[j-nsp] Graphing VCP Backplane (possible jnxUtil SNMP bug?)

Phil Shafer phil at juniper.net
Thu Nov 3 12:39:44 EDT 2011


Charlie Allom writes:
>I've come across another performance issue - running through the jnxUtil
>MIB - it drives the load up significantly:

"show system processes extensive" output while the script is running
would be helpful.  The two key questions are is the script completing
before you launch the next invocation (use jcs:dampen() to limit this)
and is the CPU spent in "cscript" (the process running the op script)
or in snmpd/mib2d (doing the snmp operations)?

Some notes:

Looks like you're still using jcs:invoke() inside fetch-object().

Avoid "//" since it has to inspect the entire document to
find nodes that match.   Use an explicit path (or two) if
possible.

Use logging and jcs:trace() to check timings to see where time
is being spent.

SLAX uses "call-by-name"; if you are passing the variable
with the same local name, you can say:

    call foo($this);

instead of:

    call foo($this = $this);

for brevity and clarity.  This is purely cosmetic and won't
affect performance.

Thanks,
 Phil


More information about the juniper-nsp mailing list