Performance impact of SIP Monitor & Trace vs. Packet-trace
We've been using both trace tools for testing in our lab (the Web GUI SIP Monitor & Trace and the packet-trace CLI command that sends to a Wireshark PC) but now our Ops team wants to use one or both of these with their production 3820 and 4500 e-SBCs.
I'm looking for guidance on the performance impact of using these tools on the SBC's processor resources. I have seen cautions against using SIP logging on production boxes but have not found any comments on using these two trace tools on prod systems.
Should we even consider using them on prod boxes or should we allow them but with limitations (such as ensuring that the filters are very specific)?