Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 84 views

    Aci Apic - Learn Work It

    Uploaded by

    ravi kant
    The Cisco Application Policy Infrastructure Controller (APIC) is the main architectural component of the Cisco ACI solution. It is the unified point of automation and management for the Cisco ACI fabric, policy enforcement, and health monitoring. APIC controllers come in different configurations depending on the size of the fabric. A minimum of three APIC controllers are required for fault tolerance, with the ability to elect a new leader and continue operations if one controller fails.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Aci Apic - Learn Work It

    Uploaded by

    ravi kant
    84 views12 pages
    The Cisco Application Policy Infrastructure Controller (APIC) is the main architectural component of the Cisco ACI solution. It is the unified point of automation and management for the Cisco ACI fabric, policy enforcement, and health monitoring. APIC controllers come in different configurations depending on the size of the fabric. A minimum of three APIC controllers are required for fault tolerance, with the ability to elect a new leader and continue operations if one controller fails.

    Original Description:

    7. ACI APIC - LEARN WORK IT

    Original Title

    7. ACI APIC - LEARN WORK IT

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The Cisco Application Policy Infrastructure Controller (APIC) is the main architectural component of the Cisco ACI solution. It is the unified point of automation and management for the Cisco ACI fabric, policy enforcement, and health monitoring. APIC controllers come in different configurations depending on the size of the fabric. A minimum of three APIC controllers are required for fault tolerance, with the ability to elect a new leader and continue operations if one controller fails.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    84 views12 pages

    Aci Apic - Learn Work It

    Uploaded by

    ravi kant
    The Cisco Application Policy Infrastructure Controller (APIC) is the main architectural component of the Cisco ACI solution. It is the unified point of automation and management for the Cisco ACI fabric, policy enforcement, and health monitoring. APIC controllers come in different configurations depending on the size of the fabric. A minimum of three APIC controllers are required for fault tolerance, with the ability to elect a new leader and continue operations if one controller fails.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 12

    LEARN WORK IT

    INFORMATION TECHNOLO GY (NE T WORK )

    C I S C O ACI BLO GS VMWARE N SX BLO G S CISCO ROUT ING B LO G

    C I S CO SW ITCHIN G BLO G IT INS TITU TES CONTACT US

    TERMS & CONDIT ION

    7. ACI APIC
     APRIL 7, 2021  LEAVE A COMMENT

    Cisco Application
    Policy
    Infrastructure
    Controller
    The C
    Ci s c o  Application Policy Infrastructure Controller
    (C
    Ci s c o A P I C ) is the main architectural component of
    the C
    Ci s c o  ACI solution. It is the uni�ed point of
    automation and management for the C
    Ci s c o  ACI fabric,
    policy enforcement, and health monitoring

    Cisco APIC Part


    Description
    configuration numb er

    APIC with
    medium-size
    CPU, hard
    APIC- drive, and
    Medium
    M3/M4 memory
    con�gurations
    (up to 1200
    edge ports)

    APIC with
    large CPU,

    Large APIC-L3 hard drive,


    and memory
    con�gurations
    Cisco APIC Part
    Description
    configuration numb er

    (more than
    1200 edge
    ports)

    Cluster of
    3 APIC-
    SERVER-M3
    with medium-
    APIC-
    Medium size CPU, hard
    CLUSTER-
    cluster drive, and
    M3/M4
    memory
    con�gurations
    (up to 1200
    edge ports)

    Cluster of
    3 APIC-
    SERVER-L3
    with large
    APIC- CPU, hard
    Large cluster CLUSTER- drive, and
    L3 memory
    con�gurations
    (more than
    1200 edge
    ports)

    1 M3*APIC
    with medium-
    size CPU, hard
    APIC-
    drive,
    XS Cluster CLUSTER-
    memory and
    XS
    2 Virtual
    APICs. XS
    Cluster is only
    Cisco APIC Part
    Description
    configuration numb er

    available as
    part of mini
    ACI fabric
    bundle part
    number –
    ACI-C9332-
    VAPIC-B1

