RMAN BACKUP FAILED ISSUES

RMAN Backup takes more time than earlier.

RMAN Backup failed with error RMAN-06059: expected archived log not found

ORA-19625: error identifying file /u02/oradata/archive/1_prod_3241.arc.

            we did not delete archive log through RMAN. we delete archive log directly using rm -rf command, so RMAN is unaware of those archive logs status. if you are experiencing same problem that your backup failed because of missing archive log then do following things.

Connect to RMAN as RMAN TARGET / and then

crosscheck archivelog all;

delete expired archivelog all;

then connect to catalog database and resync the catalog.

rman target / catalog rman/rman_2005@RMANCAT

resync catalog;

            one more thing is that my RMAN backup took more time as it also looks for unavailable  archive log as well, When an archive log crosscheck is performed, RMAN checks each archive log in turn to make sure that it exists on disk. Those that are missing are marked as unavailable. you can skip the particular range of missing archivelog while taking the backup. connect to catalog.

rman target / catalog rman/rman_2005@RMANCAT

run

{

change archivelog from logseq = xxx until logseq = yyy unavailable;

}

  rman>resync catalog;

ORA-00257: archiver error. Connect internal only, until freed.

                 It doesn’t matter, whether you working in large or small organization. Space allocation is the common issue. Management wants all prod databases should be in archive mode, archive log should not be deleted before backup and at least 3 day archive should be there on server. Yes it should be like this, but the problem is they provide that much space which is capable to maintain 3-4 days logs. If backup fails, any bat job, scheduler runs from application server which generates high archive generation and you doesn’t aware of it and the archive destination gets full. You will get error like.

ORA-00257: archiver error. Connect internal only, until freed.

Delete archive logs and make sufficient space on archive destination and then switch logs. That’s all; it will solve your problem.

SQL> alter system archive log all;

You are well aware to solve this issue, yes you have to free space in archive destination else no transaction will process on database server. Delete archive logs which are backed up and sync with standby. If the archive generation is much high and you can’t stop the job. In this case there is chance that you have to delete archive logs which are not backed up also which are not transferred to standby. You can bring standby database in sync later. By taking incremental backup of Production. Before that you should inform to your manager because he should be aware of this. Not to worry, you haven’t done anything wrong. unless application team informs you about their activity. As a core DBA You can’t control on archive generation. Use below query to find pattern of archive generation. This report will work like RCA forward it whomever the concern.

select trunc(first_time) on_date,

       thread# thread,

       min(sequence#) min_sequence,

       max(sequence#) max_sequence,

       max(sequence#) - min(sequence#) nos_archives,

       (max(sequence#) - min(sequence#)) * log_avg_mb req_space_mb

from   v$log_history,

       (select avg(bytes/1024/1024) log_avg_mb

        from   v$log)

group  by trunc(first_time), thread#, log_avg_mb

order by trunc(first_time) desc.

Don’t forget to fire full database backup.

Listener Not starting on Oracle11G gateway server.

We have Oracle 11G gateway in production and UAT environment, used to fetch data from sql database to oracle database and vise versa using DBLINKS. The issue comes whenever we required listener restart after adding entry in listener, it doesn’t start. It displays message “some services stop automatically if they are not in use” . I found some dg4msql.exe`s are running, so I tried to kill them but it doesn’t allow me. I decided to run task manager under Local System Account.

1)      Login to the server using admin console.

2)      Open command prompt and type command “AT 12:00 /interactive cmd.exe”

3)      Open task scheduler and run created task, it will launch Command prompt.

4)      Type “taskmgr” on Command prompt, it will launch task manager. Now you will able to kill services using task manager.

5)      Kill all dg4msql.exe services.

6)      Now you are able to start Listener service.

Listener Crashes in Windows

Sometimes, it doesn’t matter how expert you are! I believe I have good knowledge on datagaurd but it took me three days to solve simple issue. Oh it doesn’t mean I am not expert, because culprit was not Oracle its Windows. 

Issue was, I had configured standby database till then everything was right, problem occurs whenever I put standby database in recovery, it fetches one or two logs from productions and then listener crashes.  I configured different listener even standby database as well but no Luck.

Error prompt:

Faulting application TNSLSNR.EXE, version 0.0.0.0, faulting module orantcp9.dll, version 0.0.0.0, fault address 0×00005732

After lot of goggling I got one magical command.  

netsh winsock reset catalog

I surprised, this simple command solves my problem immediately. Oh god, I used this command many times in past but failed to relate it with this issue.

Most of the Internet connectivity problems are because of the corruption in Winsock settings. Windows sockets settings may get corrupted due to the installation of a networking software, Malware infestation. In this situation user can connect to the Internet, but randomly may get Page cannot be displayed .Above command resets the Winsock catalog to the default configuration. This can be useful if a malformed LSP is installed that results in loss of network connectivity.