WANFailover Dual WAN Failover Script

[Pak Kriss]

1. No I don't monitor bandwidth.
2. You can set ping timeout to 1 second and 3000ms pings would be considered packet loss.

OK. then I understood your script and readme correctly.

I thought I leverage the 1 second timeouts and max. ICMP package size.

Question: it's possible to define a packet loss amount? Percentage or amount or something similar, to define down state

Feature request: Switch from seconds to milliseconds scheme

 

[Ranger802004]

OK. then I understood your script and readme correctly.

I thought I leverage the 1 second timeouts and max. ICMP package size.

Question: it's possible to define a packet loss amount? Percentage or amount or something similar, to define down state

Feature request: Switch from seconds to milliseconds scheme

The % calculated is based on how many pings are dropped so if you are doing a PINGCOUNT of 6 and 1 is missed, that will be flagged as 16% loss, etc.

The ping build on the firmware doesn't have a millisecond option but for a later release I have been looking into tracking the ms and adding an option for setting a max average.

 

[Pak Kriss]

The % calculated is based on how many pings are dropped so if you are doing a PINGCOUNT of 6 and 1 is missed, that will be flagged as 16% loss, etc.

The ping build on the firmware doesn't have a millisecond option but for a later release I have been looking into tracking the ms and adding an option for setting a max average.

1. Which variable I can define 50% (1-100%) packet loss should be considered as WAN down state?

2. Network Tools - Network Analysis - may use another ping binary:
PING www.google.com (216.239.38.120): 56 data bytes
64 bytes from 216.239.38.120: seq=0 ttl=57 time=38.474 ms
64 bytes from 216.239.38.120: seq=1 ttl=57 time=38.397 ms
64 bytes from 216.239.38.120: seq=2 ttl=57 time=38.246 ms
64 bytes from 216.239.38.120: seq=3 ttl=57 time=38.307 ms
64 bytes from 216.239.38.120: seq=4 ttl=57 time=38.278 ms

--- www.google.com ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 38.246/38.340/38.474 ms

 

[SomeWhereOverTheRainBow]

1. Which variable I can define 50% (0-99%) packet loss should be considered as WAN down state?

2. Network Tools - Network Analysis - may use another ping binary:
PING www.google.com (216.239.38.120): 56 data bytes
64 bytes from 216.239.38.120: seq=0 ttl=57 time=38.474 ms
64 bytes from 216.239.38.120: seq=1 ttl=57 time=38.397 ms
64 bytes from 216.239.38.120: seq=2 ttl=57 time=38.246 ms
64 bytes from 216.239.38.120: seq=3 ttl=57 time=38.307 ms
64 bytes from 216.239.38.120: seq=4 ttl=57 time=38.278 ms

--- www.google.com ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 38.246/38.340/38.474 ms

if you really want to ping it, and ping it hard we can add a custom size.

ping -c 6 -s 65500 www.google.com
PING www.google.com (216.239.38.120): 65500 data bytes

 

[Ranger802004]

if you really want to ping it, and ping it hard we can add a custom size.

ping -c 6 -s 65500 www.google.com
PING www.google.com (216.239.38.120): 65500 data bytes

PACKETSIZE is a configuration option in my script lol

 

[Ranger802004]

1. Which variable I can define 50% (0-99%) packet loss should be considered as WAN down state?

2. Network Tools - Network Analysis - may use another ping binary:
PING www.google.com (216.239.38.120): 56 data bytes
64 bytes from 216.239.38.120: seq=0 ttl=57 time=38.474 ms
64 bytes from 216.239.38.120: seq=1 ttl=57 time=38.397 ms
64 bytes from 216.239.38.120: seq=2 ttl=57 time=38.246 ms
64 bytes from 216.239.38.120: seq=3 ttl=57 time=38.307 ms
64 bytes from 216.239.38.120: seq=4 ttl=57 time=38.278 ms

--- www.google.com ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 38.246/38.340/38.474 ms

I'm trying to work within the stock binaries as much as possible.

 

[SomeWhereOverTheRainBow]

PACKETSIZE is a configuration option in my script lol

You could also include interface as well.

 

[Ranger802004]

You could also include interface as well.

They are already specified for each WAN interface monitor.

 

[amplatfus]

Hi,
On latest v1.5.6-beta12 I am flooded with WAN Failover Notification, even the secondary WAN is Hot-Standby:

WAN0 Status: CONNECTED
WAN1 Status: DISCONNECTED

 

[Ranger802004]

Hi,
On latest v1.5.6-beta12 I am flooded with WAN Failover Notification, even the secondary WAN is Hot-Standby:

WAN0 Status: CONNECTED
WAN1 Status: DISCONNECTED

I will need debug logs to diagnose the issue

 

[Pak Kriss]

I'm trying to work within the stock binaries as much as possible.

Perfect, since I believe this is in stock firmware since decades.

Navigate via web gui to network tools or via ssh open a shell.

Both way return standard ping with milliseconds.

Which point I exactly miss?

 

[Ranger802004]

Perfect, since I believe this is in stock firmware since decades.

Navigate via web gui to network tools or via ssh open a shell.

Both way return standard ping with milliseconds.

Which point I exactly miss?

The echo/responses are measured in milliseconds and the timeout parameter is specified in seconds. I am going to consider down the road adding in the ms as a factor to be measured but not for v1.5.6

 

[Pak Kriss]

The echo/responses are measured in milliseconds and the timeout parameter is specified in seconds. I am going to consider down the road adding in the ms as a factor to be measured but not for v1.5.6

Perfect.

No rush.

I love your work.

 

[amplatfus]

I will need debug logs to diagnose the issue

Thank you for reply.
Please, how can I put the script into debugging mode?
Because I see only logs from the booting time.
I mention I have syslogng for logging with logging file in opt var log messages

Thank you!

 

[lbtboy]

v1.5.6-beta12 notifications configured via amtm are not coming, although test notifications are coming.

 

[Stephen Harrington]

v1.5.6-beta12 notifications configured via amtm are not coming

Working fine for me on Beta 12 ... in fact sometimes I get TOO many, as @Ranger802004 knows!

 

[lbtboy]

Working fine for me on Beta 12 ... in fact sometimes I get TOO many, as @Ranger802004 knows!

I am very glad that it works for you and that in fact you get too much. But I state the fact that neither more nor less works for me.

monitor.txt

drive.google.com

 

[VIper_Rus]

I pulled out a 4g modem from my wan1 device, every 15 seconds I began to receive a letter about the failure of wan1. Those. not one mail will come, but an infinite number until wan1 is restored

 

[lbtboy]

Disabling wan1 there are no alerts.

monitor_off_wan1.txt

drive.google.com

 

[VIper_Rus]

Disabling wan1 there are no alerts.

monitor_off_wan1.txt

drive.google.com

I lied, I didn’t get an infinite number of mail, but 14. Then everything stopped. After 20 minutes, I inserted a 4g modem into my wan1 Chinese router and received 1 more mail about recovery