Forum Discussion

Home_224's avatar
Home_224
Level 6
6 years ago

DB service group state to show partial

root@devuaedbs31 # hastatus -sum

-- SYSTEM STATE
-- System State Frozen

A devuaedbs31 RUNNING 0
A devuaedbs32 RUNNING 0

-- GROUP STATE
-- Group System Probed AutoDisabled State

B MNICB_iUATgroup devuaedbs31 Y N ONLINE
B MNICB_iUATgroup devuaedbs32 Y N ONLINE
B cvm devuaedbs31 Y N ONLINE
B cvm devuaedbs32 Y N ONLINE
B dbs_rac_bkup devuaedbs31 Y N ONLINE
B dbs_rac_bkup devuaedbs32 Y N OFFLINE
B oraclerac_db1 devuaedbs31 Y N ONLINE
B oraclerac_db1 devuaedbs32 Y N ONLINE
B oraclerac_db2 devuaedbs31 Y N ONLINE
B oraclerac_db2 devuaedbs32 Y N ONLINE
B oraclerac_db2s devuaedbs31 Y N PARTIAL
B oraclerac_db2s devuaedbs32 Y N PARTIAL
B oraclerac_db3 devuaedbs31 Y N ONLINE
B oraclerac_db3 devuaedbs32 Y N ONLINE
B oraclerac_db4 devuaedbs31 Y N PARTIAL
B oraclerac_db4 devuaedbs32 Y N PARTIAL
B oraclerac_db5 devuaedbs31 Y N ONLINE
B oraclerac_db5 devuaedbs32 Y N ONLINE
B oraclerac_db6 devuaedbs31 Y N OFFLINE

B oraclerac_db6 devuaedbs32 Y N OFFLINE

 


group oraclerac_db2s (
SystemList = { devuaedbs31 = 0, devuaedbs32 = 1 }
Parallel = 1
AutoStartList = { devuaedbs31, devuaedbs32 }
)

CFSMount Ora_Mnt_db2s_a01 (
MountPoint = "/db2s/a01"
BlockDevice = "/dev/vx/dsk/db2s_datadg/db2s_a01_vol"
MountOpt = rw
)

CFSMount Ora_Mnt_db2s_d01 (
MountPoint = "/db2s/d01"
BlockDevice = "/dev/vx/dsk/db2s_datadg/db2s_d01_vol"
MountOpt = rw
)

CFSMount Ora_Mnt_db2s_d02 (
Enabled = 0
MountPoint = "/db2s/d02"
BlockDevice = "/dev/vx/dsk/db2s_datadg/db2s_d02_vol"
MountOpt = rw
)

CVMVolDg diskgroup_db2s (
CVMDiskGroup = db2s_datadg
CVMVolume = { db2s_a01_vol, db2s_d01_vol, db2s_d02_vol }
CVMActivation = sw
)

Ora_Mnt_db2s_a01 requires diskgroup_db2s
Ora_Mnt_db2s_d01 requires diskgroup_db2s
Ora_Mnt_db2s_d02 requires diskgroup_db2s

 

Can you advice if I online the service group base on the server status.

  • frankgfan's avatar
    frankgfan
    6 years ago

    Was there any changes made to the cluster configuration since the last time the service group was onlined?  If there was no changes made to main.cf, thr mainm.cf should be OK.

     

    If some changes were made to nthe cluster configuration recently which was likely caused the issue, check the resource dependencies. 

    As I pointed out in my previous email, for the cvery least, you should be able to dsgnose the issue by onlining the resources per their dependencies, one by one therefore to detemine the one casued the issue.

     

    cfs related issue is most oftern storage rissue or mount point issue related

  • frankgfan's avatar
    frankgfan
    Moderator
    6 years ago

    A service group is shown up in partial meaning there is at least one critical resource in the service group is not in online state.

     

    Here is what you can do:P

    1. check the engine_A.log to find out which is/are the resource(s) in the service group which is/are not online (online timeout)

    or

    2. run hares -display | grep <system_name>   where the system name is the name of the node on which the service group is in partial state

    3. once you have determined which resoruce(s) is/are not online, take the action neccessary to fix the issue.  Once way to fix the resource not online issue is to manually bring it online outside VCS.  For instance, if its an Oracle DB, asked dba to online it, if its an IP, ask network admin to check and fix the IP issue.

    • Home_224's avatar
      Home_224
      Level 6
      6 years ago

      Thank you!

      I still not able to online the resource. From the main.cf, if there is any problem on the cfs mount configuration entry?

      I am not sure if there is one enabled entry in the main.cf. I am not sure if there is impact to make this happen 

      • frankgfan's avatar
        frankgfan
        Moderator
        6 years ago

        Was there any changes made to the cluster configuration since the last time the service group was onlined?  If there was no changes made to main.cf, thr mainm.cf should be OK.

         

        If some changes were made to nthe cluster configuration recently which was likely caused the issue, check the resource dependencies. 

        As I pointed out in my previous email, for the cvery least, you should be able to dsgnose the issue by onlining the resources per their dependencies, one by one therefore to detemine the one casued the issue.

         

        cfs related issue is most oftern storage rissue or mount point issue related