Peplink balance one WAN shutdown


#1

I have one balance one with Comcast Fiber Ethernet and every couple of days WAN goes down up - very frustrating since this is not Comcast problem. At the same time when connecting a laptop to temp switch, all is fine. I can confirm WAN ISP is ok. This is peplink issue. It happens randomly every a week, two weeks, sometimes a couple of days.

The warranty has just expired. That means almost $500 was sacrificed for nothing.

is there any known issue for balance one ? thank you


#2

Can you share the screenshot of the event log (Status > Event Log) which shows that the WAN is down?


#4


#5

screen shot attached


#6

also:

192.168.168.60 1065
192.168.168.196 340
192.168.168.82 237
192.168.168.44 177
192.168.168.40 130

these are the sessions opened !!! 1065 there is NO virus on that client - its clean!!! It smells like firmware bug - not properly closing sessions and killing WAN (fiber) . I remember there was a bug in ASA 3 years ago doing same thing open sessions bug affecting ASA to kill WAN port. Well, warranty just expired a month ago so I cannot call support. Extending warranty does not make sense, but buying another balance one also is under question with full respect to peplink.

These people have a lots of tabs opened and it makes sense whats happening.

In ASA I could remote using cli and kill active sessions, here I dont know the trick and the only thing what I can do is restarting peplink and after that, all comes to normal. But doing that during business hours for a company which relies their business on internet connection is just so frustrating. I may consider another brand which may be a more stable. Since this is not cheap but in $500 there are some good choices.

I read that place in the rack it may be sensitive to some electric interference from other devices, so I place - relocated it on the top of the rack and zip tighted.

Please feel to be in my shoes. I look like an idiot in front of my boss. They upgraded link to fiber which costs almost $1000 for same problem as they had with regular bandwidth paying $280 and all points to peplink failing. We have WAN2 with some not perfect satelite backup, but using that is so painful since all goes down for minutes. We changed satellite to coax more stable and again same thing.

there are 40 devices and 100/100 speed so all is in proper requirements. I spend days installing prtg and syslog servers but for a sake I dont see anything wrong besides the message the WAN 1 goes down … The speed at the time of downtime or just before is about 40 - 50 MB so half of our bandwidth.

Any thoughts? And I know its not ISP issue.


#7

For the issue of “WAN shutdown”, it is related to Balance One failed to ping the health check target. The health check tools is used to monitor the WAN status for WAN failover purpose. It is same when you issue a ping from your laptop to a target IP address then got the “Request timed out” (Balance One is showing “WAN failed PING test”). You may refer to the link below for further troubleshooting.

For the issue of client active session, 1065 sessions should be accepted. These are the sessions that generated by the client. You may go Status > Active Sessions > Search to check what sessions are generated by the client. You may need to take action to the client if the generated sessions are not legitimate.


#8

wow I never seen that post, the problem is it had happened before with Comcast coax ISP with Arris modem, and now happy with Comcast fiber where they use Juniper router. I can place managed Juniper switch in between, but I have feeling it’s not going to do anything.

Anyway - downloaded paesler webstress tool and did 4000 users requests -

yes - WAN 1 failed - WAN FAILED HTTPS TEST !!! bingo problem reproduced !!! At least it looks like

Should I place the managed switch between the ISP and Peplink ???


#9

the only problem is there is already small 8 port UNMANAGED switch between Comcast fiber and peplink balance one:

Trendnet GREENnet TEG-S80G 8-Port Gigabit

should I then place managed one?


#10

I dont know how these total sessions are counted but:

192.168.168.48 1676 total sessions when searched I have Total searched results: 61

and PC has only 8 google tab opened, checked its clean.

Any though


#11

Once a while peplink balance one kills WAN1

Comcast Fiber: Disconnected (Health check failed)


#12

even now - for last 35 minutes Peplink WAN was connected - disconnected - rebooting Peplink fixed problems for now.

I dont want to change firewall but looks like I have to start looking for another firewall brand since this is getting out of control.


#13

Hello MST,

Have you placed the managed switch between the modem and the Peplink? If you have not yet done this, I would recommend doing that.


#14

ok, I will work on getting managed switches, but I have to know if I have to change “ping” from peplink to managed switch MGTM IP? I am talking about health check settings.

is there any other configuration I should do on the managed switch?

thank you


#15

I would recommend setting up the ping to an IP address outside of your network. For any Health Checks that use IP addresses I would recommend using your ISP’s DNS IP as the primary record and something like Google, CloudFlare, Quad9, etc, as the second record.


#16

get it now, I thought it has to be the ping to managed switch MTGM IP, but if I can use 8.8.8.8 or 75.75.75.75 is just fine.

Thank you


#17

i will be connecting managed switch this weekend and let you know. So far what I have noticed while WAN1 and WAN2 bounced like crazy - the only fix was to reboot the Peplink device.

I really hope managed switched fixes these problems in that place. They rely on outside connection and that is not stable.