Rebooting NS, watch dog, and client routers

[garymansperger]

99.9% of the time when a client calls to tell me the internet is down the problem is solved by having them first reboot the local rounter (I supply only a cat 5 from the outside radio - they either plug it direcley into a computer or more often purchas a wireless router for the house).

Sometimes they need to reboot the NS.

Question
Is this generally what the rest of you see?

Are you using the watch dog at client side to reboot the NS when it fails to reach the AP?

The largest number of problems are the wireless routers purchased by clients at local electronic stores. I see about the same failure rate from LinkSys, Netgear, Airlink, etc.

Happy new year
Gary

[WHT]

This might be an idea for a four or five point poll of subscriber solutions.

Mad Dawg and Rory could offer some input for the poll questions. Something like:

When a customer can't connect, what is the likely solution:
1) Reboot local router (all subscribers really should have a local router but they could use the UBNT station router mode)
2) Reboot the client radio
3) Set ping

And so on and so on....

The nice thing about the UBNT radios, is you don't have to require the user buy a separate router, even if it is a cheap (which might be the core of many issues) one .

[MaximumISP]

Not often do things require a reboot that I am aware of
but it does happen once in a while
mind you It could be more often than I think
since anyone who called and that worked for them is likely
to do that again first before calling next time
Usually rebooting just the internal router works

I only use the watch dog feature if I don't have the time
to investigate a repetitive issue more closely right away ...
In that case I will use it temporarily to keep a unit in operation

a unit that requires reboots too frequently is an indication to me that
something more is going on there and warents further investigation
automatic reboots are only a temp solution for me

I think one of the key things I do that helps alot is I maintain
management control of ALL internal wireless routers on the network

This has a number of benifits but mainly It allows me to fairly quickly
troubleshoot problems and also adjust internal wireless channels
to prevent any interferance issues from popping up
which did often happen previously

The problems I see are rarely from device failure
more often its due to configuration, hidden interferance or
poor link quality

[wildwisp]

Gary,

Where I WISP here the power quality is bad and power bumps routinely lock up anything with an ethernet port.

The Ping Watchdog in the NS2 is great for this.... I get fewer support calls from customers on NS2 running the watchdog.

To counteract the power bump lockups we:

- suggest everyone get UPS power strip (some do, it helps but isn't perfect)
- Configure routing in the NS2 CPE and then if the customer wants a wireless
or wired router inside we set it up basically as a dumb bridge with no DHCP, no NAT
- Setup NS2 with ping watchdog to local AP
- Put a recording on the customer support line that instructs customers how to reset their router and CPE radio before leaving a message

Good luck.