6.6.1 on Ultra 6 Loses DNS lookup after a few days

[jamestx10]

I have my router configured as the DNS server in the network settings. It can reach the router fine but it fails to resolve addresses after a few days. I reboot the NAS and it is good to go again for a few days.

Anyone else seen this? Is there a service I can restart on the NAS rather then rebooting the whole thing?

[Retired_Member]

Why not trying to run with a static ip-address on the nas for a few days and see, whether that helps.

[jamestx10]

The NAS IP is reserved in DHCP by the router

The DNS setting on the NAS is hard coded to point to the IP of the router

I would prefer to let my Router manage the IPs unless you have found that this would fix the DNS lookup issue?

[Retired_Member]

I share your preference. I cannot offer a solution. This is just a suggestion.

My idea is: Currently the router is responsible for assigning the ip address to the nas. Now it would be interesting to see, whether the problem continues after you take away that responsibility from the router and leave the nas with a fix ip address.

If that helps, you a) would have a temporay fix and b) could continue to research for something which gets you closer to your preferred configuration.

[jak0lantash]

I would suggest the same thing... "for troubleshooting purposes".

Try to switch your NAS to static IP for now, and confirm if the issue still occurs or not. That would help narrowing down the root cause.

[jamestx10]

I have changed it from DHCP to Static now.

[jamestx10]

That did not help. A few days later and the device cannot resolve names. I have not rebooted it yet any thing I should check now in the broken state?

[jak0lantash]

Firstly, you should download the logs, at least for reference.
If you want to check via SSH, you could try to:
- ping google.com
- ping 8.8.8.8
- ping your default gateway
- ping what's supposed to be the DNS server

[jak0lantash]

You can also dump the current status of all network connections:

for a_service in $(connmanctl services | awk '{ printf $3" " }'); do connmanctl services $a_service; done

[jamestx10]

This time the unit locked up so I had to reboot it. I set the IP back to DHCP. Running your other command I noticed that I had not updated my NTP address and that the 2nd ethernet connection had a router and DNS entry set that it cannot connect to. I have Eth1 directly connected to a PC with a cross over cable. I removed the router and DNS entries from that connection. I am wondering if that was the issue??? How are the network connections weighted? Which link will it try by default? I see they are both listed as favorite.

/net/connman/service/ethernet_001f33ea8125_cable
  Type = ethernet
  Security = [  ]
  State = ready
  Favorite = True
  Immutable = False
  AutoConnect = True
  Name = Wired
  Ethernet = [ Method=auto, Interface=eth0, Address=00:1F:33:EA:81:25, MTU=1500 ]
  IPv4 = [ Method=dhcp, Address=10.10.10.20, Netmask=255.255.255.0, Gateway=10.10.10.1 ]
  IPv4.Configuration = [ Method=dhcp ]
  IPv6 = [  ]
  IPv6.Configuration = [ Method=off ]
  Nameservers = [ 10.10.10.1 ]
  Nameservers.Configuration = [  ]
  Timeservers = [ 10.10.10.26, time-a.netgear.com, 10.10.10.1 ]
  Timeservers.Configuration = [ 10.10.10.26, time-a.netgear.com ]
  Routes = [  ]
  Routes.Configuration = [  ]
  Domains = [ NULEAF ]
  Domains.Configuration = [  ]
  Proxy = [ Method=direct ]
  Proxy.Configuration = [  ]
  Provider = [  ]
/net/connman/service/ethernet_001f33ea8124_cable
  Type = ethernet
  Security = [  ]
  State = ready
  Favorite = True
  Immutable = False
  AutoConnect = True
  Name = Wired
  Ethernet = [ Method=auto, Interface=eth1, Address=00:1F:33:EA:81:24, MTU=1500 ]
  IPv4 = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0 ]
  IPv4.Configuration = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0 ]
  IPv6 = [  ]
  IPv6.Configuration = [ Method=off ]
  Nameservers = [  ]
  Nameservers.Configuration = [  ]
  Timeservers = [ 10.10.10.26, time-a.netgear.com ]
  Timeservers.Configuration = [ 10.10.10.26, time-a.netgear.com ]
  Routes = [  ]
  Routes.Configuration = [  ]
  Domains = [  ]
  Domains.Configuration = [  ]
  Proxy = [ Method=direct ]
  Proxy.Configuration = [  ]
  Provider = [  ]

Once I removed that Router and DNS entry from Eth1 I could no longer resolve DNS So it really seems like it wants to use that link

root@NAS:~# ping google.com
ping: bad address 'google.com'
root@NAS:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=46 time=9.523 ms
64 bytes from 8.8.8.8: seq=1 ttl=46 time=9.473 ms

So next I disconencted Eth1 and then I could ping again. Any ideas on how to have the two connections used but favor Eth0 for internet and DNS?

