max_failure and noreopen parameter in dataguard?

  • From: Leyi Zhang <kamusis@xxxxxxxxx>
  • To: oracle-l@xxxxxxxxxxxxx
  • Date: Wed, 12 Jan 2005 13:58:31 +0800

My database is Oracle9201 on Solaris 8
I set log_archive_dest_2='SERVICE=ctsdb.standby lgwr async=20480
net_timeout=15 max_failure=1 noreopen'
and the size of redofile in primary site is 3M.

I run a procedure which insert 600000 records in loop to test this
dataguard environment.

I unplugged the standby site's network, due to the online document, that
the primary site will only check the netwotk once, if failed then never
archive to that destination until we reenable it.

but in fact, I found primary site always check the network whenever the
log switch occured. and I checked the v$archive_dest's STATUS column,
the 'log_archive_dest_2' is 'VALID'.

After I cancel the procedure,  'log_archive_dest_2' STATUS column in
v$archive_dest change to 'ERROR'.

But after I replug the network at standby site, I ran "alter system
switch logfile;", all the gap archivelog files begin to transfer to the
standby site and all seems OK. 
I thought this is not "noreopen"'s meaning!

what did the "net_timeout=15 max_failure=1 noreopen" means on earth?

Kamus <kamusis@xxxxxxxxx>

A Oracle8i & 9i Certified DBA from China


Other related posts: