Go to main content

Managing Network Datalinks in Oracle® Solaris 11.4

Exit Print View

Updated: November 2020
 
 

Configuring Probe-Based Failure Detection for DLMP Aggregation

In DLMP aggregations, probing is disabled by default and only link-based failure detection is used. To enable probing, you must configure the probe-ip property.

In addition to probe-ip, probe-vlan-id and probe-fdt are also configurable properties related to probe-based failure detection.

How to Configure Probe-Based Failure Detection

This procedure applies only to DLMP aggregations and assumes that one already exists.

Before You Begin

Ensure that your role has the appropriate rights profile to perform this procedure. See Using Rights Profiles to Perform Network Configuration.

  1. (Optional) Display all the existing aggregations to identify the aggregation for configuring probe-based failure detection.
    $ dladm show-aggr
  2. Configure probing for the aggregation by either using automatically selected source and target IP address or by manual selection.
    • To use automatic selection, issue the following command:

      If you do not specify target addresses, the DLMP aggregation chooses a target IP automatically from one of the next hop routers that are on the same subnet as the source IP address.

      $ dladm set-linkprop -p probe-ip=+ aggr

      With this syntax, you do not specify source or target IP addresses. The process automatically selects both source address and target addresses for the probing. The source address is selected from configured addresses in the global zone. A target address is selected from one of the next hop routers that are on the same subnet as the source address.


      Note -  Probing using automatic selection is a best practice method of setting up probe-based failure detection.
    • To use manual selection, issue the following command:
      $ dladm set-linkprop -p probe-ip=[source[,...]]+[target[,...]] aggr
      source

      Specifies the source IP address for probing. Only configured addresses in the global zone can be source addresses. You can format the source IP address in any one of the following ways:

      • probe-ip=IP-address[/prefix-length]+

        IP address and its prefix length. For example, 203.0.113.1/24+.

      • probe-ip=addr-obj-name+

        Address object name. For example, vnic1/addr1+.

      • probe-ip=interface-name+

        IP Interface name, which can be the aggregated interface itself or any VNIC over the aggregation. For example, aggr1+. The IP interface must have a configured IP address.

      • probe-ip=hostname+

        Host name. For example, sdg1+

      target

      Specifies the target IP address for probing. When you specify the target IP address, it must be in any one of the following ways:

      • probe-ip=+IP-address

        For example, +203.0.113.1.

      • probe-ip=+hostname

        For example, +sdg1

      aggr

      Name of the aggregation.


      Note -  Specifying the source and the target IP addresses is optional. If you do not specify the source and the IP target addresses, the source and the probing IP addresses are automatically selected.
  3. (Optional) Set the VLAN ID for the probes.
    $ dladm set-linkprop -p probe-vlan-id=probe-VID aggr

    probe-VID is the VLAN ID to be used for both ICMP and transitive probes. The valid values of probe-VID ranges from 0 to 4094. The probe-VID value 0 indicates that the probes are untagged. The default value is 0.

  4. (Optional) Set the failure detection time.
    $ dladm set-linkprop -p probe-fdt=fdt aggr

    fdt is the failure detection time specified in seconds. The default value is 10 seconds.

  5. (Optional) Display the probe statistics.
    $ dlstat show-aggr -n -P [[t],[i],[all]]

    The –P specifies the type of the probe to display. You can specify any of the following in a comma-separated list as the probe type:

    • t – Displays the transitive probes.

    • i – Displays the ICMP probes.

    • all – Displays both the transitive and ICMP probes.

Example 7  Configuring Probe-Based Failure Detection

This example shows how you can configure probe-based failure detection by using the default automatic selection of source and probing IP addresses.

$ dladm create-aggr -m dlmp -l net0 -l net1 -l net2 aggr1
$ dladm show-aggr -x
LINK       PORT           SPEED DUPLEX   STATE     ADDRESS            PORTSTATE
aggr0      --             1000Mb full    up        2:8:20:c2:24:45    --
           net0           1000Mb full    up        8:0:27:49:10:b8    attached
           net1           1000Mb full    up        8:0:27:e4:d9:46    attached
           net2           1000Mb full    up        8:0:27:38:7a:97    attached

$ ipadm create-ip aggr1
$ ipadm create-addr -a 192.0.2.137/24 aggr1

$ dladm set-linkprop -p probe-ip=+ aggr1
$ dladm set-linkprop -p probe-vlan-id=2100 aggr1
$ dladm set-linkprop -p probe-fdt=15 aggr1

$ dladm show-linkprop -p probe-ip,probe-vlan-id,probe-fdt aggr1
LINK     PROPERTY         PERM   VALUE      EFFECTIVE       DEFAULT      POSSIBLE
aggr1    probe-ip         rw     +           +               -            -- 
aggr1    probe-vlan-id    rw     2100        2100             0           0-4094
aggr1    probe-fdt        rw     15          15              10           1-600

$ dlstat show-aggr -n -P t,i aggr1
TIME   AGGR   PORT  LOCAL           TARGET          PROBE  NETRTT     RTT
0.45s  aggr1  net0  net0            net1            t16148  --         --
0.45s  aggr1  net0  net0            net1            t16148  0.63ms     0.81ms
1.08s  aggr1  net1  net1            net0            t16148  --         --
1.08s  aggr1  net1  net1            net0            t16148  0.72ms     0.99ms
2.07s  aggr1  net1  192.0.2.137     192.0.2.137     i15535  --         --
2.07s  aggr1  net1  192.0.2.137     192.0.2.137     i15535  0.18ms     0.54ms