0

We recently had one of our locations go down for ~4 hours over the weekend due to a switch failure. Despite this being our agreed upon outage window, management was not happy with the outcome for a number of reasons.

This location is remote from all company IT resources so we had to send our remote Cisco vendor out to reprogram the Distribution switch replacement that was sent with an employee from a neighboring location.

Like they often say the best way to get your company interested in fire alarms is to have a fire. Management is now very interested in reducing this "unacceptable" time table as much as possible (They asked for 10 minutes, we laughed[Secretly of course].)

One of the things I have proposed in the past is to add a few layers of Out-of-Band to each location such as:

  • A Laptop/Micro PC with a 4G modem and Teamviewer connected with console cables to equipment.

  • A POTS line connected to a modem based Out of Band console server.

Our infrastructure is mostly centralized using VMWare in a single data center, we use SIP based VoIP.

I am having trouble finding the best way to be able to dial up and get a console session on a console server that has a built in modem from a virtual machine (WIN10). Management was very clear that we be able to connect Anytime/Anywhere to any location to have immediate access to program or troubleshoot switches.

I am ok with any solution that can solve this problem, including paid services, equipment, or Voodoo magic. I believe it would be acceptable to purchase a pots line for our data center but it would need to be able to be used from the VM.

We have a Cisco Call Manager in house but as we only use SIP we do not have a router that handles any T1 lines or other resources (Everything is over the DS3 we have).

0

There seems to be a disagreement about the service level here. Come to a compromise on what it should actually be before implementing something. Risk analysis of the downtime plus cost estimates of potential solutions will help guide the discussion.

Be sure to include costs of any additional staff needed to provide that kind of response time.

For this specific failure, a spare pre-programmed switch on site might be possible to install with remote hands. Automatic failover could also be done with a bit more design complexity.

I don't know how you would implement dial in from a Windows VM. The solution doesn't necessarily have to be a VM, if a small server with a modem works. Nor does it have to be dial up, if you can connect redundant cable/DSL/LTE connections from different providers to the branch routers.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.