[jamestx10]

Is there a service on the NAS I can restart that should correct this?

[Retired_Member]

The nas is not the problem, but the way you have setup your network topology. 

The only way to solve this I see in having two completely separated networks and that also means addresswise. eth0 should be on one network together with your router. eth1 should be on a second isolated network holding only eth1 and the workstation.

I assume you only have one nic in your workstation and that directly connects to eth1 via the crossover cable. In this case give the workstation a static ip address and do the same to eth1. Make the workstation the router and gateway of eth1. Make eth1 the router and gateway of the workstation. Make sure both ip addresses of eth1 and the workstation are not on the same network than the router and eth0.

For example: If you router is on 192.168.0.1 and the eth0 is on 192.168.0.2 the static ip addresses for the workstation could be 192.168.1.1 and for the eth1 192.168.1.2

Should you have 2 nic in your workstation, one of them directly going to the nas and one going into the router's network, disable the one into the router's network while trying to implement the proposed solution.

[jamestx10]

That is how I have it connected.

Eth0 goes to the router where it has a DHCP reserved IP of 10.10.10.20

Eth1 goes directly to the workstation with a crossover cable and static IP of 10.101.1.3

Workstation has 2 eth ports with one going to the router and one going to the NAS with an IP of 10.101.1.2

[jak0lantash]

---

@jamestx10 wrote:

/net/connman/service/ethernet_001f33ea8125_cable
  Ethernet = [ Method=auto, Interface=eth0, Address=00:1F:33:EA:81:25, MTU=1500 ]
  IPv4 = [ Method=dhcp, Address=10.10.10.20, Netmask=255.255.255.0, Gateway=10.10.10.1 ]

/net/connman/service/ethernet_001f33ea8124_cable
  Ethernet = [ Method=auto, Interface=eth1, Address=00:1F:33:EA:81:24, MTU=1500 ]
  IPv4.Configuration = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0 ]

---

Your configuration is correct.

Eth0: 10.10.10.20/24, GW 10.10.10.1, DNS 10.10.10.1

Eth1: 10.101.1.3/24, no GW, no DNS

This is exactly the configuration I have on my NAS and it's working OK.

You could try to restart connmand, but it's just easier to reboot the NAS.

After the reboot, double check that both NICs still have the same configuration, from the GUI and using the command I give you.

Try again to resolve google.com with a ping.

[Retired_Member]

In case you want to exclude the slot in router the nas' nic going to is the culprit, just try a different slot on the router.

In case you want to exclude any of the nics involved being the culprit you now got to do the works.

I do not know which nics are currently talking to each other, but I would just start with swapping the tasks the nas' nics are currently conducting, as the nas is loosing the ip address.

That means, the nic in the nas currently talking to the router should talk to the workstation only and the other nic in the nas should talk to the router.

There are the following cases

With nas_eth0 talking to the router nas_eth1  can either talk to ws_eth0 or ws_eth1 (2 cases to check)

Case1: nas_eth0 to router and nas_eth1 to ws_eth0

Case2: nas_eth0 to router and nas_eth1 to ws_eth1

With nas_eth1 talking to the router nas_eth0  can either talk to ws_eth0 or ws_eth1 (2 cases to check)

Case3: nas_eth1 to router and nas_eth0 to ws_eth0

Case2: nas_eth1 to router and nas_eth0 to ws_eth1

[jamestx10]

The NAS does not lose the IP it only loses the ability to resolve DNS entries. I can still access shares and SSH into it. I just can no longer resolve a DNS name.

Fresh boot, SSH into NAS, Ping google.com, resolves fine.

A few days later, CrashPlan cannot connect, SSH into NAS, Ping google.com, no resolution.

Every other device on my network can resolve fine so I don't think it is a router issue or port issue. I did recently change the LAN subnet from a 192.168.x.x to 10.10.x.x but do not see anything in the GUI pointing to the old address and like I mentioned it works fine for a few days.

[jak0lantash]

---

@jak0lantash wrote:

---

@jamestx10 wrote:

/net/connman/service/ethernet_001f33ea8125_cable
  Ethernet = [ Method=auto, Interface=eth0, Address=00:1F:33:EA:81:25, MTU=1500 ]
  IPv4 = [ Method=dhcp, Address=10.10.10.20, Netmask=255.255.255.0, Gateway=10.10.10.1 ]

/net/connman/service/ethernet_001f33ea8124_cable
  Ethernet = [ Method=auto, Interface=eth1, Address=00:1F:33:EA:81:24, MTU=1500 ]
  IPv4.Configuration = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0 ]

---

Your configuration is correct.

