Sporadic timeout in internal network to reach BR200/BR500 (see attached network map)

[Aristomastrosi]

Dear community,

You were very helpful in the past, therefore I try it again. I need some troubleshooting advice on the following after many tries, but the problem persists on my network with 6 netgear devices (see map attached):

• During normal network load, I was getting from my macbook (ethernet wired) around every 20-30min no internet access for 15-20 seconds (estimate)
• I stared to constantly ping my router (internal IP), and I get an ICMP timeout after 15 seconds (destination not reachable). It is not only one package, all packages during the time of 15-20 seconds are not going through.
• This issue is also noticable with wireless clients on either of the access points.
• I have not found a reliable way how to force this error (e.g. ICMP floods with large package sizes do not trigger the behaviour more often).

I have attached you a map of my network including VLAN and SSID configurations. The host above (macbook) is attached to switch "2nd Floor, 192.168.1.242" on port 3 (untagged VLAN 20). Furthermore, the router is receiving a public IP address at the internet port. STP is disabled.

I'm very thankful for any hints or suggestions how to troubleshoot. I can also capture packages (e.g. on the switch "master") with port mirroring if you think that might be useful.

Thank you very much and best regards,

[Aristomastrosi · Accepted solution]

Finally I found the solution. The issue does disappear once I changed from "Auto Detect" to "Auto Config" in the IPv6 Config of the BR200. It seems that "Auto Detect" is creating this lag every 15min a DHCP lease extension is received from the ISP. I think this is a bug, because auto detect should not bring down the router for 5-15 seconds after such a harmless message.

Unfortunately I was unable to file a bug report (purchase date older than 3 month). Can anyone help me how to contact netgear that they are aware of that bug? I'm happy to assist.

[Aristomastrosi]

Dear community,

I now have captured all traffic going in/out of the BR200 on all three ports and VLANs.

Monitoring Setup:

- Port Mirroring of Port 1-3 (Master Switch) to Port 5 (capture)

- ICMP pings from 192.168.20.32 (wireless client)

As you can see, at time 13:03:39.272945, the router on 192.168.20.250 becomes unreachable (visible in the screenshot) and resumes operation 5-15 seconds later.

From long-term tests I can can confirm that his happens exactly every 15min.

I can't see any suspicious activity before on any of the VLAN's, floods or any race condition in ARP requests or similar.

Thanks for any hints!

[Aristomastrosi]

Finally I was able to isolate the problem. The ISP has a lease time of 30min, but extends every 15min. Exactly at this time, the BR200 logs "Internet connected". This is the time where all ports go down for a couple of seconds (destination not reachable). I attached you a snapshop of the BR200 logs.

Configuration or BR200:

- IPv6 for the ISP (Auto Config).

I'll file a bug report since I'm unable to do further investigations, the log function of the BR200 is quite limited (no syslog or debug logs as far as I can see).

[Aristomastrosi · Accepted solution]

Finally I found the solution. The issue does disappear once I changed from "Auto Detect" to "Auto Config" in the IPv6 Config of the BR200. It seems that "Auto Detect" is creating this lag every 15min a DHCP lease extension is received from the ISP. I think this is a bug, because auto detect should not bring down the router for 5-15 seconds after such a harmless message.

Unfortunately I was unable to file a bug report (purchase date older than 3 month). Can anyone help me how to contact netgear that they are aware of that bug? I'm happy to assist.