If I choose the RV042 mode as load balancer and bind some protocol to specific WANs, what will happen to the protocols that I didn't bind to any specific WAN? will they go to WAN1 by default?
That will depend on what you choose to be the primary WAN on your network, if you go to the top specificly in under SYSTEM MANAGMENT tab on the router it self, you can use the WAN 1 or WAN2 to the primary one, but by default it is WAN 1.
If you have any further questions about this issues please feel free to contact us at 1866-606-1866 U.S/CA.
If RV042 runs in load balance mode, the traffic that is not bound to specific WAN port will be sent over 2 WAN ports by some type of round-robin.
Thank you tekliu, that's what I fugured out later after posting here.
I was wondering however, if I change the Kbit/Sec bandwidth of each WAN, will that only affect the likehoud of selection of the WAN (based on the round-robin) or something else also?
As I understood if I have both WANs having the same speed, the likehood of selection is 50/50.
So assuming I have WAN1 and WAN2 as 1024 Kbit/sec download, and I put WAN1 as 512 instead; then WAN2 will be selected more than WAN 1 correct?
When Load Balance is enabled, RV042 will dispatch (using weighted Round Robin) incoming sessions to WAN1 and WAN2 according to the ratio of the WAN bandwidth. A session is defined by a TCP connection or a UDP stream with certain timeout.
I found this in the users guide.
* Dual WAN. There are two functions provided for users – Smart Link Backup and Load Balance..
* If Smart Link Backup is selected, you only need to choose which WAN port is the primary and then the other will be the backup. See Figure 6-22.
* If Load Balance is selected, there will be two main choices:
• By Traffic – Intelligent Balancer (Auto) and user defined. See Figure 6-23. First, choose the Max. Bandwidth of Upstream (64K/128K/256K/384K/512K/1024K/1.5M/2M/2.5M or above) and Downstream (512K/1024K/1.5M/2M/2.5M or above) for WAN1 and WAN2, as provided by your ISP.
• Intelligent Balancer (Auto): When choosing Intelligent Balancer, it will automatically compute the maximum bandwidth of WAN1 and WAN2 by using Weighted Round Robin to balance the loading. If (upstream / downstream / upstream or downstream) bandwidth is excessive (30%, 40%, 50%, 60%, 70%, 80%, 90%), bring up the second link. When there is an inactivity time-out (None/10min/20min/30min/40min/50min/60min), the second link will be terminated.
Adding to this is the operation of Network Service Detection (NSD), which allows RV042 to ping the liveliness of the default gateway of the WAN port or a specified Host. The retry timeout and retry count will affect how long RV042 determines the WAN is down. By default the detection time is
(Retry timeout) + (Retry count) * (5 sec) // the 5 sec is hard coded (which is about 1 minute)
Please, someone educate me just a little bit. Here is the deal. I have a customer who has a rural setting. Actually this is a hunting camp and I have cellular routers connected to both Verizon and AT&T. So not "necessarily" the greatest bandwidth. So the customer is talking about load balancing between the two cellular Internet providers. So help me with this. Load balancing will distribute the individual "sessions" between the two WAN interfaces. Now here is where it gets techie. Does the router see each WAN connection as basically the same. So lets say it has two gigabit connections. Does it see each pipe as being equal regardless of how capable the Internet connection attached to each wan port might be. Will it distribute traffic based on the saturation of the Internet connection rather than the saturation of the rated WAN connection?
Now correctly me if I am wrong, but with Load balancing user A is directed to WAN port X. All his traffic will go through that WAN port regardless. I look forward to the discussion.