Cacti

Would this script query all the routers I'm monitoring to see if any new interfaces have been added to device? I've been trying to find out how cacti handles that. There is a cli script that can recheck a host for changes/new interfaces on polled devices: poller_reindex_hosts.php That is in the <ca...

95th Percentile is a factor of the traffic passing through the interface, not the capability of the bandwidth allowed through, so the only way that changes is if the traffic actually changes. If you have a 10Mbs burstable limit, and your actual traffic through the interface was 10Mbs, then the 95th ...

Not uncommon to have both inbound and outbound the same on an interface if that particular interface is both the inbound and outbound path for the traffic in question. An example is the classic 'router on a stick' scenario, where all traffic goes to the router, only to be routed back out the same in...

See Phalek's nmidCreatePDF plugin...this should have what you are looking for, although it's still in development. Yes, I looked at it, but what I have already done myself is much cleaner at this time, although I think the nmid version could be very useful, if some of the requirements are changed (...

I thought I saw some work being done on this somewhere, but I can't find it. The Reportit plugin is fine for scheduled reports, although I'd like to be able to both export to CSV and PDF, and for more than just the current report. What Is needed is an adhoc report tool, so that you can select what g...

Have you tried checking to see how many crons are running when this is happening? When I have seen similar in the past, I would have 3-4 running. I would stop the crond daemon, stop the poller, then kill the crons, then wait until it all settled, then restart the cron daemon, and restart the poller,...

I am running in a VirtualBox VM on CentOS 4.7 without these issues. The host is actually Windows Server 2003 (not my preference!), and on the non-virtualized machine, I am running a Netflow collection package that is receiving several thousands flows per second. The host has 4Gb of RAM, with 4 Xeon ...

I'm a month late again, but, the data is not wrong. The problem here is that you are trying to measure traffic of a virtual interface across which very little traffic actually traverses. The reason the data numbers seem so small is that most of the traffic will be forwarded via CEF on this device, u...

This is a bit old, but, the easier way is to adjust the bandwidth statement on the interface itself if it is a Cisco device. (other manufacturers devices may have a similar function/command) The bandwidth statement in no way affects or controls interface speed, but is there for use by routing protoc...

I missed this post back when you made it, but wanted to respond anyway, as I see you are not putting your percentiles next to your legend relating to them (and you may want it that way!). I did that by creating the appropriate formulas next to my legends for 95th in and out. I separated my 95th perc...

FWIW, you mention collisions, but keep in mind, that on Ethernet networks, collisions are not errors, but events. Ethernet is a CSCDMA technology, and collisons are expected, and transfer of data without collision detection would be very much a problem. (it's what the CD in CSCDMA means) Now, I know...

That actually brought to mind a 'Hold' and 'Resume' feature for when cycling is turned on. In the event someone has thier system set to cycle through maps at a fast interval, and sees a map with something 'red' (or for whatever reason) they may want to hold that map, to take a longer look. Of course...