Thursday, June 21, 2012

Why Cisco Live is so much fun for Uplogix

Cisco CEO John Chambers visits the
Uplogix Launch Your Network booth
at Cisco Live. (Or at least visits
near our booth...)
Exhibiting in the World of Solutions at Cisco Live and talking with the network engineers, admins, CCNAs and CCIEs attending the show is totally within the Uplogix wheelhouse. They just get what we do. Sometimes it might take a minute of explanation, but they all come around to see that Uplogix is solving the issues they face on a daily basis.

Most of the conversations at our booth on the show floor go kind of like this:

[Attendee looking curiously at our demo gear, or maybe eying one of the slingshot astronaut toys we're giving away]

Uplogix: Hi, are you trying to figure out what Uplogix does?

Attendee: Yes I am. (No offense taken here - we're not one of the Big Names at the show, that's alright.)

Uplogix: We do network management automation from a local perspective - we deploy in the rack with the gear you care about and manage over the console port instead of using SNMP to poll from the NOC -- because that's using the network to manage the network, which seems kind of silly when you think about it.

Attendee: So you are managing out-of-band?

Uplogix: You've got it. Plus when you combine strong access control and intelligence & storage placed locally to managed gear, you can do some really cool stuff when it comes to configuration management (like automatic rollbacks of failed changes) and mass upgrades (through our Uplogix Control Center).

Attendee: Hmm. [Soaking all of this in.]

What we were giving away
at our booth.
Uplogix: And that's just the start of it. Since we're local, we can monitor intensively without putting all that traffic on the network. We'll know within 30 seconds of a problem and start down an abbreviated timeline of recovery, usually solving a problem before you even knew it existed with traditional network management.

Say we can't solve the problem and the network connection is down, the Local Manager will automatically spin up an out-of-band connection over just about anything short of smoke signals -- a POTS line, cell modem, a secondary network, even an Iridium modem if you are in the middle of nowhere.

Attendee: So you guys take the place of a console server, but do more than just get me access. Your product monitors anything with a console port, and does real work. This is pretty cool.

Uplogix: Now you've got it. Want me to scan your badge for some more information?

Attendee: Sure thing. Let me get my buddy over here and see a demo.

Proving we have a better mousetrap

The proof is in the conceptOne of the challenges we face at Uplogix is that the better mousetrap we bring to network management is such a leap from traditional network management that people really want us to prove ourselves. We've always been more than happy to do this, challenging potential customers to put us into their most troublesome sites and situations.

Over the years we've developed a rigorous proof of concept (POC) plan to ensure that all questions are answered and that we get to show off some of the cooler functionality of our product. Below are some of the evaluation scenarios from the POC project plan that deal with configuration issues. Depending on which analyst you choose to believe, configuration changes are the cause of many to most downtime events, so the config-related features of Uplogix are especially critical.

Configuration Loss or Corruption

A network device loses its configuration and becomes unusable, or a corrupt configuration file is accidentally distributed as the new standard. 

Uplogix Solution – Configuration Recovery
Issuing the recover configuration command on a port with a managed network device results in:
  • The Uplogix Local Manager (LM) cycles power to the network device
  • Breaks into boot sequence
  • Ignores current configuration
  • Network device comes up with no configuration 
  • LM pushes last known good configuration
Configuration Change with Human Errors

There have been incidents where a user accidentally commits a change that results in a network outage.

Uplogix Solution – Surgical Rollback
  • The rollback functionality is a configurable component of the network device driver.
  • The LM caches the current running configuration when initializing a terminal session. When the terminal session ends, a running configuration is pulled again and a list of changes made during the session is generated. The user is prompted to accept, reject or delay the changes made within the specified time. If time expires or the user rejects the changes, the LM backs out all uncommitted changes made during the session.
  • If the user session times out due to network failure or general inactivity after a configurable amount of time, the LM backs out all uncommitted changes made during that session.
Mass Configuration / OS Change

Situations arise that require the mass change or update of a configuration file, passwords or operating system. Common examples of these mass changes occur as a result of security requirements or during enterprise-wide device upgrades.

Uplogix Solution – Centralized Push
  • From the Uplogix Control Center (UCC) users have the ability to schedule a mass push or update to occur either instantly, or stage the change to take place at a scheduled time. 
  • In the case of OS and configuration file pushes, the system checks operational status after the change. If an error is detected, Uplogix rolls back to the last known good configuration or OS. 
  • The UCC allows the creation of custom filters to include or exclude specific devices needing the operation.
But wait there's more!
In a future post we'll cover the other scenarios in the Uplogix proof of concept project plan where we solve issues like primary outages, out-of-band, power management and AAA.

Friday, June 1, 2012

Local Management enables global entertainment

Uplogix Customer Case Study: E! Entertainment Television

Uplogix Local Management provides a network-independent platform that E!’s California-based IT staff uses to manage a network of global broadcast centers. Remote management of the broadcast centers is through Uplogix Local Managers that are directly connected to critical network infrastructure and installed with DSL out-of-band connections. Combining the out-of-band with automated features for monitoring and recovery, E! staff manages the remote sites 24x7.

Monitoring the network with the tenacity of a paparatzziUplogix gathers much more granular diagnostic data more frequently than SNMPbased systems without affecting the performance of the devices or the network. Analyzing this data, Uplogix can then either automatically resolve issues based on preapproved guidelines, or communicate the problem back to E!’s IT staff for resolution.

Remote access gives IT staff a (secure) backstage passUplogix Local Managers are collocated and directly connected to devices and servers to deliver persistent connectivity—regardless of the state of the network or device. Should the network go down, the LM immediately establishes remote connectivity to the E! network team via a secure out-of-band path over DSL.

Automation saves time, enables remote managementWith best-in-class problem diagnosis and recovery capabilities, Uplogix autonomously detects and fixes issues without having to deploy expensive resources onsite. Local Management enables E! to execute routine remote maintenance tasks like patch upgrades and configuration changes consistently and confidently.

For more, see the full case study in the Resource Center on the Uplogix website.