Skip to Main Content

Oracle Database Discussions

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

Impossible high values on seconds_in_wait column from V$SESSION/

Luis Claudio SantosJun 13 2018 — edited Jun 14 2018

In some databases on my site (all of them on Linux host and 11.2.0.4 version) I noticed that SECOND_IN_WAIT column from V$SESSION or V$SESSION_WAIT has impossible high values. To be more precise, more then 47 years.

This only happens on background sessions (SESSION_TYPE='BACKGROUND').

I have some monitoring scripts that are been raised due this problem.

When I focus on DB Writers I see that all of them are normally waiting for rdbms ipc message. Ok, this is the normal behaviour. My monitoring script are not been raised in this situation.

[11g]> @dbwr_waits

PROGRA   SID SPID      EVENT                                SEQ# wait_time (s) SECONDS_IN_WAIT STATE

------ ----- --------- ------------------------------ ---------- ------------- --------------- -------------------

(DBWA)  6021 66809     rdbms ipc message                     193          0,00      1528903014 WAITING

(DBWB)  6322 66811     rdbms ipc message                   46961          0,00      1528903014 WAITING

(DBWC)  6623 66813     rdbms ipc message                    5592          0,00      1528903014 WAITING

(DBWD)  6924 66815     rdbms ipc message                    7218          0,00      1528903014 WAITING

...

But If I get some write on db file I got this situation:

[11g]> @dbwr_waits

PROGRA   SID SPID      EVENT                                SEQ# wait_time (s) SECONDS_IN_WAIT STATE

------ ----- --------- ------------------------------ ---------- ------------- --------------- -------------------

(DBWA)  6021 66809     rdbms ipc message                    3893          1,99      1528903205 WAITED KNOWN TIME

(DBWB)  6322 66811     rdbms ipc message                   50603          1,64      1528903205 WAITED KNOWN TIME

(DBWC)  6623 66813     rdbms ipc message                    9322          1,46      1528903204 WAITED KNOWN TIME

(DBWD)  6924 66815     rdbms ipc message                   10765          2,22      1528903205 WAITED KNOWN TIME

(DBWE)  7225 66817     rdbms ipc message                   31151          0,40      1528903203 WAITED KNOWN TIME

(DBWF)  7526 66819     rdbms ipc message                    3027          1,69      1528903205 WAITED KNOWN TIME

(DBWG)  7827 66821     rdbms ipc message                   24244          2,64      1528903206 WAITED KNOWN TIME

(DBWH)  8128 66823     rdbms ipc message                   18634          2,76      1528903206 WAITED KNOWN TIME

(DBWI)  8429 66825     rdbms ipc message                   61980          1,79      1528903205 WAITED KNOWN TIME

(DBWJ)  8730 66827     rdbms ipc message                   38905          2,84      1528903206 WAITED KNOWN TIME

(DBWK)  9031 66829     rdbms ipc message                   63753          2,99      1528903206 WAITED KNOWN TIME

(DBWL)  9332 66831     db file parallel write              19677         -0,01      1528903203 WAITED SHORT TIME

(DBWM)  9633 66833     rdbms ipc message                   35403          1,48      1528903204 WAITED KNOWN TIME

In this situation my monitoring script thinks I have a stuck write on storage, and the 24/7 DBA is alarmed.

I could not find no doc or bug about this problem, except a very old one, for 8.1.6 RDBMS version.

Comments
Locked Post
New comments cannot be posted to this locked post.
Post Details
Locked on Jul 12 2018
Added on Jun 13 2018
9 comments
1,432 views