avayner at cisco
Jul 1, 2012, 10:07 PM
Post #2 of 2
(241 views)
Permalink
|
Hank,
I think it is related to CSCtc38611 "need performance enhancements to 'show ip cache flow' on ASR1k"
Symptom:
"show ip cache flow" and "show ip cache flow | include WORD" may take a very long time to run
Conditions:
Several thousand flows being learned
Workaround:
Use "show ip cache x.x.x.x flow" to see specific flows
Basically, ASR1000 is a distributed platform, and the cache info is stored on ESP (forwarding path). It means that every time you issue this command it would have to transfer the whole table to the RP...
If you use the "| include" option, it still has to transfer the whole file, and then apply the regexp on the text...
If you want to do multiple searches on a large table, the best solution would most likely be to copy the output to a file, and then use that file...
If you want to monitor a specific flow, using the specific command above works faster, as it does not transfer the whole file...
Arie
-----Original Message-----
From: cisco-nsp-bounces [at] puck [mailto:cisco-nsp-bounces [at] puck] On Behalf Of Hank Nussbacher
Sent: Sunday, July 01, 2012 07:47
To: cisco-nsp [at] puck
Subject: [c-nsp] "show ip cache flow" is slow
Ever since we switched to ASR1004 running XE15.1(2)S1, we have seen that the output of "show ip cache flow" stalls and is super slow to complete. We have a few interfaces with "ip flow ingress" defined. What can be causing this slowness? Any recommendations of commands to speed up the output?
Thanks!
-Hank
_______________________________________________
cisco-nsp mailing list cisco-nsp [at] puck https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/
_______________________________________________
cisco-nsp mailing list cisco-nsp [at] puck
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/
|
