Reduce Downtime with up.time

Agreed that downtime cannot be avoided, but it can be reduced by monitoring

network devices regularly. There are several tools and procedures, which can be

used for monitoring and up.time is one such server/client monitoring solution

supporting both agent and agentless monitoring.

It is easy to use and implement, and can monitor network devices through SNMP

(Simple network management protocol). It has support for Windows 2000/2003, AIX,

Linux, Solaris, AIX and FreeBSD 5.x for server monitoring, while for

workstations, it has support for Windows XP, Linux and Solaris. This tool can

monitor Application Servers, Host Services, Windows, Databases and Network

Services. For Application Server monitoring it has default monitors for Exchange

Server such as SMTP local queue length, SMTP bytes sent/received per second,

current SMTP inbound and outbound connections.

Likewise, for database monitoring the tool has support for MySQL, Oracle, SQL

Server and Sybase. It can monitor databases for various parameters such as Total

Transactions persec, Table Space usage, user connections etc. up.time also

supports SNMP monitoring and comes with SNMP MIB Browser and has default

monitors for IIS, WebLogic, WebSphere and ESX v3.

Whenever a failure occurs, up.time immediately sends an alert so you can

handle the problem effectively and resolve it instantly. The alert could be in

the form of an e-mail, SMS, a Windows pop-up message or you can even page it.

The automatic alert escalation system of this utility takes care of an alert

when not addressed within a particular instant of time. Moreover, you get the

liberty to create custom monitors for running custom scripts and generating

specific alerts. In addition, this utility also provides User Access Control,

which lets you create users and assign specific roles as well as access rights

to them.

In Global scan, you can see real time view of all machines being monitored with their trends over the past 24 hours

How to use



We installed up.time on Windows 2003. It took us only 15 minutes to have the

software up, running and start monitoring. When you run it for the first time,

it will ask you to define a password for admin login and provide an e-mail id on

which notifications should be sent. Before you start monitoring, you first need

to configure the Archive policy to define for how long the data should be

monitored. For this go to the menu bar, click on the CONFIG tab and then on

Archive policy, specify the number of months for which you want to keep the

statistics, check the 'Enable Archiving' option and click on 'Set Archive

Policy' button. Now to start monitoring, you need to add

systems/services/network devices that you would like to monitor. For this click

on ' MY ENTERPRISE ' tab in the main Web console of up.time and then select

'Auto-Discovery' option from the pane on left. Now in the display on the same

page provide Network Address and Default SNMP read community fields. In Network

Address field specify the broadcast IP address of the network, in the next field

specify SNMP community string for the SNMP enabled devices present on the

network, and then click on 'Continue' button. It will scan your network and

detect all systems and devices attached. To add a new system to the up.time,

click on the 'Add' button present on the display where name of the detected

machine appears. This opens up the 'Add System' window, here in the 'Type of

System/Device' field select the 'Agent' option if the system that you are adding

runs up.time agent. For agent based monitoring, you have to manually install

agent on the systems you want to monitor and then add those systems to up.time.

You can see performance summary of each machine being monitored through easy to understand graphs

Once you have added machines, you can see them under the Global Scan window.

In Global Scan panel you can see the live summary of systems with details such

as number of recent service outages, CPU usage, hard disk usage, memory usage

etc. To view complete details of the system, click on the name of the system.

For viewing a system's performance in graphs, click on the 'Graphing' option

and then select the component whose graphical details you want to see. For

instance, if you want to view free memory of the machine, go to Memory menu,

select 'Free' option, then choose the time period for which you want to see the

graph, and then click on 'Generate Graph' option. A new window displaying the

details of free memory will open up.

This report shows CPU utilization of all servers across the enterprise and power units consumed by their CPUs

Resource usage reports



You can also use this tool for generating comprehensive reports based on the

data collected over a period of time. To view reports click on the REPORTS tab

and then click on 'Resource Usage' option present under the pane on left. On

right hand side display you have the 'Date and Time Range' option to choose the

period for which you want to generate reports. In the 'Report' option, you can

select reports on CPU Performance, Disk Statistics, TCP Retransmits etc, select

the systems that you want to include in the report, from the List of Entities.

To see the reports right away, click on 'Print to Screen' button present

under the 'Next in Generate Now' option. This will generate the report and

display it in your Web browser. Similarly, you can also create reports in PDF,

XML and e-mail them. If you want to schedule automatic generation of reports

then you have to check the 'Scheduled Report'option and schedule report

generation on a daily, weekly or monthly basis.