Forum Discussion

allaboutunix's avatar
11 years ago

Regarding resource grp faults

Hi,
 
 
Clearing the faulty resource with this way,
Clearing faulted resources in a service group
 
Clear a resource to remove a fault and make the resource available to go online.
 
To clear faulted, non-persistent resources in a service group
 
 Type the following command:
hagrp -clear service_group [-sys system]
 
and
 
Clearing the faulty resource in this way,
 
server14# hares -display | grep FAULT
DBAppOramon State server14 FAULTED
 
OK, Now you got the resource name. In order to online it back you must clear the resource first.
Use the resource name that you grep just now.
 
server14# hares -clear DBAppOramon -sys server14
 
Check back whether it is working fine.
 
server14# hares -display | grep FAULT
 
Now, you can online it back. But before that you must make sure where it was supposed to be online.
You can check it back at main.cf
 
server14# hares -online DBAppOramon -sys server14
 
Now you can check whether the reource will go online. It will take some time to online back.
 
server14# hares -display | grep DBAppOramon | grep ONLINE
DBAppOramon State server14 ONLINE
 
=====================================================================
 
Is there any difference between these two ways, both of these ways are used to clear the faults, right,Correct me if I am wrong please.
 
  • You can use either method to clear the fault, but the difference between the two methods is that the "hagrp -clear" command will clear ALL faulted, non-persistent resources in the specified group.

    Whereas, the "hares -clear" command will only clear the faulted flag on the ONE resource. 

    Also, you probably should not clear a faulted resource until you have satisfied yourself as to why it has faulted in the first place -- else you may end up having to clear it again, and again... (and again...)

    Review the engine log and the resource's type log, as well as any other log that the resource may write to in order to gain an understanding as to why the resource faulted. 

    And of course you may not be able to attempt another online of the resource if the service group it is in has faulted on that server, which would be the case if the "DBAppOramon" resource is marked at Critical (the Critical attribute is set to 1).

     

    Have you managed to arrange VCS training yet?  It do recommend it. 

     

  • You can use either method to clear the fault, but the difference between the two methods is that the "hagrp -clear" command will clear ALL faulted, non-persistent resources in the specified group.

    Whereas, the "hares -clear" command will only clear the faulted flag on the ONE resource. 

    Also, you probably should not clear a faulted resource until you have satisfied yourself as to why it has faulted in the first place -- else you may end up having to clear it again, and again... (and again...)

    Review the engine log and the resource's type log, as well as any other log that the resource may write to in order to gain an understanding as to why the resource faulted. 

    And of course you may not be able to attempt another online of the resource if the service group it is in has faulted on that server, which would be the case if the "DBAppOramon" resource is marked at Critical (the Critical attribute is set to 1).

     

    Have you managed to arrange VCS training yet?  It do recommend it. 

     

  • Hi

    I will have to do, please asist how to do but I can be able to do online free classes or do you have video based training?

     

    Thanks,

    Allaboutunix

  • I personally do not have any video training available - sorry. 

    I believe that I heard that much of the SFHA taining is now available online for the general public -- but do not quote me on that -- I think I saw some posts referring to the same.