1

Topic: DB hangup

Good afternoon.
There was the strange piece.
At DB start there was an error:
[spoiler] ORA-12012:
ORA-06502: PL/SQL::
ORA-06512:
ORA-06502:
ORA-06512:
ORA-12514: TNS:
ORA-06512: [/spoiler]
After that the DB was launched, all norms.
After time there was an error:
[spoiler] Fatal NI connect error 12514, connecting to:
(DESCRIPTION = (ADDRESS = (PROTOCOL=TCP) (HOST=10.0.0.21) (PORT=1521)) (CONNECT_DATA = (SERVER=DEDICATED) (SERVICE_NAME=test) (CID = (PROGRAM=oracle) (HOST=dboracle.dom) (USER=o
racle))))
VERSION INFORMATION:
TNS for Linux: Version 12.1.0.2.0 - Production
TCP/IP NT Protocol Adapter for Linux: Version 12.1.0.2.0 - Production
Time: 07-AUG-2018 4:45:24 AM
Tracing not turned on.
Tns error struct:
ns main err code: 12564
TNS-12564: TNS:connection refused
ns secondary err code: 0
nt main err code: 0
nt secondary err code: 0
nt OS err code: 0 [/spoiler]
In result of the given error possibility to be connected to a DB of everything, except sysdba disappeared. As the DB hung up at any reversal to it, for example at show parameter name.
Serviceability of a DB recovered correction pfile, namely  repeating parameters of type: "*." And "sid.", different values had the majority of these parameters. For an example:
[spoiler] sid.log_archive_dest_1 ="
*.log_archive_dest_1 ='location=USE_DB_RECOVERY_FILE_DEST ', ' valid_for = (ALL_LOGFILES, ALL_ROLES) ' [/spoiler]
Question: what was from a DB and where to dig?

2

Re: DB hangup

Ilnov;
I can something is finite I do not understand, but how you could appropriate two different values for one parameter of initialization?
And where you saw such?

 [color=red] sid. [/color] log_archive_dest_1 = "

3

Re: DB hangup

maverick2104 wrote:

Ilnov;
I can something is finite I do not understand, but how you could appropriate two different values for one parameter of initialization?
And where you saw such?

 [color=red] sid. [/color] log_archive_dest_1 = "

In the documentation :

 alter system set sga_target=6016M scope=spfile [color=red] sid ='testdb ' [/color]; 

Under the total in spfile:

...
*.sga_target=5033164800
>>> testdb.sga_target=6308233216 <<<
...

Ilnov, to begin with to dig in alert.log on a subject alter system.

4

Re: DB hangup

maverick2104, for example in  pfile/spfile:
[spoiler] n. __ db_cache_size=1761607680
tn. __ java_pool_size=33554432
tn. __ large_pool_size=83886080
tn. __ oracle_base ='C:\app'#ORACLE_BASE set from environment
tn. __ pga_aggregate_target=838860800
tn. __ sga_target=2516582400
tn. __ shared_io_pool_size=117440512
tn. __ shared_pool_size=486539264
tn. __ streams_pool_size=16777216 [/spoiler]
Parameter "sid." It is applied before parameter "*." At instructions of the same parameter.
To specify the second same parameter in spfile it is possible alter system set... sid ='db_name ';

5

Re: DB hangup

If the data is so invaluable that does not demand the qualified supervision not only from regular DBA, but even remote - that the given DB has no practical value and can be without serious consequences disconnected.
At such approach with switch-off standby (possibly settled resources or it is banal filled up) it is possible to consider experiments quite to itself harmless.
For additional effect it is possible to recommend to translate a DB in nologging,  stored archive logs and to nobody the necessary backup copies, and a vacant place to use under something more valuable to business - for example, under a collection of video with cats.

6

Re: DB hangup

If to me does not change a sclerosis - similar parameters ( sid. log_archive_dest etc. at existing *. log_archive_dest) liked to push at each opportunity DG Broker
Here only on serviceability of a DB (that-more, registration in ) it hardly should influence...
Whether and is cool informative error report a case at  the trigger on switching of roles? smile))

7

Re: DB hangup

Ljubomudrov wrote:

whether And is cool informative error report a case at  the trigger on switching
Roles? smile))

The question only - where exactly the HARDWARE watched the "underground knock".
Judging by presentation - somewhere at level of client connection.
And the effect of "hangup" of a DB could be an impossibility consequence  archive log to destination - for example, is banal the empty seat in destination was settled.

8

Re: DB hangup

andrey_anonymous wrote:

Judging by presentation - somewhere at level of client connection.
And the effect of "hangup" of a DB could be an impossibility consequence  archive log to destination - for example, is banal the empty seat in destination was settled.

Not only on client side, but also on the server at connection sqlplus / as sysdba hang happened, is not dependent on a type of reversal to a DB.
On a question archlog, there were such thoughts, but the place for them was free on 70 %.

9

Re: DB hangup

Ljubomudrov wrote:

Whether and is cool informative error report a case at  the trigger on switching of roles? smile))

There is no it an error in  job. It resulted from reasons, and suddenly. It is not enough easier variants for understanding of a problem.

10

Re: DB hangup

Evrika!
Analyzed alert a broad gull more attentively and saw following interesting things:
Cutting from dens during DB start:
[spoiler] WARNING: No local destinations have been defined for
archival of the Redo Log. [/spoiler]
It before editing spfile.
Probably something in parameter log_archive_dest_1 was not pleasant to it.
andrey_anonymous Thanks for the help in a search direction.
After editing, in a broad gull of start of the given warning there is no also a DB almost days works in a normal mode.
All thanks for the help!
P.S.
Read broad gulls! :-)