Basic question about probing resources

Hello,

Probing of resource means to run the monitor script for the resource for VCS to know the status of the resource.

By design, all agents in VCS have some entry points,

Agent functions

Agents carry out specific functions on resources. The functions an agent
performs are called entry points. For details on any of the following agent
functions, see the Veritas Cluster Server Agent Developer’s Guide.
■ Online—Brings a specific resource ONLINE from an OFFLINE state.
■ Offline—Takes a resource from an ONLINE state to an OFFLINE state.
■ Monitor—Tests the status of a resource to determine if the resource is
online or offline. The function runs at the following times:
■ During initial node startup, to probe and determine status of all
resources on the system.
■ After every online and offline operation.
■ Periodically, to verify that the resource remains in its correct state.
Under normal circumstances, the monitor entry point is run every 60
seconds when a resource is online. The entry point is run every 300
seconds when a resource is expected to be offline.
■ Clean—Cleans up after a resource fails to come online, fails to go offline, or
fails while in an ONLINE state. The clean entry point is designed to clean up
after an application. The function ensures that the host system is returned
to a valid state. For example, the clean function may remove shared
memory segments or IPC resources that are left behind by a database.
■ Action—Performs actions that can be completed in a short time and which
are outside the scope of traditional activities such as online and offline.
Some agents have predefined action scripts that you can run by invoking
the action function.
■ Info—Retrieves specific information for an online resource.

In VCS all the monitor scripts (for bundled or custom agents) are encoded to have two exit codes ..  110 for successful & 100 for unsuccessful.

If monitor script returns a 110 status code , VCS treats that resource is probed & successfully monitoring.

G

Yes as Gaurav says - "Probing of resource means to run the monitor script for the resource for VCS to know the status of the resource".

All resources need to be probed and the probe by default runs every minute for every ONLINE resource and every 5 mins for OFFLINE resources (so VCS can check resources are not brought online outside of its control).

Mike

Putting it in a simple manner, it has two main meanings:

1. When first configuring or editing a resource configuration, probing is the process that VCS uses to verify if the informations/configurations provided are right and VCS is able to check resource status either online or offline.
2. After first configuration, probe is part of regular resource status monitoring and will fail if VCS is not able to determine resource status, either online or offline. If probe fails, it likely means that resource configuration is wrong. Something might have changed in the APP/OS and VCS configuration need to be updated to be able to monitor that.

Raphael

Thank you very much all of you. Here is my problem

When I try to bring mount resource I go the following error.

Resouce hasn't probed on system 2, 

2013/12/03 03:35:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:36:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:37:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:38:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:39:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:40:11 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:41:10 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:42:10 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.
2013/12/03 03:43:10 VCS ERROR V-16-10001-5565 Mount:apache_mnt:monitor:No FSType is specified.

bash-3.2# hastatus  -sum

-- SYSTEM STATE
-- System               State                Frozen

A  node1                RUNNING              0
A  node2                RUNNING              0

-- GROUP STATE
-- Group           System               Probed     AutoDisabled    State

B  ClusterService  node1                Y          N               OFFLINE
B  ClusterService  node2                Y          N               ONLINE
B  cvm             node1                Y          N               ONLINE
B  cvm             node2                Y          N               ONLINE
B  disk_group      node1                N          N               OFFLINE
B  disk_group      node2                N          N               PARTIAL
B  oracle          node1                Y          N               OFFLINE
B  oracle          node2                Y          N               OFFLINE

-- RESOURCES NOT PROBED
-- Group           Type                 Resource             System

E  disk_group      Mount                apache_mnt           node1
E  disk_group      Mount                apache_mnt           node2
E  oracle          IP                   oracleIP             node1
E  oracle          IP                   oracleIP             node2
 

I did try to probe and got the same error. What am I missing?

