RE: [jffnms-users] BGP peer prefix trending

2004-12-09 Thread JDeane
An interesting read... http://www.juniper.net/techpubs/software/junos/junos63/swconfig63-net-mg mt/html/mib-jnx-bgpmib2.txt Jade -Original Message- From: Javier Szyszlican [mailto:[EMAIL PROTECTED] Sent: Thursday, December 09, 2004 10:39 AM Cc: [EMAIL PROTECTED] Subject: Re: [jffnms-us

Re: [jffnms-users] Triggers Problem

2004-12-09 Thread Javier Szyszlican
Hi Tim, Can you show us the contents of the triggers and triggers_rules tables? Is this JFFNMS 0.7.9 ? Is this a Match Events or Match Alarms trigger? Javier Tim Pepitone wrote: Hello all, I recently began using jffnms in my office with a great deal of success except for email alerts. I've

Re: [jffnms-users] BGP peer prefix trending

2004-12-09 Thread Javier Szyszlican
Jade, It seems this addition to the BGP4 mib allows you to query the routes learned by peer, but not the total number, you will still need to walk a sub-tree. Javier [EMAIL PROTECTED] wrote: Have a look at this: http://www.cisco.com/en/US/products/sw/iosswrel/ps1839/products_feature_ guide09186a0

RE: [jffnms-users] BGP peer prefix trending

2004-12-09 Thread JDeane
Have a look at this: http://www.cisco.com/en/US/products/sw/iosswrel/ps1839/products_feature_ guide09186a0080110bbc.html If I'm reading this correctly, and I'm no SNMP expert, it's at least per-peer based. And you're absolutely right regarding processing waste from walking an entire routing tabl

[jffnms-users] Triggers Problem

2004-12-09 Thread Tim Pepitone
Title: Triggers Problem Hello all,     I recently began using jffnms in my office with a great deal of success except for email alerts.  I've check that the PHP mail() function works properly, enabled logging, created an alert, assigned the alert to a user, set user's email profile va

Re: [jffnms-users] BGP peer prefix trending

2004-12-09 Thread Javier Szyszlican
Jade, As I see it this is one thing we can't do. The BGP4 MIB does not have the number of prefixes received as an OID, the only way to count them is to walk the complete routing table, which is slow and causes high CPU load. Javier [EMAIL PROTECTED] wrote: Would it be possible, as an element of