Polisy Rude Network Behavior

[whywork]

I use and enjoy my ISY every day.

However, polisy is acting rudely on my network.

Thousands of DNS queries to my.isy.io, proxy.isy.io, proxy2.isy.io.

Yesterday ran pkg update upgrade.  Today my pihole reports 51K queries to my.isy.io from my polisy box. seems like every 2 seconds

2021-07-06 17:55:40    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:37    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:36    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.1ms)    
2021-07-06 17:55:35    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:33    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:32    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.3ms)    
2021-07-06 17:55:30    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.3ms)    
2021-07-06 17:55:28    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:21    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)    
2021-07-06 17:55:21    A    my.isy.io    192.168.1.194    OK (cached)    IP (0.2ms)

Cold rebooted polisy.  Still multiple rapid DNS queries.  (my network topology has not changed)

 

Suggestions?

[Michel Kohanim]

Hi @whywork,

Apologies. This is the side effect of ISY Portal being active by default on the ISY port on Polisy. Please do this for now:

sudo service isy stop

With kind regards,
Michel

[whywork]

Hi,

Thanks for the quick reply.  

A suggestion - Polisy should be respectful of the network whether it is successful or not in connecting.

Use a fallback mode, for example:

Try to connect every second for 1 minute

Try to connect every minute for 5 minutes

Try to connect every 10 minutes for an hour

Try to connect every hour for as long as polisy is booted.

After 1 hour - send an email alert to the administrator that connection is unsuccesfull.

 

 

[albrandwood]

Also, it should be respecting the DNS and caching the results for the TTL specified time...

edit: Guess I should have checked the DNS first ... Why is the TTL for a CNAME to an AWS ELB set to 59seconds ? 

 

Edited July 29, 2021 by albrandwood