    APIC with
    medium-size
    CPU, hard
    Medium drive, and
    APIC-M3=
    (spare) memory
    con�gurations
    (up to 1200
    edge ports)

    APIC with
    large CPU,
    hard drive,
    and memory
    Large (spare) APIC-L3=
    con�gurations
    (more than
    1200 edge
    ports)

    APIC Controller Hardware:

    For APIC controller Cisco uses UCS series


    Servers, Below is the speci�cation:

    UCS C Series Server: C220 M3 ( used as generation 1 )


    UCS C Series Server: C220 M4 ( used as generation 2 )
    ACI DC Sizing:

    up to 80 leaf: use 3 APIC cluster


    up to 300 leaf: use 5 APIC cluster
    up to 400 leaf: use 7 APIC cluster
    Currently, APIC controller’s uses are between (3-7) N+2,
    But the roadmap is to scale up between (3-31).

    APIC cluster -Sharding

    •APIC’s management tree divided into DB units called


    shards to allow load balancing and scaling
    •Shards assigned to appliances by a static shard layout
    •Each DN is mapped to a shard number (0 –31)
    •Data accessed together is put in the same shard
    •All data for an application pro�le maps to one
    particular shard
    •Topology data or switch stats shared by a switch ID
    •Appliance vector (AV) : Vector of APIC with their
    address, state
    •Replica vector (RV) : Vector of shard, replica with state.
    Leader for each shard
    •Fabric node vector (FNV) : Vector of switch nodes with
    address, state
    •Check using acidiag[fnvread|avread|rvread]

    Some useful Controller Commands

    acidiag avread

    acidiag fnvread
    acidiag rvread

    APIC -Cluster Formation

    •APICs connected to leaf switches on infra network


    •IP of APIC learned via LLDP and distributed through
    ISIS
    •On discovering new APIC start sending periodic HB
    •APICs which are heart-beating with each other are said
    to be in a Cluster

    APIC Cluster Information : GUI


    admin@p o d2-apic1:tmp> show c ontroller
    controllers:

    ID NAME INFRASTRUCTURE-IP IN-BAND-


    MANAGEMENT-IP OUT-OF-BAND-MANAGEMENT-
    UP-TIME STATUS

    1 pod2-apic1 10.0.0.1 0.0.0.0 10.48.16.171 26:01:54:31.000 in-


    service
    2 pod2-apic2 10.0.0.2 0.0.0.0 10.48.16.172 26:01:35:44.000
    in-service
    3 pod2-apic3 10.0.0.3 0.0.0.0 10.48.16.173 26:01:17:00.000
    in-service

    Why we need minimum 3 APIC?

    Since leader election is distributed we need the majority


    of replicas to agree to the leader. So minimum number
    of replicas/APIC to continue operation with a single
    failure is 3.
    •1 APIC : obviously no fault tolerance. Data lost for a
    single failure
    •2 APIC : Write unavailability with single failure. Can
    recover if APIC is replaced
    •3 APIC : If 1 APIC lost, the other two can elect a new
    leader and continue writes. If 2 APIC lost we go to
    minority behavior (no writes, only read)

    APIC Fabric Failures Scenarios:


    All APIC in ACI , works in the majority and minority
    function,
    I f o n e A P I C i s d o w n : Two are still in Active and
    APIC cluster is in majority, the read Write function can
    still work properly. You can use HOT standby APIC for
    quick manual recovery.
    I f t w o A P I C i s d o w n : Cluster is in the minority and
    Single left APIC goes in to read mode, ACI fabric will
    work properly but there would be no con�guration
    change.
    I f m o r e t h a n t w o A P I C i s d o w n , you should
    immediately call TAC to marge external con�guration
    from backup and to restore fabric use command:
    trigger fabric re c over y.
    All APIC controller has two interfaces to be connected
    to the fabric, and it is recommended that this two
    interface of APIC should be connected to two separate
    leafs. All APIC must be connected to different leafs.

    APIC Cluster Fault Tolerance

    APIC Images

    Link to download Image

    https://software.cisco.com/download
    /home/285968390/type/286278832/release
    /4.2(7f)

    You might also like