reinstalling calendar server - node id - Calendar(Archived)

Hi,
When i try reinstalling calendar server if i give the same node id as previous installation then i get a msg saying the node id already exist, so i added a new node id using uniaddnode utility but i am not able to access the users under the previous node id. when i use the utility uniuser -ls -n <node id> it says invalid node id, any solution for this, am i wrong anywhere in the installation process.
regards,
shantha 

Here, you have cused some irreversible side-effects...
If you delete reinstall the calednar server, you will be prompted to create a node.
If you did not clean up the old installation, you will still have teh existing node.
No two nodes can share the same number.
I would log a support ticket and ask for the steps to follow.
Jean-Marc Robillard 

As far as I know, some of the Calendar Server Information is saved in the OID.
So, simply deleting the calendar server software does not help. You have to remove the OID information too.
Regards,
Silvio

Related

user equivalence test failed (RAC setup in Windows)

I am trying to set up user equivalence between 2 nodes, following instructions from "Oracle® Database Oracle Clusterware and Oracle Real Application Clusters Installation Guide 10g Release 2 (10.2) for Windows"
The Administrator account on both nodes have the same password. But the test failed. I can't find any trouble shooting doc from metalink.
Anyone can help?
Thank you very much.
Laura 
As a minimum you should be able to log on to each node and execute a remote operation on every other node.
You have to also be able to copy a file from every node to every other node.
eg. ssh node_x date
and
scp testfile node_x:/dir/
The first time you execute these commands you are usually asked to confirm the trust relationship.
It is important you do this. Otherwise scripts will start breaking because they are waiting for input from the user. 
Thank you very much. 
well,
i have the same problem with you and haven't get the answer.
U might have done this.Would you email the solution to me at hdj342502#hotmail.com?
thank you very much
regurds
dyluen 
Hi, Try to create a domain account and this domain account should be explicitly set in the local admin group for each node.
Bye
also check Note:388730.1 that can help on windows.
Message was edited by:
Romain VILCOQ
Note:412118.1 is valuable also
Message was edited by:
Romain VILCOQ

ERROR: Wrong password for user

Oracle 10g version 10.2.0.1. RHEL 4. I was trying to setup a backup, but everytime I enter my User ID and Password it keeps telling me it's wrong.
I have tried this with EVERY user ID I have, I have verified the Password by loging in via SSH, and still, it gives me that same error again and again.
I'm sure I must have messed something up during the installation now, the question is, is there anyway to fix this?
Thanks in advance! 
Is a proxy server between you and the internet (to access metalink for patching)?
If yes you should add your hostname to the proxy exception list (dontProxyFor). In 10gR2 this can be done through GridControl or DBConsole (proxy settings) , but also manually:
1. Open the file <GRID_HOME>/sysman/config/emoms.properties
2. Modify the property dontProxyFor with adding the hostname
dontProxyFor=<hostname>
=> replace <hostname> by the actual hostname.
3. Restart the oms or dbconsole.
Another complety different reason for this error could be accounts SYSMAN and/or DBSNMP are locked. These must be unlocked.
Werner 
oradba,
Thank for the response. No I do not have a proxy server between me and the internet. SYSMAN and DBSNMP are NOT locked at this time. Be that as it may, I did update the emoms.properties file as listed, but the results were the same.
BTW: When you say GRID_HOME, you mean the directory for the database instance itself, and not ORACLE_HOME, correct? Yes I'm a bit of a n00b.
Thanks again! 
It depends. If we're talking about centralized GridControl, GRID_HOME is the ORACLE_HOME for the GridControl software, this ORACLE_HOME is not a database home. Standalone DBConsole, which is dedicated to a specific 10g database, has its configuration files under <Database_ORACLE_HOME>/<Hostname_SID>/sysman/config.
I'm wondering could it be a ssh connection issue, I don't know how Oracle makes actually the connection to the host.
Werner 
The latest advice I was given was to make the oinstall group a member of the dba group. Any thoughts? 
I don't understand what you mean with 'make one group member of another group'.
It's recommend to define oinstall as primary and dba as secondary group of the oracle software owner, that's the id which installed the software.
Werner 
oradba,
Never mind, I'm new to Linux and I was given incorrect information. I've also been told that the ID I did the installation with needs to change due to a mistake. (The ID exists already in LDAP)
Let me sort that out and I'll get back to you if the problem still exists.
Thanks! 
Your hint with LDAP is interesting, because there's a bug reported (3507516), when LDAP authentication is used. The workaround is:
1. Create a local password for the OS user
2. Disable the use of remote authentication of OS users
However, you shouldn't be affected, because the bug note states the issue is fixed in 10.2.
Werner 
Would that be 10.2.0.2? Or 10.2.0.1?
Cause I'm using 10.2.0.1. 
It is said 10.2, that means for me starting with 10.2.0.1 .
Werner 
But first the problem occurred on SUN Solaris and - due to my experience - Oracle sometimes 'forgets' to fix such a problem on other platforms, too.
Werner 
Lovely, I'll have to keep my eyes open then.
Thanks! 
Ok, I've upgraded to 10.2.0.2, and I'm still having the same issue.
ANY help would be appreciated. THANKS! 
Could you paste the out put of command <AOH>\bin\ls -l nm* 
Forgive my n00bness. I understand you mean the Agent directory, but I'm not sure where that is.
The default installation directory for oracle is /u01/app/oracle/product/10.2.0/db_1, is that what you mean?
If it is, this is the result:
-bash: /u01/app/oracle/product/10.2.0/db_1/bin/ls: No such file or directory
Scratch the above. Here's the results:
nmb nmcbufp nmccollector0 nmei0 nmo0 nmocat0 nmupm nmus nmb0 nmccollector nmei nmo nmocat nmuct nmupm0 nmus0
Message was edited by:
CooperHawkes Because I realized the mistake.

