I'd like to know if maybe we could have a new function in the next firmware release to get a watchdog functionality over ePMP!
We work for a WISP in Italy and we're using ePMP a lot lately, even in ePTP mode to get ptp link from tower to tower (with light data throughput). We've noticed that sometimes we've to reboot the ePTP devices since they stop working together.
Watchdog functionality in this case would be REALLY very nice! Like the one's over Ubiquiti devices.
There are some known issues with ePTP mode that have been resolved in the latest beta... try that out OR if you use TDD-PTP mode and flexible timing, that will solve the issue as well.
Better use a external ping watchdog. I suggest Netonix WISP PoE switch that include the watchdog feature and will reboot your radios if those stop pinging.
We already use Mikrotik RB750 in our small CPE, they're perfect with POE ports.
But on large installations where we use RB2011 or RB1100, we use gigabit POE injectors because they're cheaper than POE switches... anyway we didn't know those products... I'll take a look. Thanks.
By the way, I still think it'd be nice to have a watchdog functionality directly on ePMP radio.
Yes, regardless of a hardware rebooting solution at the remote side, Cambium absolutely still needs to add a 'test' mode, and a 'Ping/Watchdog' function. Â WISP's often need to be able to test different settings and need to be able to trust that if they pick a bad setting, that the radios can reset back to the 'last saved' or 'last known good' settings, and they can relink.
We are going to have all issues releated to ePTP mode fixed in the nearest future, so there is no sense to spend time for watchdog development, instead of this we will be totally focused on bug fixing.
Test mode is already included to short list of features Cambium Networks is going to deliver.
Hello everyone, I would like to add something important concerning our network that includes many point-to-point links, and that in the future will increase more and more. It is frustrating when you have to spend so much time in controlling that the antennas' operativity does not exceed 100 days for example, as this is often cause of malfunction. To overcome this problem, it would be great to have an instrument other than watchdog.... for instance an additional function giving the opportunity to reboot the system "every XX days at 4am" . This would allow us not to lose valuable time. Also watchdog would be great. Thank you
I understand your point, but maybe you could provide as much debug information as possible about your particular issue in order to solve it instead of asking for development time aiming to implement dirty work-arounds.
Hello everyone, I would like to add something important concerning our network that includes many point-to-point links, and that in the future will increase more and more. It is frustrating when you have to spend so much time in controlling that the antennas' operativity does not exceed 100 days for example, as this is often cause of malfunction. To overcome this problem, it would be great to have an instrument other than watchdog.... for instance an additional function giving the opportunity to reboot the system "every XX days at 4am" . This would allow us not to lose valuable time. Also watchdog would be great. Thank you
If you've got a linux/unix server available you can do something pretty simple like adding the following cronjob:
1 4 1/16 * * for ip in "10.10.12.1" "10.10.15.3" "10.18.12.3"; do /usr/bin/snmpset -v2c -cprivate $ip .1.3.6.1.4.1.17713.21.4.1.0 i 1;done
This would trigger on the first of each month at 4:01am, (then again on the 17th, 16 days later - for once a month just put in the day-of-month number by itself) and step through the list of IPs and send them a reboot command via SNMP. (of course you have to have snmpwrite installed, and change the path if needed to find it)
Interesting your vision ... we do not use Linux systems to control the network, we have a server with Win dude. Are there any alternative solutions, but for me the idea of automatic reeboot is essential. We have hundreds of active links, and every day someone is not working properly causing disconnections to our users/client. Â And 'so complicated include this function ?? I'm not a programmer and do not know the complexity of your fantastic work ... and I apologize if I'm asking difficult things.
âTEST MODEâ and âWATCHDOGâ functions are much needed. I mean, i find ePMP to be very very stable, but still⌠we need to be make changes and test, without sawing the branch we are standing on. And, in the event of a FUBAR, we need systems to be able to reboot themselves.
I'd like to see a test mode as well... as far as a watchdog feature is concerned, I've never had an issue with an AP or SM that needed this feature, and typically if you need it there are other more serious issues that need to be addressed on your network. We do have all of our AP's plugged into remote managed power devices (packetflux rackinjectors), but I can't remember the last time we ever had to remote power cycle a Cambium device.
By âwatchdogâ I mean a ping watchdog, so that if its unable to ping a certain programmed IP for a proframmed time limit, itâll reboot.
We use thst on other systems, particularly on PTP links, where the far side might be an hour away. That way we can make setting changes and âactivateâ without saving⌠and if its unable to ping the AP for x minutes, it will reboot back to the last known good settings. Same functuon as âtestâ mode in a way, but a bit different.
