today i noticed one of my production VM's had crashed a couple of times (see crash reports attached).
the default WAN seems to be acting up, and after a zillion times going up->down->up  the VM ends up with a Fatal Kernel Trap.
I also seem to notice a lot more CPU usage on up–>down-->up events. This wasn't the case on 2.1 AFAIK, i didn't notice it till today on 2.1.3

An hour later it happened again. Bad isp's shouldn't trigger this behaviour right ?
Could someone take a look at the the TXTs attached please

Also noticed a BUG in the desciption or calculation of the openvpn-client "limit outgoing bandwidth" (this is what caused the gateway hopping in the first place …

the description specifies:

Maximum outgoing bandwidth for this tunnel. Leave empty for no limit. The input value has to be something between 100 bytes/sec and 100 Mbytes/sec (entered as bytes per second).

In reality if you fill in a value of 50000 , this should get you around 50KB/s ==> in reality you get around 50Kbit/s or 5KB/s. after further testing, it appears it isnt a simple as that, its just completely off
when I input a value of <=1000000 (one million)  around <=30KByte/s upload over the vpn (client –> server)
when I input a value of  1500000 (one point 5 million) i get around 100KByte/s upload over the vpn = max upload speed of WAN (client --> server)
does anyone else notice the similar inconsistencies ?

crashreport_gvl.txt
crashreport_gsh.txt