bash-3.2# more main.cf
include "types.cf"
include "CFSTypes.cf"
include "CVMTypes.cf"

cluster n (
        UserNames = { admin = cJGeGDfDGkJQiIJkJOiEJoJEiM,
                 ananthan = cJJdJQiQJdJEjDKcJEfEGeGG }
        ClusterAddress = "192.168.0.7"
        Administrators = { admin, ananthan }
        HacliUserLevel = COMMANDROOT
        )

system node1 (
        )

system node2 (
        )

group ClusterService (
        SystemList = { node1 = 0, node2 = 1 }
        AutoStartList = { node1, node2 }
        OnlineRetryLimit = 3
        OnlineRetryInterval = 120
        )

        IP webip (
                Device = e1000g0
                Address = "192.168.0.7"
                NetMask = "255.255.255.0"
                )

        NIC csgnic (
                Device = e1000g0
                )

        webip requires csgnic


        // resource dependency tree
        //
        //      group ClusterService
        //      {
        //      IP webip
        //          {
        //          NIC csgnic
        //          }
        //      }


group cvm (
        SystemList = { node1 = 0, node2 = 1 }
        AutoFailOver = 0
        Parallel = 1
        AutoStartList = { node1, node2 }
        )

        CFSfsckd vxfsckd (
                )

        CVMCluster cvm_clus (
                CVMClustName = n
                CVMNodeId = { node1 = 0, node2 = 1 }
                CVMTransport = gab
                CVMTimeout = 200
                )

        CVMVxconfigd cvm_vxconfigd (
                Critical = 0
                CVMVxconfigdArgs = { syslog }
                )

        cvm_clus requires cvm_vxconfigd
        vxfsckd requires cvm_clus


        // resource dependency tree
        //
        //      group cvm
        //      {
        //      CFSfsckd vxfsckd
        //          {
        //          CVMCluster cvm_clus
        //              {
        //              CVMVxconfigd cvm_vxconfigd
        //              }
        //          }
        //      }


group disk_group (
        SystemList = { node2 = 0, node1 = 1 }
        AutoStartList = { node2 }
        )

        DiskGroup my_dg_group (
                DiskGroup = appdg
                )

        Mount apache_mnt (
                Critical = 0
                MountPoint = "/opt/apps/apache"
                BlockDevice = "/dev/vx/dsk/appdg/vol1"
                FsckOpt = "[-y]"
                )

        Volume appvol (
                Critical = 0
                Volume = vol1
                DiskGroup = appdg
                )

        apache_mnt requires appvol
        appvol requires my_dg_group


        // resource dependency tree
        //
        //      group disk_group
        //      {
        //      Mount apache_mnt
        //          {
        //          Volume appvol
        //              {
        //              DiskGroup my_dg_group
        //              }
        //          }
        //      }


group oracle (
        SystemList = { node1 = 0, node2 = 1 }
        )

        IP oracleIP (
                Enabled = 0
                )



        // resource dependency tree
        //
        //      group oracle
        //      {
        //      IP oracleIP
        //      }
 

Hi,

As Gaurav explained one option is missing.

run below commands to set correctly:

#haconf -makerw

#hares -modify apache_mnt FSType vxfs (or any other file system you are using for this mount point)

#haconf -dump -makero

Then resource should get probed and VCS be able to online the resource.

regards,
Dan

Note: the FsckOpt attribute also does not appear to be set correctly - it should not contain brackets - so in addition to setting the required FSType attribute, you should also set FsckOpt to a valid value.

# hares -modify apache_mnt FsckOpt %-y

(% required to escape the '-')

For more details, refer to the bundled agents reference guide

eg: VCS 5.1SP1 (Solaris) Bundled Agents Reference Guide

http://sort-cdn.symantec.com/public/documents/sfha/5.1sp1/solaris/productguides/pdf/vcs_bundled_agents_51sp1_sol.pdf

(if using a different version/platform, go to https://sort.symantec.com/documents, and refer to the product guides for the relevant version/platform combination)

It works, thank you very much all of you.