webiis
asked on
Sonicwall TZ 190
I have one subnet 192.168.2.0
I have two TZ190 routers with both WAN and OPT being assigned public IP's from 4 different providers.
TZ190 router 1 has a gateway of 192.168.2.1
TZ190 router 2 has a gateway of 192.168.2.2
Several servers on my LAN. some servers use 192.168.2.1 as the gateway and the router 1 has NAT policy's with ports for web server as an example
Several servers on my LAN use 192.168.2.2 as the gateway and the router 2 has NAT policys with web email servers as an example.
All i have is the two sonic walls and since they are both on the same subnet I have an ethernet cable going from router1 to router2. All my servers are pluged in to router 1 and I run vmware ESX.
Everything has been working fine for months. Suddenly, yesterday router 1 stopped responding and all the servers that were using that router went down. I logged in to one of the servers by coming in from router2 and did a continuous ping on router 1 192.168.2.1
It would stop responding for many pings and then would respond every 4th 8th ping with a long time like 800ms. I replaced the router with an exact one and imported the exact configuration and the problem was gone. Next day at 8:00 AM the issue kinda came back so rebooted the router this time and it went away. Same day at 5PM the router refused to respond and then I replaced it as well with a new one. This was 3 hrs ago and everything is fine. I'm terrified that since I don't know the root cause this issue will appear again. Just to recap this configuration and setup has been running the same way for months without an issue.
Any suggestions?
I have two TZ190 routers with both WAN and OPT being assigned public IP's from 4 different providers.
TZ190 router 1 has a gateway of 192.168.2.1
TZ190 router 2 has a gateway of 192.168.2.2
Several servers on my LAN. some servers use 192.168.2.1 as the gateway and the router 1 has NAT policy's with ports for web server as an example
Several servers on my LAN use 192.168.2.2 as the gateway and the router 2 has NAT policys with web email servers as an example.
All i have is the two sonic walls and since they are both on the same subnet I have an ethernet cable going from router1 to router2. All my servers are pluged in to router 1 and I run vmware ESX.
Everything has been working fine for months. Suddenly, yesterday router 1 stopped responding and all the servers that were using that router went down. I logged in to one of the servers by coming in from router2 and did a continuous ping on router 1 192.168.2.1
It would stop responding for many pings and then would respond every 4th 8th ping with a long time like 800ms. I replaced the router with an exact one and imported the exact configuration and the problem was gone. Next day at 8:00 AM the issue kinda came back so rebooted the router this time and it went away. Same day at 5PM the router refused to respond and then I replaced it as well with a new one. This was 3 hrs ago and everything is fine. I'm terrified that since I don't know the root cause this issue will appear again. Just to recap this configuration and setup has been running the same way for months without an issue.
Any suggestions?
ASKER
Good point! From any server I can ping router 192.168.2.2 without issue. The one that has the issue is 192.168.2.1. When I ping I'm pinging from many servers that are plugged in to 192.168.2.1 and the result is the same, separate cables, all in a data center rack. There is not switch on the LAN side. Just a cable between the two sonic walls. Each sonic wall has an 8port LAN switch built in.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
oh and my particular favorite - a loose power cable sufficiently loose such that whenever a datacenter employee walks past your rack, the device reboots.
i had that happen once, took out an MSA cabinet of hard drives... demanded to see the video tapes, and then the datacenter babysitter admitted that my door looked loose so he pushed on it - didn't realize he pushed a server 2008 dvd box which flipped the switch on the back
i had that happen once, took out an MSA cabinet of hard drives... demanded to see the video tapes, and then the datacenter babysitter admitted that my door looked loose so he pushed on it - didn't realize he pushed a server 2008 dvd box which flipped the switch on the back
So, how do you have the sonicwalls connected to each other? If you have the LAN interfaces configured on the same subnet AND you have other servers connected to them, then you must have them all connected to the same switch, right? have you looked at the switch as a problem? have you looked at half/full duplex or setting the speed of the port?
so, what was the solution? i see lots of possibilities in the solution you chose.
did you replace the wire that connects the two routers? is it straight-thru and you're relying on auto-crossover? what if you use a real crossover cable instead? what if you put a dumb 5port switch in between the two routers?
did anyone run a new power cable right next to the ethernet cord that connects the routers? looking for some kind of interference here. it's probably not the router you replaced... is it the router you didn't replace yet? do constant pings to other things stay up, while the ping to the other router went down?
is the problematic router on the bottom of a stack of stuff, maybe getting too hot?