Tampilkan postingan dengan label Others. Tampilkan semua postingan
Tampilkan postingan dengan label Others. Tampilkan semua postingan

Minggu, 11 Juli 2010

Redistribution Overview

- Redistribution rules and guidelines:
 -- Redistributed routes cannot be redistributed again on the same router (RIP > OSPF > EIGRP)
 -- Manual Split-Horizon - Never redistribute a prefix injected from domain-A into domain-B back to  domain-A.
-- You cannot change the EIGRP external AD (170) per-route. It can only be done for all prefixes or none.
--  Sub-optimal routing in the lab is not a problem unless specified, as long as there is full reachability.
-- The redistribute command redistributes only routes which are in the router’s current IP routing table (RIB).
-- Before enabling any redistribution, make sure each protocol have full reachability within itself.
-- The metric assigned using the "redistribute metric" command takes precedence over metrics assigned with the "default-metric" command.

- When redistributing into RIP and EIGRP:
-- The metrics must be set via configuration as RIP and EIGRP have no default values.
-- RIP cannot use a 0 metric, the hop count must be between 1 and 16.
-- The 0 metric is also incompatible with the EIGRP multi-metric format.

- When redistributing into OSPF:
-- By default, routes are redistributed into OSPF as external type-2 (E2) routes, with a metric of 20.

- These logical steps happen when redistribution is enabled:
-- The router ONLY looks at the routing table to get the routes that are to be redistributed.
-- Not all the routes that the redistributed protocol sends to the routing table, will be redistributed.
-- Verify what routes, with "sh ip route " before redistribution is enabled.
-- The router takes all connected subnets matched by that routing protocol’s network commands.
-- Verify these interfaces, by looking at the redistributed protocol's network statements OR look at the individual routes with "sh ip route x.x.x.x" as listed with "advertised by".
-- Passive-interfaces for the redistributed protocol ARE included when redistributing.
-- This hidden step is the equivalent of the following: (Hidden: H>)
#router ospf 1
#redistribute rip subnets
    >redistribute connected subnets route-map NAME
    >route-map NAME permit
    >match interface fa0/0 s0/1 <--- All the RIP enabled interfaces
-- If ever asked to redistribute specific interfaces, ALWAYS INCLUDE the interfaces that the redistributed
protocol runs on.
- Mutual Router Redistribution
-- Redistributing from a low AD protocol(eg OSPF) to a higher AD protocol(eg RIP) won’t cause feedback as the lower AD is always preferred.
-- But redistributing from a high AD protocol to a low AD protocol could create problems, because the high AD protocol routes might prefer the redistributed low AD routes to a destination.

Rabu, 16 Desember 2009

key element for proactive engineering of IT services

It is a key element for proactive engineering of IT services, making sure that certain IT resources and infrastructure can be utilized or are made available should the need arises. Capacity Management activities include the following :

> Monitoring and modeling to simulate the performance of IT services and its supporting components, and understand future resource needs
>Analyzing, tuning, and implementing necessary activities to make efficient uses and changes in resource utilization
> Managing and understanding the current demands for computing resources and deriving forecasts for future requirements, which involves an understanding of business priorities
> Application sizing to ensure required service levels can be met
> Building the yearly infrastructure growth plan with input from other team members
> Storing capacity management data while predicting the IT resources needed to achieve agreed service levels and
> Producing a capacity plan that documents current utilization and forecasted requirements, as well as support costs for new applications or releases.

Sabtu, 14 November 2009