FAQ
Hello

I am getting the *User equivalence unavailable on all the nodes* while
running cluvfy verification. (10gR2 RAC)

ssh on both nodes are working correctly without needing password. I also
enabled ssh user equivalency on the node that will launch OUI with exec
/usr/bin/ssh-agent $SHELL and /usr/bin/ssh-add. I can't find any additional
troublesheeting message from metalink and anywhere else and never get this
message before.

Appreciate all the help. Many thanks.

$ ./runcluvfy.sh stage -pre crsinst -n node1,node2 -verbose
*notice: JAVA_HOME not set in the environment. <---- is this another
issue???*

Performing pre-checks for cluster services setup

Checking node reachability...

Check: Node reachability from node "node1"

Destination Node Reachable?
------------------------------------ ------------------------
node1 yes
node2 yes

Result: Node reachability check passed from node "node1".

Checking user equivalence...

Check: User equivalence for user "oracle"

Node Name Comment
------------------------------------ ------------------------
node1 failed
node2 failed

*Result: User equivalence check failed for user "oracle".*
**
*ERROR:

User equivalence unavailable on all the nodes.
Verification cannot proceed.
*
**
**
*Thanks for all the help!!*
**
*Cindy
*

Search Discussions

  • Niall Litchfield at Sep 6, 2006 at 5:38 pm
    Have you done this

    Set up user equivalence. (If on Unix, set up remote copy (rcp) to all

    the nodes, including the local node. If on Windows, issue "NET
    USE \\node_name\C$" for each node.)from note 206037.1? That is it

    is correct setting up of rcp that you should look at rather than ssh
    per-se
    On 9/6/06, oracle sos wrote:


    Hello

    I am getting the User equivalence unavailable on all the nodes while running cluvfy verification. (10gR2 RAC)

    ssh on both nodes are working correctly without needing password. I also enabled ssh user equivalency on the node that will launch OUI with exec /usr/bin/ssh-agent $SHELL and /usr/bin/ssh-add. I can't find any additional troublesheeting message from metalink and anywhere else and never get this message before.

    Appreciate all the help. Many thanks.


    $ ./runcluvfy.sh stage -pre crsinst -n node1,node2 -verbose
    notice: JAVA_HOME not set in the environment. <---- is this another issue???

    Performing pre-checks for cluster services setup

    Checking node reachability...

    Check: Node reachability from node "node1"
    Destination Node Reachable?
    ------------------------------------ ------------------------
    node1 yes
    node2 yes
    Result: Node reachability check passed from node "node1".


    Checking user equivalence...

    Check: User equivalence for user "oracle"
    Node Name Comment
    ------------------------------------ ------------------------
    node1 failed
    node2 failed
    Result: User equivalence check failed for user "oracle".

    ERROR:
    User equivalence unavailable on all the nodes.
    Verification cannot proceed.



    Thanks for all the help!!

    Cindy
    --
    Niall Litchfield
    Oracle DBA
    http://www.orawin.info
    --
    http://www.freelists.org/webpage/oracle-l
  • Paula Stankus at Sep 6, 2006 at 8:37 pm
    Then I tried to just run standard.


    It said it created the package.


    Then I tried to run utlirp.sql


    It hangs and comes back with messages standard not accessible.


    Now my whole machine is hanging.


    Is there anyway to cleanup the catalog without recoverying the entire database?


    I read that it might be necessary to just have online a large rbs, create standard, run catalog.sql, run catproc.sql and utlirp.sql


    Any advice is greatly appreciated. The good thing is this is test but it means contract $$ if we have to loose a day.


    sigh....



    Get your email and more, right on the new Yahoo.com
  • Bobak, Mark at Sep 6, 2006 at 8:45 pm
    What happens if you do:
    alter package standard compile;
    ??


    Then try utlrp.sql not utlirp.sql (no need to invalidate everything and
    then recompile it all. Just recompile what's not valid.)



    --
    Mark J. Bobak
    Senior Oracle Architect
    ProQuest Information & Learning

    "A human being should be able to change a diaper, plan an invasion,
    butcher a hog, conn a ship, design a building, write a sonnet, balance
    accounts, build a wall, set a bone, comfort the dying, take orders, give
    orders, cooperate, act alone, solve equations, analyze a new problem,
    pitch manure, program a computer, cook a tasty meal, fight efficiently,
    die gallantly. Specialization is for insects." --Robert A. Heinlein

    ________________________________

    From: oracle-l-bounce_at_freelists.org
    On Behalf Of Paula Stankus
    Sent: Wednesday, September 06, 2006 4:37 PM
    To: oracle-l_at_freelists.org
    Subject: Re: URGENT!!! - Inadvertantly ran catalog.sql in an existing
    database

    Then I tried to just run standard.

    It said it created the package.

    Then I tried to run utlirp.sql

    It hangs and comes back with messages standard not accessible.

    Now my whole machine is hanging.

    Is there anyway to cleanup the catalog without recoverying the entire
    database?

    I read that it might be necessary to just have online a large rbs,
    create standard, run catalog.sql, run catproc.sql and utlirp.sql

    Any advice is greatly appreciated. The good thing is this is test but
    it means contract $$ if we have to loose a day.

    sigh....

    ________________________________

    Get your email and more, right on the new Yahoo.com
    <http://us.rd.yahoo.com/evt=42973/*http://www.yahoo.com/preview>

    --
    http://www.freelists.org/webpage/oracle-l
  • Paul Drake at Sep 6, 2006 at 8:46 pm

    On 9/6/06, Paula Stankus wrote:
    Then I tried to just run standard.

