I try to look in to what I can do to enhance my IPFire experience and also make use of some of the addons listed on the Wiki. I am not a good friend of CLI but can make do if necessary, meaning I rather not install anything that way.
One of the first I looked at was a way to get e-mail notifications for downtime, but that seems to be currently unavailable as a Pakfire Addon, from what I read on the Wiki. Ipfstatusmail Is that still in working condition?
WIO, that has been installed since almost day 1, has some useful things sending e-mal notifications for UP / DOWN on selected devices. Took me some extra glaring on the icons to figure that one out… testing it via restart of my monitored NAS did not trigger any mail, perhaps it was just to fast to be detected, testing with a VM instead, leaving it off a few minutes. Got the WIO settings at checking interval 5 minutes and ping at 3 seconds. I think I would like to reduce the checking interval to 30 seconds but does not seem possible in the interface.
I realize that I should perhaps look at a “bigger” solution to achieve better monitoring, like Observium or Zabbix, but then again my essential devices are like half a dozen so feels a bit over the top.
I use email notification with several IPFire modules (including WIO, Proxy accounting and IPFStatusMail), that gives me overall satisfaction.
WIO : Refreshing the status can take more than a minute, depending on the number of devices / OVPN processed and the responsiveness of the network…
So I think that’s why the period can’t be less than 5 minutes (executing a new status refresh while the previous one hasn’t finished would not be a good thing) ; likewise, modifications are not possible during the execution of the refresh, which could cause a loss of control of the module.
Proxy Accounting : I still receive emails every month with the consumption and invoice of each entity.
IPFStatusMail : I actually haven’t received any notifications for some time and I’m afraid that the module is no longer functional.
Edit: For sending mail, I use the core function (GUI > System > Mail service) and not an add-on
in my case, I have activated email notification on all the NAS that I manage (Synology) and I receive real-time alerts for all the events I need (power loss (+ inverter), degradation of volumes, material defect, temperatures, attempted attacks [black listed], etc.).
Perhaps other manufacturers offer the same possibilities…
Note: In WIO, if your devices are on the same local network as IPFire, the 3 second timeout seems huge to me and you might have better global responsiveness by reducing it to 1 second
Well, having e-mail notifications on other devices is one thing, like your Synology example, I use Asustor.
But you need an external service to detect devices ping response and up/down-times. The timeout set by WIO is indeed on a local GB network so ping rate to 1 sec could of course be relevant, but the other setting, the time interval, can not be set lower than 5 minutes, not really sure what that is.
To test, you can edit the following file (at your own risk):
nano -l /srv/web/ipfire/cgi-bin/wio.cgi
and replace (line 975) :
for ($i=5; $i<=60; $i+=5) {
by…
for ($i=1; $i<=15; $i+=1) {
Thus, you now have a step of 1 to 15 mns for “Time interval for checking”
Note 1 : This change may be overwritten with the next WIO update…
Note 2 : Allowing less than 1 min is much more complicated
Note 3 : My proposal is not a suitable solution if you have many network devices to manage (the smallest values certainly too short between two refreshes)
I changed the title of this thread since I wanted to see if anyone can explain what the first time interval in WIO actually does? The upper one in below table…
When it is doing its check of all listed systems then you can not do any activity such as editing an entry or creating a new one. You will get the message that the action can’t be completed due to checking being ongoing.
So too frequent a check frequency will very frequently stop you making any updates or changes.
The time interval is from starting one checking to the start of the next check. If that is set at 1 minute and you have enough systems being checked that the check time takes 45 seconds then you only have 15 secs for doing any updates etc.