Oracle clustering server error with redhat 10

This is the problem
1- We have installed linux redhat 10 enterprise edition on our servers
2- 2 node server, type HP
3- we have done the all the necessary configuration like mentioned in the oracle documentation and we have set the following files:
a- csh.login
b- hosts
c- limits.conf
d- login
e- profile
f- rc.local
g- sysctl.conf
after all of this setting and updating the kernel version and setting up the boot loader order
and did the rsa/dsa keys to the two servers, but they are not sync or let us say they sync and when we run the check on that sync we get this massage
Man In The Middle Attack " MITMA'
and after some time it come back to sync then it goes back to the same massage
Man In The Middle Attack " MITMA'
whats wrong and why ?!!!!!
because of this error we cant make the installation to the oracle clustering server
Please advice us on this problem........
Thank you. 
Move your ~.ssh/know_hosts file on both servers to another location and try it again... 
I think you mean ~/.ssh/known_hosts. Remember to make attempts to connect (via SSH) to the remote node via the public and private networks. When you do that, you'll be prompted to confirm that the remote node is okay to connect to. When you answer "yes", the known_hosts file will be created/updated. 
Yep forgot that /....
No wonder why my files keep disappearing when I do a rm -f *
;-)

Installing Oracle 10g Grid Agent on Windows Workgroup

Hi All,
I am now in the process of setting up a Grid-based environment in my company. I have managed to get most of the servers, on the same domain on the Grid, with quite a bit of help from the forum...:)))
There are a couple of legacy systems, which are not on the domain like the other servers, including my Grid server. The databases on these systems are working fine.
I am now attempting to bring these legacy servers into the Grid as well. I was not sure if it were possible, since the servers on which the databases exist were configured as Workgroups.
So, I created a test server as a WORKGROUP and created a test database there successfully. But when I tried to install the agent, it went till the second screen, where we key in the physical location for the agent to be installed,when I click next it just exited the OUI, without any error messages.
I checked the install logs in c:\program files\oracle\bin\log and found only a 0 KB file of the install.
Could one of you please tell me what is going wrong here? Or is it not possible to install a Grid Agent on a Workgroup server?
Thanking you all in advance,
Regards,
Sriraman 
It shouldn’t matter.. As long as you have admin rights (or you log on as a admin) to that server, it shouldn’t have an issue. I suggest you read this
http://download.oracle.com/docs/cd/B16240_01/doc/install.102/e10953/installing_em.htm#CJGGHJEF
If you have any further question please let me know. 
Hi mate....
Thanks for your reply.
I went through the document you suggested, but I am afraid it does not help me. Perhaps I have not stated my problem clearly....I am giving a step by step account of what I did below:
1. Removed DB server from domain and made it WORKGROUP.
2. Created local OS account and made it part of the "admin!" group, so I could have administrator rights.
3. Restarted the server.
4. Started the OUI by clicking on set up icon.
5. Chose the option "Additional Management Agent" and clicked "Next".
6. Defined location where to install the agent "D:\OraGAgent" and clicked "Next".
At this point, the OUI exits without any error message. When I checked the installation log in C:\Program Files\Oracle\bin\log, there is only one file containing a log of the above steps and no error message there.
Could you please explain what mistake I have made for this problem to occur....and how I can correct it? Only then will I be able to bring the remaining servers as well into the Grid......
Thanks and Regards,
Sriraman

Oracle Install Problem : multiple UID 0

hi All. I am going to install 12c on my Linux , however there are two account with UID 0which throw an error like below during installation:INFO: ERROR: [Result.addErrorDescription:624]  PRVF-4132 : Multiple users "root,abcd" with UID "0" exist on nodes "server1".INFO: FINE: [Task.perform:594]TaskUsersWithSameID:Users With Same UID[TASKUSERSWITHSAMEID]:TASK_SUMMARY:FAILED:CRITICAL:VERIFICATION_FAILED          ERRORMSG(server1): PRVF-4132 : Multiple users "root,abcd" with UID "0" exist on nodes "server1".INFO: *************** End of CVU Error logs ***************  are there anyway to ignore this check/error, I would like to keep the user abcd
You've messed up your OS installation.  Your problem is with the OS, not the database. https://community.oracle.com/community/server_%26_storage_systems/linux/oracle_linux/overview
So what's the solution ?
Edit /etc/passwd to change abcd's UID, and then chown all his files to abcd.
>hi All. I am going to install 12c on my Linux , Consider to actually Read & Follow the Fine Manual below https://docs.oracle.com/database/121/LADBI/toc.htm Alternatiely install VirtualBox & then VM which already contains a working Oracle DB
I manually edit /etc/passwd and process the install again. and its works thanks!
So mark this thread as answered.
2670775 wrote:
 
I manually edit /etc/passwd and process the install again. and its works thanks!
So now, do you understand why that worked and why it was necessary?  Do you know HOW you managed to get the same UID assigned to 2 users ... especially UID 0 which should always be root? You could take the fix and move on.  Or you could take this as a learning opportunity.

Categories

Resources