ISPPlanet • Budget Network Management System Series - Wrap-Up

Monitoring and Alerts
Ad hoc and real-time queries are helpful to pinpoint a problem or better understand how a device is operating. But, at the heart of every NMS is an event monitor that issues periodic polls and/or receives asynchronous notifications. Every product in this series supports periodic polling. All can determine device reachability with ping and/or SNMP. Most can determine server status with basic port-level polling (if the listening port is absent, the service must be down).

ipMonitor, WhatsUp Gold, and the NetInspector Internet Plug-In (beta) go a step further. Each can evaluate service health using application-level protocols. These pollers transfer a test file, retrieve a web page, or query a RADIUS server with varying levels of proficiency. ipMonitor is the master in this arena. While not directly comparable, creative tools in the SNMP ToolChest evaluate router health, end-to-end reachability, and network delay for selected services.

Vendor	ACE*COMM	MediaHouse	SNMP ToolChest Company	MG-SOFT	Ipswitch	Castle Rock Computing
Product	NetPlus AutoManage	ipMonitor	SNMP ToolChest	MIB Browser NetInspector LITE	WhatsUp Gold	SNMPc
Device polling	SNMP	Ping, SNMP	SNMP	Ping, UDP Echo, SNMP	Ping, IPX, NetBIOS, SNMP	Ping, IPX, SNMP
Service polling	No	SSL, LDAP, DNS, SMTP, IRC, RADIUS, HTTP, POP, DNS, FTP, IMAP, SQL, Telnet, DriveSpace, URL, Notes, Active Directory, Kerberos, Finger, Gopher, NTP, WhoIs, SNPP	RouterScope, IPScope, NetDelay	Via Internet Plug-In: SMTP, POP, FTP, HTTP, NNTP, IMAP, DNS, Time, Daytime, Quote, Echo	SMTP, POP, FTP, HTTP, NNTP, IMAP, DNS, Telnet, Gopher, Echo, HTTP content, RADIUS, SSL, NT Services, SNMP Threshold, Custom	SMTP, FTP, HTTP, Telnet,4 user-specified
Dependent polls	No	Yes	No	No	Yes	Yes
SNMP TRAPs	Yes	Yes	No	Yes	Yes	Yes
Correlated TRAPs	Yes	No	No	No	Yes	Yes
Bubble-up status	Map	Groups	Buttons	Map	Map	Map
Event List	Scroll Individually	History	IPScope	Current, History	Event Log	Current, History
List filters	Source, Context, OID	Time, Monitor, Status, Action	None	Severity, Time	Device	Device, Severity
Alert methods	Forward Trap	Mail, Page, ICQ, PopUp, Log, Forward Trap, Run Program, Auto Report	None	Mail,HTML Reports	Mail, Page, Sound, PopUp, Run Program, Voice Message	Mail, Page, Sound, PopUp, Run Program, Forward Trap
Alert escalation	No	Yes	No	No	Yes	Yes
Thresholds	No	Yes	Yes	No	Yes	Yes

Only SNMP ToolChest lacks an SNMP TRAP handler; the others factor incoming TRAPs into device status. NetPlus AutoManager, WhatsUp Gold, and SNMPc can self-clear correlated TRAPs (e.g., clear linkDown upon receipt of linkUp). NetPlus AutoManager, ipMonitor, and SNMPc can forward TRAPs, enabling event integration by HP OpenView or another enterprise manager.

You might staff your NOC 24x7, but you don't want to rely on a human to notice status changes. A good NMS lets you specify criteria that trigger alerts to keep you informed and initiate corrective action. ipMonitor, WhatsUp Gold, and SNMPc provide strong alert handling—they can notify you by email, page, pop-up window, or all three in escalating sequence.

Alerts can be a double-edged sword. Frequent, inconsequential alerts caused by transient conditions can often be avoided with monitor thresholds (e.g., page me only if utilization exceeds 90% for 10 minutes). SNMPc will even observe trends during a learning period to establish baselines and thresholds.

Pt. 1: Introduction / What We Tested	Pt. 5: Device Control <
Pt. 2: Installation & Setup	Pt. 6: Network Montoring & Alerts
Pt. 3: Network Discovery	> Pt. 7: Reporting
Pt. 4: SNMP Browse and Query	Pt. 8: Product Support / Overall Evaluation