cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
982
Views
0
Helpful
10
Replies
Highlighted
Beginner

NCS 5501

Hi,

 

I'm getting this error whenever i issued the ping command:

 

RP/0/RP0/CPU0:NCS03#ping 10.77.77.77
Fri May 11 05:33:19.276 UTC
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.77.77.77, timeout is 2 seconds:ping: vsocket_pak -- Unknown error -1

Can someone help me fix this issue?  Thank you in advance.

HN

 

 

 

10 REPLIES 10
Highlighted
Beginner

this error doesn't ring a bell, but if you can share the "show route 10.77.77.77" and "show cef 10.77.77.77", this could give us some clues.

/Aleksandar
Highlighted

When we open a ping_sock (socket for ping) we are getting an error.

 

Are you able to ping any IP address, loopback, physical, remote peer, etc?

 

Please also collect the following details and open a TAC case if possible.

show raw statistics clients loc <primary RP>
ping <address that doesn't work>
show raw statistics clients loc <primary RP>
show raw brief loc <primary RP>
ping <address that doesn't work> count 10000 timeout 0
<on another terminal collect show raw brief loc <primary RP>> The socket for ICMP (protocol 1) is only open when a ping is initiated
show raw trace reverse
show ping ipv4 trace reverse

 

Thanks,

Sam

Highlighted

Hi Sam,

Thank you for your reply.

Are you able to ping any IP address, loopback, physical, remote peer, etc?

None of the above is the answer to this question.  I can ping the interface from the far end without any issue.  I have never seen this error before and please see attached for the commands that you wanted me to run.

Highlighted

I'm in discussions with the socket team and they want us to turn on the following debugs and then run the ping again.

debug socket api

debug socket events

debug socket error

debug socket io

 

I would recommend letting these debugs log to the buffer instead of to console or monitor and then copying out the show logging.

 

Sam

Highlighted

Thank you Sam.  The attached is "sh logging" after I ran those commands that you recommended.  I did a quick link from my PC to the Mgmt and pinged.  Do you think restore this switch to a factory default will fix this issue?

 

HN

Highlighted

Most likely a reload or factory reset will fix the issue, but I think its best to try and find the RCA for the issue. If its a bug then we would like to fix it.

 

Some of the other syslogs look to be useful, looks like the kim process is unable to bind to the global-vrf namespace, that might be part of the issue. I am pending feedback from the development team on next steps.

 

Sam

Highlighted

Can you send me an email with some times/timezone you are available for a WebEx? The development team has asked for more detailed troubleshooting which will be difficult to do on here. My name is my username at cisco.com

Also if you have the ability to open an SR that will be helpful.

 

Thanks,

Sam

Highlighted

Thank you Sam.  I just sent you an email.  

Highlighted

Hi Aleksandar, see below for the commands that you wanted to see:

 

RP/0/RP0/CPU0:NCS03#sh route 10.77.77.77
Fri May 11 23:44:32.442 UTC

Routing entry for 10.77.77.0/24
Known via "connected", distance 0, metric 0 (connected)
Installed May 11 23:42:56.820 for 00:01:35
Routing Descriptor Blocks
directly connected, via HundredGigE0/0/1/2
Route metric is 0
No advertising protos.
RP/0/RP0/CPU0:NCS03#
RP/0/RP0/CPU0:NCS03#sh cef 10.77.77.77
Fri May 11 23:44:54.698 UTC
10.77.77.0/24, version 58, attached, connected, glean adjacency, internal 0x3000061 0x0 (ptr 0x87e528e0) [1], 0x0 (0x881a4370), 0x0 (0x0)
Updated May 11 23:42:56.827
Prefix Len 24, traffic index 0, precedence n/a, priority 0
via HundredGigE0/0/1/2, 2 dependencies, weight 0, class 0 [flags 0x8]
path-idx 0 NHID 0x0 [0x87619908 0x0]
glean adjacency

Highlighted

if this is directly connected interface which from show route it seems like, the local adjacency should be shown in show cef, its showing glean.

can you also check ARP if it is resolved.