We have inherited the maintenance of a WIFI Mesh made up of all Mikrotik products. The mesh is made up of 11 single and 3 dual radio devices starting with a Mikrotik Firewall running Mikrotik OS 6.7.

Most of the access points single and dual radio are running OpenWrt backfire r31348. Two dual radios are inexplicably running Mikrotik RouterOS 6.7.

The Dude is used to monitor the infrastructure. Sections have gone up and down and we have fixed the issue, by either replacing parts, reinstalling the OS or simply rebooting the device.

This morning every single device all 14 radios and the Firewall are coming up orange in THE DUDE, indicating that the CPU and disk are down.

Flag,Type,Problem,Notes
up,ssh,ok,
up,ping,ok,
up,dns,ok,
up,http,ok,
down,mikrotik,timeout,
down,router,timeout,
down,cpu,down,
down,disk,down,

Note The dude and the probe settings were also inherited

There is no impact on performance and every device behind the various radios are green in the DUDE. All devices and the radios themselves remain accessible.

We have no idea what gives. Any suggestions?

We first created new probes , the new probes said OK. However the new probes were different. They did not just check the presence of CPU , disk , memory and virtual memory , they were actually supposed to plot a graph showing percentage of usage. ( thanks to winkelman, probes provided by " just joined" ) http://forum.mikrotik.com/viewtopic.php?f=8&t=12402

Although the new probes came up green , they did not show any percentage of usage, not even on a Windows computer where we could actually see the usage in Task manager.

We had earlier performed a soft reboot of the Mikrotik Firewall running the Dude server, but that did not did not make a difference. Sent a Dude ( LOL) to perform a hard power reset on the Firewall running the DUDE server.
Lo and behold , everything is hunk dory , the new probes are now actually show a graph ( did not catch their absence earlier) and the old probes are all green. MY whole network is green.

I am now going to go and blow my brains out.

Hi Iansend, I have had similar problems when upgrading to 6.36rc16.

I have had lots of problems with The Dude and random things happening. I find that disabling the "graph bit rate" / "graph poll times"check box, applying, and then enabling clears lots of errors (along with all my graph history )

I actually had to restart all my graphs the above way because somehow some bad data was being inserted into future time in the graph, not sure how it happens.

I have CHR running on VMware Workstation, I have enabled SNTP in CHR, but still I get random errors.