cancel
Showing results for 
Search instead for 
Did you mean: 

cvm node

vucni11
Level 4

hi,

i have on node test6 and test88 vcs and cvm.

normally when i issue hastart on both nodes then vcs and cvm are up.

i started on test88 with gabconfig -c -n1 and i issue hastart.

test88:~ # gabconfig -a
GAB Port Memberships
===============================================================
Port a gen  1c14301 membership ;1
Port h gen  1c14303 membership ;1

why in this case cvm will not start automatically?is this because is a parallel service in main.cf?

test88:~ # cat /etc/VRTSvcs/conf/config/main.cf
include "types.cf"
include "CFSTypes.cf"
include "CVMTypes.cf"

cluster clust (
        HacliUserLevel = COMMANDROOT
        )

system test6 (
        )

system test88 (
        )

group cvm (
        SystemList = { test6 = 0, test88 = 1 }
        AutoFailOver = 0
        Parallel = 1
        AutoStartList = { test6, test88 }
        )

        CFSfsckd vxfsckd (
                ActivationMode @Test6 = { tests = sw }
                ActivationMode @test88 = { tests = sw }
                )

        CVMCluster cvm_clus (
                CVMClustName = clust
                CVMNodeId = { test6 = 0, test88 = 1 }
                CVMTransport = gab
                CVMTimeout = 200
                )

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

        ProcessOnOnly vxattachd (
                Critical = 0
                PathName = "/bin/sh"
                Arguments = "- /usr/lib/vxvm/bin/vxattachd root"
                RestartLimit = 3
                )

        cvm_clus requires cvm_vxconfigd
        vxfsckd requires cvm_clus

 

i try to start cvm

test88:~ # vxclustadm -m vcs -t gab startnode
VxVM vxclustadm ERROR V-5-1-11076 Fencing driver not present. Not starting.

why should i start by hand vxfen?

test88:~ # service vxfen start
Starting vxfen..
Loaded 3.0.76-0.11.1-default on kernel 3.0.76-0.11-default
WARNING:  No modules found for 3.0.76-0.11-default, using compatible modules for 3.0.76-0.11.1-default.
Starting vxfen.. Done
Please see the log file /var/VRTSvcs/log/vxfen/vxfen.log
test88:~ # vxclustadm -m vcs -t gab startnode
VxVM vxclustadm INFO V-5-2-9687 vxclustadm: Fencing driver is in disabled mode
test88:~ # gabconfig -a
GAB Port Memberships
===============================================================
Port a gen  1c14301 membership ;1
Port b gen  1c14305 membership ;1
Port h gen  1c14303 membership ;1
Port m gen  1c14307 membership ;1
Port u gen  1c1430d membership ;1
Port v gen  1c14308 membership ;1
Port w gen  1c1430b membership ;1
Port y gen  1c14309 membership ;1

 

test88:~ # vxclustadm nidmap
Name                             CVM Nid    CM Nid     State
test6                            1          0          Out of Cluster
test88                           0          1          Joined: Master

test88:~ # vxdg list
NAME         STATE           ID
tests        enabled,shared,cds   1491557254.23.te

test88:~ # vxprint
Disk group: test88

TY NAME         ASSOC        KSTATE   LENGTH   PLOFFS   STATE    TUTIL0  PUTIL0

Disk group: tests

TY NAME         ASSOC        KSTATE   LENGTH   PLOFFS   STATE    TUTIL0  PUTIL0
dg tests        tests        -        -        -        -        -       -

dm disk_0       disk_0       -        4120320  -        -        -       -

v  testn        fsgen        ENABLED  204800   -        ACTIVE   -       -
pl testn-01     testn        ENABLED  204800   -        ACTIVE   -       -
sd disk_0-01    testn-01     ENABLED  204800   0        -        -       -

i try to mount testn under /vol5s on test88.

test88:~ # cfsmntadm add tests testn /vol5s all=cluster
  Mount Point is being added...

test88:~ # cfsmount /vol5s
  Error: V-35-39: Cluster Manager is not running on system : test6
  Mounting...
  Error: V-35-50: Could not mount [/dev/vx/dsk/tests/testn] at /vol5s on test88

 

why testn was not able to be mounted under /vol5s?is this because the service is parallel?

group vrts_vea_cfs_int_cfsmount1 (
        SystemList = { test6 = 0, test88 = 1 }
        AutoFailOver = 0
        Parallel = 1
        AutoStartList = { test6, test88 }
        )

        CFSMount cfsmount1 (
                Critical = 0
                MountPoint = "/vol5s"
                BlockDevice = "/dev/vx/dsk/tests/testn"
                MountOpt @Test6 = "cluster"
                MountOpt @test88 = "cluster"
                NodeList = { test6, test88 }
                )

        CVMVolDg cvmvoldg1 (
                Critical = 0
                CVMDiskGroup = tests
                CVMVolume = { testn }
                CVMActivation @Test6 = sw
                CVMActivation @test88 = sw
                )

        requires group cvm online local firm
        cfsmount1 requires cvmvoldg1

 

tnx

 

1 ACCEPTED SOLUTION

Accepted Solutions

RiaanBadenhorst
Moderator
Moderator
Partner    VIP    Accredited Certified

Fencing is not configured in the cluster, that is why its saying its starting in disabled mode.

Also, why is your other node down?

Get everything running with a normal setup before you start playing around with more complex scenarios.

run /opt/VRTS/install/installer -fencing to reconfigure fencing.

View solution in original post

2 REPLIES 2

RiaanBadenhorst
Moderator
Moderator
Partner    VIP    Accredited Certified

Fencing is not configured in the cluster, that is why its saying its starting in disabled mode.

Also, why is your other node down?

Get everything running with a normal setup before you start playing around with more complex scenarios.

run /opt/VRTS/install/installer -fencing to reconfigure fencing.

View solution in original post

hi,

tnx