Thinking Out Loud

April 28, 2017

Bug 18411339 – Low performance V$ARCHIVE_GAP ( fix

Filed under: 11g,12c,Dataguard — mdinh @ 12:31 am

Just came across bug from not fixed until 12.2 base release. Seriously Oracle?
In the test case below, it looks to have only affected – 64bit for AIX Version 7.1 since I recall this was not an issues for Linux.
select * from v$archive_gap;
Elapsed: 00:01:48.93
select * from v$archive_gap;
Elapsed: 00:00:06.60

Work Around

select USERENV('Instance'), high.thread#, low.lsq, high.hsq
  (select a.thread#, rcvsq, min(a.sequence#)-1 hsq
   from v$archived_log a,
        (select lh.thread#, lh.resetlogs_change#, max(lh.sequence#) rcvsq
           from v$log_history lh, v$database_incarnation di
          where lh.resetlogs_time = di.resetlogs_time
            and lh.resetlogs_change# = di.resetlogs_change#
            and di.status = 'CURRENT'
            and lh.thread# is not null
            and lh.resetlogs_change# is not null
            and lh.resetlogs_time is not null
         group by lh.thread#, lh.resetlogs_change#
        ) b
   where a.thread# = b.thread#
     and a.resetlogs_change# = b.resetlogs_change#
     and a.sequence# > rcvsq
   group by a.thread#, rcvsq) high,
 (select srl_lsq.thread#, nvl(lh_lsq.lsq, srl_lsq.lsq) lsq
     (select thread#, min(sequence#)+1 lsq
        v$log_history lh, x$kccfe fe, v$database_incarnation di
      where to_number(fe.fecps) <= lh.next_change# and to_number(fe.fecps) >= lh.first_change#
        and fe.fedup!=0 and bitand(fe.festa, 12) = 12
        and di.resetlogs_time = lh.resetlogs_time
        and lh.resetlogs_change# = di.resetlogs_change#
        and di.status = 'CURRENT'
      group by thread#) lh_lsq,
     (select thread#, max(sequence#)+1 lsq
      where (select min( to_number(fe.fecps))
             from x$kccfe fe
             where fe.fedup!=0 and bitand(fe.festa, 12) = 12)
      >= next_change#
      group by thread#) srl_lsq
   where srl_lsq.thread# = lh_lsq.thread#(+)
  ) low
 where low.thread# = high.thread#
 and lsq < = hsq and hsq > rcvsq;


  1. You beauty! I was struggling with this then found your blog! This has saved me a lot of head pounding!

    Comment by Bob Mycroft — May 15, 2018 @ 6:38 pm | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Blog at

%d bloggers like this: