Ora-17629, Ora-17627 Errors When Active Duplication

Today I get the next error from target database when I was perform a Active Duplicate;

released channel: c5
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of Duplicate Db command at 06/18/2019 11:40:29
RMAN-05501: aborting duplication of target database
RMAN-03015: error occurred in stored script Memory Script
RMAN-06136: ORACLE error from auxiliary database: ORA-17629: Cannot connect to the remote database server
ORA-17627: ORA-01017: invalid username/password; logon denied
ORA-17629: Cannot connect to the remote database server

Reading note 2119741.1 in ML/MOS it says the problem is caused by changed the password for user sys. It need bounce the listener. This is my case, but when I restarted the listener the problem continued and gave the same error. After some tests recreate the password file in the auxiliary and solve the problem.

HTH – Antonio NAVARRO

Advertisements

ORA-01017 RMAN-00554 RMAN-04005 Connecting From Rman

Today I have been reported an error that occurs when making a backup of a database, which previously did not occur and has been working correctly until yesterday. It seems that there has been no no change (pacht installation, change of parameterization, etc.) The error sent to me is an ORA-0107 when executing the backup.

Doing some connection test, I can see that If I perform the connection inside the rman utility I get the next error;

 
RMAN> connect target backup/XXXX@cloud;  

RMAN-00571: =========================================================== 
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS =============== 
RMAN-00571: =========================================================== 
ORA-01017: invalid username/password; logon denied 

But I perfom the conection like a parameter at command line

 
pluton> rman  target backup/XXXX@cloud
RMAN-00571: =========================================================== 
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS =============== 
RMAN-00571: =========================================================== 
RMAN-00554: initialization of internal recovery manager package failed 
RMAN-04005: error from target database: 
ORA-01017: invalid username/password; logon denied 

This is a classic Oracle and some of its utilities, it depends how you run commands can give you more or less information. Remember that the correct way is

rman target / debug=ALL log=/diag/anavarro/mydebug.log

Connection by using sqlplus and listener work fine, so it can be issues of permissions or the password file.

sqlplus backup/XXXX@cloud

The first option is Grant sysdba to user don’t solve the problem, although the user already had this privilege before;

grant sysdba to backup

This fails so I have finally deleted the password file and recreated it with the following command;

orapwd file=’+CLOUD_DG’ password=ecc3005 dbuniquename=cloud entries=20 FORMAT=12

This solve the problem

HTH – Antonio NAVARRO

 

Instance Does Not Start In Node 2

Today performing a series of tasks in a database I have seen that some of the nodes of the cluster had the database raised but outside the cluster, possibly started from the sqlplus with a startup command. Attempting to raise the instance from the srvctl I get this error;

 
orap12@starsw-1:…idades/SOP/Mscript$ srvctl start instance -db cloud -i cloud2
PRCR-1013 : Failed to start resource ora.cloud.db
PRCR-1064 : Failed to start resource ora.cloud.db on node starsw-2
CRS-2546: Server 'starsw-2' is not online

Check the CRS Daemon;

 
orap12@starsw-2:~$ ps -fe | grep crsd.bin
  orap12 24430 23483   0 17:39:53 pts/5       0:00 grep crsd.bin

I then try to lift the CRS and I get this other error

 
root@starsw-2 # crsctl start res ora.crsd -init
CRS-2672: Attempting to start 'ora.storage' on 'starsw-2'
diskgroup OCRVTD_DG not mounted ()
CRS-2674: Start of 'ora.storage' on 'starsw-2' failed
CRS-2679: Attempting to clean 'ora.storage' on 'starsw-2'
CRS-2681: Clean of 'ora.storage' on 'starsw-2' succeeded
CRS-4000: Command Start failed, or completed with errors.

So it seems the OCR diskgroup is not mounted. this seems to be the problem of everything

 
asmcmd>  mount OCRVTD_DG

Retry start the CRS

 
root@starsw-2 # crsctl start res ora.crsd -init
CRS-2672: Attempting to start 'ora.storage' on 'starsw-2'
CRS-2676: Start of 'ora.storage' on 'starsw-2' succeeded
CRS-2672: Attempting to start 'ora.crsd' on 'starsw-2'
CRS-2676: Start of 'ora.crsd' on 'starsw-2' succeeded

Although the previous command has worked, we do a check more;

 
root@starsw-2 #  crsctl check cluster                                
CRS-4537: Cluster Ready Services is online          
CRS-4529: Cluster Synchronization Services is online
CRS-4533: Event Manager is online             

after this I can now raise the instance of node 2 within the cluster with the svrctl command;

 
HTH – Antonio NAVARRO

ORA-00604 When Tracing Using Trigger For Logon

Today I was making a trace (10046 + 10053) of a user, for it activated the same on the current connection. The issue is that the application in a window to give the button to get report, below generated a new connection to the database, so the current trace did not capture anything of interest. In these cases an old trick is to generate a logon trigger for the user, which first activates the trace. The problem was that in my case I only captured this in the trace and it was aborted;

 

dbkedDefDump(): Starting a non-incident diagnostic dump (flags=0x0, level=0, mask=0x0)
----- Error Stack Dump -----
ORA-00604: se ha producido un error a nivel 1 de SQL recursivo
ORA-01031: privilegios insuficientes
ORA-06512: en l▒nea 4

 

This is a version 12c. The first step was to give DBA directly to the user because I was in a hurry to solve the problem but the DBA role did not solve the ORA-00600, which left me a little puzzled. Looking at the security model (Oracle is making great efforts in this area in recent years) was something as simple as giving an alter session to the user, that is, directly.

GRANT ALTER SESSION TO <USERNAME>;

I include a link to my GITHUB where I have my script to generate a logon trigger to activate traces.

Create Trigger Logon for Tracing

HTH – Antonio NAVARRO

 

 

Node 1 Does Not Support Connections (RAC)

This morning I have been reported a problem in a test environment. Some connections were hung when trying to enter. After connecting and taking a look, the environment, it is a cluster of two nodes version 12c, note that node one is the one that does not support connections and the connections were left is waiting for a change of log. It’s funny but it did not give any error, it’s usually this situations shows a ora-00257, but this is not the case, maybe because of the engine version.

Checking the situation at the instance level;

 
SQL> select inst_id, version, status, thread#, archiver, log_switch_wait, logins from gv$instance

   INST_ID VERSION           STATUS          THREAD# ARCHIVE LOG_SWITCH_WAIT LOGINS
---------- ----------------- ------------ ---------- ------- --------------- ----------
         2 12.1.0.2.0        OPEN                  2 STARTED                 ALLOWED
         1 12.1.0.2.0        OPEN                  1 STARTED ARCHIVE LOG     ALLOWED

 
Sessions are waiting on event “log file switch (archiving needed)” like show below, first column is node (in this case the problem is in node 1);

 
N     SID TIME    SQL_ID            EXECS EVENT                              Marker
- ------- ------- -------------- -------- ---------------------------------- --------
1      22 0       6vm7g6qj4mqhd        0  PX Deq: Execution Msg
2     327 0       6vm7g6qj4mqhd        4  PX Deq: Execution Msg
2     649 0       6vm7g6qj4mqhd        4  PX Deq: Execute Reply
1     361 1471    6228pzdt28kzd     1008  log file switch (archiving needed)      <<<
2     121 106     61tssjb6hj8x7      329  gc buffer busy acquire
1      88 1645    9zg9qd9bm4spu     7787  log file switch (archiving needed)      <<<
1     804 1704    3mptsg6h27zg9        1  log file switch (archiving needed)      <<<
1      89 1705    0bfdn75zn75pw        2  log file switch (archiving needed)      <<<
2      20 1699    6nauzjpthp1w7        3  db file sequential read
1       3 +1H     1aa2fpqtx557g     3877  log file switch (archiving needed)      <<<
1     328 1705    g8bkp70myp46t        2  log file switch (archiving needed)      <<<
1     257 +1H     dr6d1upgkc1g3        1  log file switch (archiving needed)      <<<
2     615 505     aq8yqxyyb40nn     1255  gc current request
1     800 817     aq8yqxyyb40nn     1282  buffer busy waits
1     480 817     aq8yqxyyb40nn     1282  buffer busy waits
2     342 +1H     5ms6rbzdnq16t    15644  gc buffer busy acquire
1     377 817     aq8yqxyyb40nn     1282  buffer busy waits
1     345 817     aq8yqxyyb40nn     1282  log file switch (archiving needed)      <<<
1      21 817     aq8yqxyyb40nn     1282  buffer busy waits

 
After perform archive backup (with delete option);

 
SQL> select inst_id, version, status, thread#, archiver, log_switch_wait, logins from gv$instance

   INST_ID VERSION           STATUS          THREAD# ARCHIVE LOG_SWITCH_WAIT LOGINS
---------- ----------------- ------------ ---------- ------- --------------- ----------
         2 12.1.0.2.0        OPEN                  2 STARTED                 ALLOWED
         1 12.1.0.2.0        OPEN                  1 STARTED                 ALLOWED

 
HTH – Antonio NAVARRO

 

ORA-00245 Error

Today when I was executing a archive backup after of full refresh (by OGG) I get the next error;

 
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of show command at 04/10/2019 09:26:58
RMAN-03014: implicit resync of recovery catalog failed
RMAN-03009: failure of full resync command on default channel at 04/10/2019 09:26:58
ORA-00245: control file backup failed; in Oracle RAC, target might not be on shared storage

The same is generated in the first sentence of the RMAN script (resyn catalog). The problem is that snap CF backup is in a DG, which is not permited. It must be out of DG. In this case ACFS or local disk (ZFS) is possible.

Like first action I performed a specified backup like show below;

 

SQL> alter database backup controlfile to ‘/prod/backup/CONTROLFILE/snapcf_nube.f’ reuse;

 
The second action is make this change permanent. From RMAN you can use;

 

CONFIGURE CONTROLFILE AUTOBACKUP OFF;

CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK CLEAR;

CONFIGURE SNAPSHOT CONTROLFILE NAME clear;

CONFIGURE SNAPSHOT CONTROLFILE NAME TO ‘/prod/backup/CONTROLFILE/snapcf_nube.f’;

 
HTH – Antonio NAVARRO

Error ORA-01924

Recently I have been migrating a database from version 10g to 12c. In the functionality testing phase the error ORA-01924 arose. This occurs (in version 12.1) when the 600 roles assigned to a user are exceeded, Oracle has a specific oneoff collected in bug 18934948. But after applying this oneoff, the problem. After doing several tests we saw that it was a problem of lack of privileges. The user who made the mistake He lacked the ability to interact with roles.

In version 10g I did not have any problem with the level of privileges assigned, but in version 12c I needed to add two new permits. It may seem strange the difference of operation between versions, but it really makes sense if you look at the model of continuity. Oracle in the last has accelerated this part of the databases, in general it has been a leader in many aspects, but here has always been one step behind other  rdbms such as SQL Server.

In this case giving these two privileges to the user solved the problem;

  • drop role
  • grant any role

HTH – Antonio NAVARRO.