    It said it created the package.

    Then I tried to run utlirp.sql

    It hangs and comes back with messages standard not accessible.

    Now my whole machine is hanging.

    Is there anyway to cleanup the catalog without recoverying the entire
    database?

    I read that it might be necessary to just have online a large rbs, create
    standard, run catalog.sql, run catproc.sql and utlirp.sql

    Any advice is greatly appreciated. The good thing is this is test but it
    means contract $$ if we have to loose a day.

    sigh....

    ------------------------------
    Paula,

    What version of the oracle database server software? (8.1.7.0)
    Were you connected as sysdba? (yes)
    Is this RAC? (no)
    Was the database instance started up migrate or upgrade, as in

    SQL> startup migrate

    (no)?
    What was the value of the following:

    sga_target or shared_pool_size

    As its recommended that the shared_pool_size be at least 150M when running
    catalog.sql (or catpatch.sql).

    More than likely, processes (sessions or background jobs) have pins on
    packages that are attempting to be invalidated - hence the "hang".

    Bounce the instance.
    startup migrate and re-run catalog.sql, then recompile the invalid objects.
    Check for invalid objects.
    Then let users back in and worry about implications.

    hth.

    Paul
  • Powell, Mark D at Sep 6, 2006 at 8:54 pm
    Where you user SYS (sysdba) when you ran catalog? If not you should be
    able to drop the wrong user owned objects and then fix all public
    synonyms to point to the correct user, that is, SYS.


    Anyway try this:


    Drop any should be SYS owned objects owned by any other user
    shutdown immediate
    startup restrict
    connect / as sysdba
    @catalog
    @catproc
    @utlrp

    -- take a look
    -- if OK unrestrict the db or shutdown and startup normal
    -- Also note that any direct grants on SYS owned objects to users,

    package owners, etc... will likely need to be re-issued right before you
    run utlrp


    Note - I have seen special instructions for dealing with revalidation of
    STANDARD on metalink that you may want to search for if the above does
    not work. I think the special instructions were due to a bug that
    caused package STANDARD to go invalid.


    HTH -- Mark D Powell --



    From: oracle-l-bounce_at_freelists.org
    On Behalf Of Paula Stankus
    Sent: Wednesday, September 06, 2006 4:37 PM
    To: oracle-l_at_freelists.org
    Subject: Re: URGENT!!! - Inadvertantly ran catalog.sql in an existing
    database

    Then I tried to just run standard.


    It said it created the package.


    Then I tried to run utlirp.sql


    It hangs and comes back with messages standard not accessible.


    Now my whole machine is hanging.


    Is there anyway to cleanup the catalog without recoverying the
    entire database?


    I read that it might be necessary to just have online a large
    rbs, create standard, run catalog.sql, run catproc.sql and utlirp.sql


    Any advice is greatly appreciated. The good thing is this is
    test but it means contract $$ if we have to loose a day.


    sigh....



    Get your email and more, right on the new Yahoo.com
    <http://us.rd.yahoo.com/evt=42973/*http://www.yahoo.com/preview>
  • Paula Stankus at Sep 7, 2006 at 10:00 am
    I would like to thank everyone that responded to this e-mail. I believe that having Standard invalid then trying to run utlrp.sql caused core dumps and eventually led to the entire server being hung maybe as I was in some kind of a loop. Once I was able to get access to the machine and cycle the server I was able to try some things which was a combination of the advice I got from you all. This is what worked:


    1-I brought the database up in restricted mode
    2-I tried to run catalog.sql but it hung
    3-I compiled the Standard package and package body which were there but not validated.
    4-I commented out the @Standard.sql so the Standard package would not get recreated or invalidated
    5-I ran the rest of catalog.sql
    6-I ran catproc.sql
    7-I ran utlrp.sql (to validate just invalid objects).

    It all worked and didn't take long to do once I got access to the server - that was the longest part. :)


    Thanks very much for the help.
    Paula




    Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small Business.
  • K Gopalakrishnan at Sep 6, 2006 at 6:39 pm
    Hi,

    Check the following things..

    Make sure the UIDs and GIDs are same across the nodes.
    Try creating some files on the other nodes (touch, rm)
    If nothing works trace the cluvfy. You can trace this by setting the
    trace variable
    SRVM_TRACE to true.

    -Gopal

    --
    Best Regards,
    K Gopalakrishnan
    Co-Author: Oracle Wait Interface, Oracle Press 2004
    http://www.amazon.com/exec/obidos/tg/detail/-/007222729X/

    Coming Soon... Oracle RAC Handbook, Oracle Press 2006
    http://www.amazon.com/gp/product/007146509X/

    --
    http://www.freelists.org/webpage/oracle-l
  • Niall Litchfield at Sep 7, 2006 at 7:33 am

    On 9/6/06, K Gopalakrishnan wrote:
    3. If nothing works trace the cluvfy. You can trace this by setting the
    trace variable
    SRVM_TRACE to true.
    talking of which has anyone successfully UNSET this variable? In my
    environment RH3, 10.1 setting SRVM_TRACE to true generates the trace,
    but I have to create a new session to stop the tracing. In my
    innocence I'd have hoped that FALSE or NONE would have stopped
    tracing...

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouporacle-l @
categoriesoracle
postedSep 6, '06 at 4:30p
activeSep 7, '06 at 10:00a
posts9
users7
websiteoracle.com

People

Translate

site design / logo © 2022 Grokbase