are doing and then fallback to opening a TAC case to do some 1-on-1 debugging.
http://www.cisco.com/en/US/docs/switches/lan/smart_call_home/QuickStart_NX7000.pdf
Lawrence
04-18-2012 07:40 AM
Hell Johan,
My name is Sameh and I'm from the Smart Services Bureau.
Thank you for posting your question here. We are currently investigating and will be posting answer hopefullly soon.
Regards,
Sameh Taha
Cisco Smart Services Bureau
04-18-2012 11:47 AM
First, the version helps identify if you are hitting a bug. To get the entire call home configuration, use the command show run callhome all.
Now on to your specific issue. Check the output to
show callhome destination-profile
There is an issue in Nexus 7000 when you upgrade, the email configuration shows in the running configuration but not in the destination profile. But it is an old bug. You can check if your email server shows up there in the command output.
But the answer might be easier than that. Some user generated Call Home messages are sent at severity level 2. Since you have
destination-profile XXX-XXX message-level 4
in the configuration, these severity 2 messages will *NOT* be sent - only messages above severity 4 will ever be sent.
Please try again after configuring the following command
destination-profile XXX-XXX message-level 0
In case you are wondering, Call Home severities are opposite of syslog severities. Here is the mapping
Call Home Level | Keyword | Syslog Level | Description |
9 | catastrophic | N/A | Network-wide catastrophic failure |
8 | disaster | N/A | Significant network impact |
7 | fatal | Emergency (0) | System is unusable |
6 | critical | Alert (1) | Critical conditions, immediate attention needed |
5 | major | Critical (2) | Major conditions
|
4 | minor | Error (3) | Minor conditions |
3 | warning | Warning (4) | Warning conditions |
2 | notification | Notice (5) | Basic notification and informational messages. Possibly independently insignificant. |
1 | normal | Information (6) | Normal event signifying return to normal state. |
0 | debugging | Debug (7) | Debugging messages. |
The last suggestion if that does not, is look at the quickstart configuration guide for what you are doing and then fallback to opening a TAC case to do some 1-on-1 debugging.
http://www.cisco.com/en/US/docs/switches/lan/smart_call_home/QuickStart_NX7000.pdf
Lawrence
04-18-2012 12:40 PM
I noticed I did a typo when I wrote "only messages above severity 4 will ever be sent". I meant to say that only messages at severity 4 or above will ever be sent.
BTW, callhome send diagnostic is a severity 2 message.
Lawrence
04-19-2012 06:44 AM
Hi Lawrence!
We are using 5.1(3)N1(1A)
Yes, the destination-profile XXX-XXX message-level 4 seems to be the problem, great.
Another thing a noticed that looks suspicious is that some configururation lines seems to get duplicated.
transport email smtp-server 10.10.10.24 port 25 use-vrf default
transport email from callhome@xx.xx
transport email smtp-server 10.10.10.24 port 25 use-vrf default
transport email from callhome@xx.xx
If I remove one of the lines, the other one also disapears. Everything works, so I guess it´s cosmetic.
Thank you
Johan
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide