NSA310S suddenly not accessible at LAN IP (but responds to ping)

erkme73

For the second time in as many weeks, my NSA310S cannot be reached with any browser, or from windows file explorers on various PCs.   

The first time it happened, I power-cycled the NAS, and when it booted back up, everything was working as it should.

Now that it's happened a second time, I'm trying to run diagnostics to figure out what's causing the problem before power-cycling again.   Here's what I've figured out so far:

1. NAS has a static IP (fixed at device and static DHCP via router).
2. Ping responds with <1ms response
3. Trying to access NAS at LAN IP from any local PC on the same LAN (and same IP subnet) using multiple browsers results in no connection - though it never times out or errors out... tab just spins and spins.  
4. Attempts to SSH fails "FATAL ERROR: Network error: Connection refused" - though that may be normal behavior as I don't recall whether I have SSH enabled on the NAS

Without accessing the web GUI, short of any other efforts that I'm unaware of, I cannot confirm the current firmware version.

Can anyone think of a scenario where these issues might occur?

Mijzelf

Do you have any new devices in your network? I have seen this behaviour on my NSA325 (which happens to have the same kernel), and as far as I could find the TCP stack was damaged.

According to Wireshark connections where refused on closed ports (like your attempt to connect over ssh) but not established on open ports (causing the browser to wait forever). Meanwhile UDP still works. And ping, which is ICMP.

The only reason I can think of which causes this is some malformed or badly timed TCP packet. On my network the problem stopped before I could really investigate it.

erkme73

Thanks for the reply.  Shortly after posting my request for help, I discovered a router in my barn - which I had been using as a guest router (vs. access point) - which I recently moved.  When I plugged in the LAN data connection, I accidentally plugged it into one of the LAN ports instead of the WAN port.  Thus, I had two DHCP servers on my network.  The barn was on a different range (10.0.0.x) vs. the remaining network (192.168.x.x).


That shouldn't  have affected the scanner since its IP was statically assigned by my primary/gateway router - but I think that's what happened.  Connecting the scanner via USB cable and verifying its IP confirmed it had jumped to the 10.0.0.x subnet.  

I tell ya, if it weren't for self-inflicted problems, I'd probably have a much simpler life.

Thanks again.

Mijzelf

erkme73 said:

That shouldn't  have affected the scanner since its IP was statically assigned by my primary/gateway router

A static assignment in a DHCP server only means the IP is reserved for, and handed over, to a specific MAC address. The DHCP client doesn't know about this, and will try to renew it's IP address when the leasetime is half over. Normally it gets the same address again, but if you indeed have 2 DHCP servers that is not guaranteed.

erkme73

Yup, that's exactly what I suspected happened.  For grins, I've added a true static address at the NAS now so no matter what, it'll maintain the correct one.

But still, doesn't explain why the GUI and file sharing stopped, but the ping at the original correct IP was still responding.

ETA:  The ip was statically assigned in the NAS... So there goes that theory.  Maybe it's on the blink.  

ETA2:  Now that I'm logged into the admin GUI, I'm poking around for the first time in years.  The CPU seems pegged at 100% - nothing less than 90%.  I'm not using this as a media server, so I'm not quite sure why it would be that high.

ETA3:  Looks like I somehow turned on Twonky media share.  I disabled it and CPU dropped to 6%.

ETA4:  Well, that didn't take care of the CPU afterall...