Hi all,

I am really new to Dude, so I don't know if my question has been answered before or not. Tried reading several posts, can't understand them. ^^

Anyway, I have this problem here (if you can call it a problem):
I'm using Dude to monitor the number of clients that are on or off at any given time (among other things). I usually do this using RouterOS's netwatch, but I just want a more graphical report. I can still do that at the moment, but The Dude seems kind of slow in updating the on/off status of each clients. My guess is it's something to do with the ping interval. Well, in routeros, I can set the interval setting in Netwatch to x seconds, but I can't find anything like it in The Dude. Tried looking all over the place, too.

Anyone can tell me where to set this interval, please?

Thanks.

higher powers correct me if I'm wrong but this is what I've found. if you go into the settings and modify the polling times this should do what you want it to do. you can do this on a global level or you can do it on each individual machine. I don't recall what the defaults are but there are three main settings in regards.

polling interval = how often it does it
polling Timeout = how long it'll take before the probe times out
polling Count = how many times it can fail before notifying you

this settings is specific i believe to all services running on a device. i've limited my services on all but a few to just ping. so say you setup the interval for 15 seconds timeout for 3 and count for 2. if a ping takes longer than 3 seconds it'll register as a fail and the device will re-list as unstable. then if the next ping takes longer than three seconds it'll it'll re-list as down. and you can set the notifications to tell you when it's unstable and when it's down and when it's up. the notification settings are actually rather intuitive if you can figure out how to use them. in the above scenario if you've got a machine down the longest you'll wait before dude turns it red and lists it as down is 30 seconds. keep in mind that settings this low could seriously affect your network traffic though. I personally have mine set (for PC's anyway) to interval 60, timeout 15, and count 3. so i'll know within 3 minutes if a PC has gone down. but as a note on how it can be adjusted i've got my core devices and routers set to interval 10, timeout 4, count 2. so i'll know in 20 seconds or less if my router has died on me.

hope that helps a little.

Neat, I'll try it and see how it goes.

Thanks so much for the answer.

Hmm, looks like it works. Thanks much beerfiend.

One question though:
If I want to update the status of my clients RIGHT NOW, what do I do? There are some instances when I need to know which clients are up and down immediately. I tried drag-selecting them all, then right-click, settings, then "reprobe". It seems to work, but there's still some delay. So it's not instant.

Is that the best the software can do?

Thanks.

might be a few different things i can think of. First is how fast is the server you're running dude on. if the servers lagging obviously your response times would too. second how many devices are you polling, how many devices are being monitored. i guess a limitation on the software would be that there is no escalation, such as polling a downed device every second until it comes back up. aside from doing a re probe it'll list as back up on the next polling interval. as for doing a reprobe, well I've got roughly 400 nodes between my five maps and when i do a reprobe the status response (provided the device is up) is nearly instantaneous. of course if a device is that critical you might want to set it to have a very low polling time too. just be careful not to overload the networking with snmp and ping polls, or the device for that matter. i had some Aironet1200's that if i set the polling at 5 seconds it would overload them, and crash my entire wireless net.

Gah, I forgot about this post entirely. ^^

Yeh, I tried doing a reprobe on a faster computer, and it's updated almost instantly. Guess it's the old computer that's causing the delay.

Thanks much for all your explanations. ^^