I am using a Max BR1 Mini with my connection currently provided via Wifi as WAN from a Sprint Hotspot. This ran rock-solid for me for about two months.
In the last few weeks, however, I started getting “WAN Failed DNS Test” errors here and there, and in the past week, almost continuously to the point where my rock-solid connection has become almost unusable.
I initially thought this was a Sprint issue, but I decided to connect some machines directly to the Sprint hotspot, and they do not seem to have the problem when the machines connected to the BR1 Mini still do.
We pulled in a firmware update a few weeks ago and maybe this is when the problems started. We have been really pleased with the BR1 Mini and this Sprint solution (never could get Verizon card working in it). I would like to get it back to the stable condition it was in before.
Any suggestions for troubleshooting and resolving this?
Agreed with Martin, normally is the WAN connection or health check destination failing the WAN health check. WAN health check is just like a guard dog help you to monitor the WAN connectivity. Guard dog will bark when found some stranger and it’s same that WAN health check packets sent and doesn’t get reply then this will failed the WAN health check.
Further isolation need to do to confirm the issue.
Change health check method as suggested by Martin.
We disabled the health check yesterday and really haven’t seen a problem since. Health check appeared to be checking every 5 seconds and was really choking the connection. Also appears that the timeout can’t be set to more than 10 seconds and for whatever reason, Sprint kept failing with that, then the connection would get reset and choke out.
I like the thought of having the health check but am really hesitant to have it back on with any settings if we’re going to experience what we were over the past couple of weeks - by yesterday when I posted and we turned it off in desperation, the connection was basically not usable. Students couldn’t even pull in simple syncs on their tablets and we were just at a loss after having things run with no issues for so long.
It may be something on Sprint’s end, but without the health check resetting things, it’s apparently not bad enough to really notice…
It sounds like if anything the Peplink has highlit a problem with your carrier DNS! There are plenty of public DNS services am sure you know such as OpenDNS and others like that. Turning on DNS caching on the unit will help the clients DNS resolution speed if theirs are flaky.
As Sit Loong mentioned if you only have a single WAN connection you can happily turn it off, its used as a check/decision maker to switch traffic to another connection, if you dont have one connected then its a bit of a pointless check.
Hi guys, thanks again for the help. There is no backup WAN for the device. The connection is very rural and options are severely limited. We purchased the Peplink to use with a Verizon card that we were ultimately unable to troublshoot problems with (throughput dropped to essentially 0 every 20-30 minutes). Then we went with the Sprint connection which is through a non-profit for educational status. But we are using their hotspot (a ZTE WarpConnect) and I worry that putting the card directly into our Mini might cause us to lose the unlimited connection or get throttled… so we’re using Wifi as WAN. The ZTE Warp Connect with NOT let me change the DNS. Also, I’ve read that even if you change it, Sprint still hijacks. Things seem better right now though we’re still having some weird issues with wireless devices (that isn’t happening with wired connections) and timeouts here and there. I will contact Sprint if things seem to get really bad again.
Same thing happening with my peplink balance 710. We have 2 ISP’s and one our isp is giving this Wan failed dns test message. It occurs only 3-5 times every night between 8pm to 12am . We are using dns server provided by our ISP. It happens for almost 3 weeks until today. I hope anyone here can help me to fix this issue. Thank you
It appears you’ve posted essentially the same message twice.
As an initial step, we’d probably recommend changing to PING from DNS and set a target of well known and reliable hosts e.g., 8.8.8.8 and 1.1.1.1. In general, we have not found DNS inquiries of ISP DNS servers to be the best solution for WAN health checks.