I seem to be having issues with Dude on my multi-core ESXi VM. I'm running the CHR with 4 assigned cores, but there are times, usually when updating a custom function, when one of the cores max out and Dude hangs, even though the other three cores are barely being used. Often times this lasts for several minutes causing several monitored devices to be listed as down. I tried reducing the number of services I'm monitoring and increasing the time between probes. I've included a snapshot cpu use and one of the my customer functions. I don't think it's the scripts fault as this occurs even with my simple custom functions.


Function: Brand
------------------------------------------------
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Brother")<>4294967295,"Brother",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Cisco")<>4294967295,"Cisco",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"EPSON")<>4294967295,"Epson",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"TOSHIBA")<>4294967295,"Toshiba",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Windows Version")<>4294967295,"Microsoft",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"ADP Printing System")<>4294967295,"ADP",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"RouterOS")<>4294967295,"Mikrotik",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Meraki")<>4294967295,"Meraki",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Dell")<>4294967295,"Dell",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"EdgeSwitch")<>4294967295,"Ubiquiti",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"EdgeOS")<>4294967295,"Ubiquiti",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Xerox")<>4294967295,"Xerox",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Hewlett Packard")<>4294967295,"HP",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"HP ETHERNET")<>4294967295,"HP",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Ubiquiti")<>4294967295,"Ubiquiti",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"USW-24")<>4294967295,"Ubiquiti",
if(string_find(oid("1.2.840.10036.3.1.2.1.2.10"),"Ubiquiti")<>4294967295,"Ubiquiti",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Fortinet")<>4294967295,"Fortinet",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"VMware")<>4294967295,"VMware",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Aerohive")<>4294967295,"Windows 10 v1703",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"Juniper Networks")<>4294967295,"Juniper Networks",
if(string_find(oid("1.0.8802.1.1.2.1.3.4.0"),"Juniper Networks")<>4294967295,"Juniper Networks",
if(string_find(oid("1.0.8802.1.1.2.1.3.4.0"),"Netgear")<>4294967295,"Netgear",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"PowerConnect")<>4294967295,"Dell",
if(string_find(oid("1.3.6.1.2.1.1.2.0"),"dell")<>4294967295,"Dell",
if(string_find(oid("1.3.6.1.4.1.674.10892.2.1.1.4.0"),"DELL")<>4294967295,"Dell",
if(string_find(oid("1.0.8802.1.1.2.1.3.4.0"),"ProSafe")<>4294967295,"Netgear",
if(string_find(oid("1.3.6.1.2.1.1.1.0"),"SonicWALL")<>4294967295,"SonicWALL",""))))))))))))))))))))))))))))

Did you set up the VM cores to be of the same CPU "package"/"socket"? Kernels act differently between multiple cpus on multiple cpu sockets vs multiple cpus/cores on a single package.

Last working on CCR version dude is 6.45.9. After years of fighting the NV2 TDMA mikrotik trash, the migration to Cambium has followed. Now it's time to finish the dude server. The next one will probably be CCR.