Router DHCP woes?

[scyto]

With my RC68 it works brilliantly, however I noticed with the latest few firmware's some strange DHCP oddities:

After some significant uptime (+15 days) machines suddenly drop off the network (,y synology NAS, my windows machine - not all mind you and those machine all revert to AIPA addresses of 169.x it appears that the outer is no longer issuing DHCP leases. Machines that have different lease periods seem to stay connected. This DHCP issue appears to affect both wired and Ethernet connections.

Looking in the log stuff gets weirder, for example my router was working perfectly until ~ 2015-07-16 23:17:03 EDT however it appears all logging ceased on August 7th (some 9 days ago)

Aug 7 09:08:37 dnsmasq-dhcp[480]: DHCPREQUEST(br0) 192.168.1.134 d8:a2:5e:01:c1:07
Aug 7 09:08:37 dnsmasq-dhcp[480]: DHCPACK(br0) 192.168.1.134 d8:a2:5e:01:c1:07 Caroles-iPad-2
Dec 31 16:00:24 syslogd started: BusyBox v1.20.2
Dec 31 16:00:24 kernel: klogd started: BusyBox v1.20.2 (2015-07-16 23:17:03 EDT)

This discontinuity is basically the reboot event. I assume some critical processes are getting stuffed up, which eventually breaks logging and eventually breaks DHCP over the longhaul?

Is this a known issue? this has happened over several releases (last 3 I think).

Alex

 

[steve500]

With my RC68 it works brilliantly, however I noticed with the latest few firmware's some strange DHCP oddities:

After some significant uptime (+15 days) machines suddenly drop off the network (,y synology NAS, my windows machine - not all mind you and those machine all revert to AIPA addresses of 169.x it appears that the outer is no longer issuing DHCP leases. Machines that have different lease periods seem to stay connected. This DHCP issue appears to affect both wired and Ethernet connections.

Looking in the log stuff gets weirder, for example my router was working perfectly until ~ 2015-07-16 23:17:03 EDT however it appears all logging ceased on August 7th (some 9 days ago)

Aug 7 09:08:37 dnsmasq-dhcp[480]: DHCPREQUEST(br0) 192.168.1.134 d8:a2:5e:01:c1:07
Aug 7 09:08:37 dnsmasq-dhcp[480]: DHCPACK(br0) 192.168.1.134 d8:a2:5e:01:c1:07 Caroles-iPad-2
Dec 31 16:00:24 syslogd started: BusyBox v1.20.2
Dec 31 16:00:24 kernel: klogd started: BusyBox v1.20.2 (2015-07-16 23:17:03 EDT)

This discontinuity is basically the reboot event. I assume some critical processes are getting stuffed up, which eventually breaks logging and eventually breaks DHCP over the longhaul?

Is this a known issue? this has happened over several releases (last 3 I think).

Alex

Alex,
How many devices do you have, and how many are connected at a time on average?

 

[scyto]

37 +/- devices offline at any one time.

the devices alive are across one subnet that is serviced by two asus routers (one configured as AP),

I had one Hi-flying electronics technology Co.,Ltd device with MAC AC:CF:23:0F:E5:5D appears as 23 devices with 23 addresses (same MAC for all).

I figured out what it is, reconfigured and rebooted it, so I just eliminated that issue (it was a brultech green eye energy meter).

Also I have one dumb device that does stupid number of request and acks... nothing I change about that device though - that's what it does for better or worse.

Aug 16 20:04:50 dnsmasq-dhcp[1098]: DHCPOFFER(br0) 192.168.1.75 cc:fb:65:78:ff:60
Aug 16 20:04:58 dnsmasq-dhcp[1098]: DHCPREQUEST(br0) 192.168.1.151 00:1e:c0:10:86:8c
Aug 16 20:04:58 dnsmasq-dhcp[1098]: DHCPACK(br0) 192.168.1.151 00:1e:c0:10:86:8c PCMWIFIG
Aug 16 20:05:39 dnsmasq-dhcp[1098]: DHCPREQUEST(br0) 192.168.1.151 00:1e:c0:10:86:8c
Aug 16 20:05:39 dnsmasq-dhcp[1098]: DHCPACK(br0) 192.168.1.151 00:1e:c0:10:86:8c PCMWIFIG
Aug 16 20:06:30 dnsmasq-dhcp[1098]: DHCPREQUEST(br0) 192.168.1.151 00:1e:c0:10:86:8c
Aug 16 20:06:30 dnsmasq-dhcp[1098]: DHCPACK(br0) 192.168.1.151 00:1e:c0:10:86:8c PCMWIFIG