Eth0: 10.10.10.20/24, GW 10.10.10.1, DNS 10.10.10.1

Eth1: 10.101.1.3/24, no GW, no DNS

This is exactly the configuration I have on my NAS and it's working OK.

You could try to restart connmand, but it's just easier to reboot the NAS.

After the reboot, double check that both NICs still have the same configuration, from the GUI and using the command I give you.

Try again to resolve google.com with a ping.

---

[jamestx10]

It is currently in a failure state. Cannot resolve DNS names.

root@NAS:~# ping google.com
ping: bad address 'google.com'
root@NAS:~# ping 8.8.8.8
PING 8.8.8.8 (8.8.8.8): 56 data bytes
64 bytes from 8.8.8.8: seq=0 ttl=58 time=2.615 ms
64 bytes from 8.8.8.8: seq=1 ttl=58 time=2.498 ms
64 bytes from 8.8.8.8: seq=2 ttl=58 time=2.539 ms
^C
--- 8.8.8.8 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 2.498/2.550/2.615 ms

I can SSH in on both Eth0 and Eth1 so the network link is up. I can also access the web GUI and it shows no change in the network settings. I DL the logs but have no idea where to look in them for any issues.

 for a_service in $(connmanctl services | awk '{ printf $3" " }'); do connmanctl services $a_service; done
/net/connman/service/ethernet_001f33ea8125_cable
  Type = ethernet
  Security = [  ]
  State = ready
  Favorite = True
  Immutable = False
  AutoConnect = True
  Name = Wired
  Ethernet = [ Method=auto, Interface=eth0, Address=00:1F:33:EA:81:25, MTU=1500 ]
  IPv4 = [ Method=dhcp, Address=10.10.10.20, Netmask=255.255.255.0, Gateway=10.10.10.1 ]
  IPv4.Configuration = [ Method=dhcp ]
  IPv6 = [  ]
  IPv6.Configuration = [ Method=off ]
  Nameservers = [ 10.10.10.1 ]
  Nameservers.Configuration = [ 10.10.10.1 ]
  Timeservers = [ 10.10.10.26, time-a.netgear.com, 10.10.10.1 ]
  Timeservers.Configuration = [ 10.10.10.26, time-a.netgear.com ]
  Routes = [  ]
  Routes.Configuration = [  ]
  Domains = [ NULEAF ]
  Domains.Configuration = [  ]
  Proxy = [ Method=direct ]
  Proxy.Configuration = [  ]
  Provider = [  ]
/net/connman/service/ethernet_001f33ea8124_cable
  Type = ethernet
  Security = [  ]
  State = ready
  Favorite = True
  Immutable = False
  AutoConnect = True
  Name = Wired
  Ethernet = [ Method=auto, Interface=eth1, Address=00:1F:33:EA:81:24, MTU=1500 ]
  IPv4 = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0 ]
  IPv4.Configuration = [ Method=manual, Address=10.101.1.3, Netmask=255.255.255.0, Gateway=10.10.10.1 ]
  IPv6 = [  ]
  IPv6.Configuration = [ Method=off ]
  Nameservers = [  ]
  Nameservers.Configuration = [  ]
  Timeservers = [ 10.10.10.26, time-a.netgear.com, 10.10.10.1 ]
  Timeservers.Configuration = [ 10.10.10.26, time-a.netgear.com ]
  Routes = [  ]
  Routes.Configuration = [  ]
  Domains = [  ]
  Domains.Configuration = [  ]
  Proxy = [ Method=direct ]
  Proxy.Configuration = [  ]
  Provider = [  ]

[jak0lantash]

Not sure if that's the issue, but there is indeed a difference between now and earlier. Eth1 shouldn't have a default gateway and it didn't, but now it does. Any chance you could run Wireshark on the PC connected to eth1, ping google.com on the NAS and look at Wireshark to see if you receive DNS queries from the NAS?

[jamestx10]

Well the issue is worse then I first thought. I did not have time last night to wireshark so I left it. This moring the NAS has locked up. No GUI, No SSH, No respond to ping, even the front power button would not make the display light up. Hard reboot and it is back up.

So it seems that the failure of DNS to resolve is just one of the first signs that I get that the system is going to lock up in the next 12 - 24 hours. I downloaded logs again as soon as the system was back up but I don't know what to look for in them. Should I try reinstalling 6.1.1

[jak0lantash]

I don't know anything about running OS6 on a legacy NAS. That's too specific for me to answer. Sorry

[jamestx10]

Thanks for your help. Hopefully someone else stumbles on this thread. @mdgm-ntgr

[jamestx10]

Nothing has changed that I am aware of and my system has been up for almost 2 weeks now.

[jamestx10]

Made it just under a month and locked up again requiring a power cycle to get it back up.