Removing CDR

11.50.FC8 I have two servers that replicate between each other. I recently did a restore of one of these severs to a third server. My objective is to define a third er server on this machine and have three way replication between these servers. The issue I am facing is that I need to first remove the existing CDR database and all its associated baggage (cdr delete tables in all databases) on the server that I just restored . However I cannot even get ER to start. cdr start complains ER is already active and cdr stop says its not active. All this is I presume the result of the restore etc. Any quick way to drop the syscdr database without just dropping it and combing thru all the databases looking for cdr_delete tables that would also need to be cleaned up ? I tried cdr delete server xxx -force but get the same "er is not active" error. I will be re-doing the restore again so if there is way to do this more cleanly I surely will appreciate it. Thanks, Mark

Hi, We do this regularly as follows: Bring up the database with oninit -D so that replication is bypassed. Drop the syscdr database. Rebuild sysmaster by running $INFORMIXDIR/etc/buildsmi. Restart the engine. Then you can rebuild the replication as needed. I hope this helps. Pam ________________________________________ From: ids-bounces@iiug.org [ids-bounces@iiug.org] On Behalf Of MARK JALKIEWICZ [mark.jalkiewicz@verizon.net] Sent: Thursday, March 06, 2014 5:11 PM To: ids@iiug.org Subject: Removing CDR [32675] 11.50.FC8 I have two servers that replicate between each other. I recently did a restore of one of these severs to a third server. My objective is to define a third er server on this machine and have three way replication between these servers. The issue I am facing is that I need to first remove the existing CDR database and all its associated baggage (cdr delete tables in all databases) on the server that I just restored . However I cannot even get ER to start. cdr start complains ER is already active and cdr stop says its not active. All this is I presume the result of the restore etc. Any quick way to drop the syscdr database without just dropping it and combing thru all the databases looking for cdr_delete tables that would also need to be cleaned up ? I tried cdr delete server xxx -force but get the same "er is not active" error. I will be re-doing the restore again so if there is way to do this more cleanly I surely will appreciate it. Thanks, Mark ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

I was once told by support that you don't need to rebuild sysmaster if you are not using update anywhere, even if you are then a restart after dropping syscdr is just as good Cheers Paul > -----Original Message----- > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of > Ekstrand, Pamela A. -ND > Sent: Thursday, March 06, 2014 4:38 PM > To: ids@iiug.org > Subject: RE: Removing CDR [32676] > > Hi, > > We do this regularly as follows: > > Bring up the database with oninit -D so that replication is bypassed. > Drop the syscdr database. > Rebuild sysmaster by running $INFORMIXDIR/etc/buildsmi. > Restart the engine. > > Then you can rebuild the replication as needed. > > I hope this helps. > > Pam > ________________________________________ > From: ids-bounces@iiug.org [ids-bounces@iiug.org] On Behalf Of MARK > JALKIEWICZ > [mark.jalkiewicz@verizon.net] > Sent: Thursday, March 06, 2014 5:11 PM > To: ids@iiug.org > Subject: Removing CDR [32675] > > 11.50.FC8 > > I have two servers that replicate between each other. I recently did a restore > of one of these severs to a third server. My objective is to define a third er > server on this machine and have three way replication between these > servers. > > The issue I am facing is that I need to first remove the existing CDR database > and all its associated baggage (cdr delete tables in all databases) on the > server that I just restored . However I cannot even get ER to start. cdr start > complains ER is already active and cdr stop says its not active. All this is I > presume the result of the restore etc. Any quick way to drop the syscdr > database without just dropping it and combing thru all the databases looking > for cdr_delete tables that would also need to be cleaned up ? I tried cdr > delete server xxx -force but get the same "er is not active" error. > > I will be re-doing the restore again so if there is way to do this more > cleanly I surely will appreciate it. > > Thanks, > > Mark > > > ********************************************************** > ********************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > ********************************************************** > ********************* > Forum Note: Use "Reply" to post a response in the discussion forum.

Have you tried bringing up your engine with oninit -D And then drop the syscdr database and then remove all the cdr del tables? -----Original Message----- From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of MARK JALKIEWICZ Sent: Thursday, March 06, 2014 4:12 PM To: ids@iiug.org Subject: Removing CDR [32675] 11.50.FC8 I have two servers that replicate between each other. I recently did a restore of one of these severs to a third server. My objective is to define a third er server on this machine and have three way replication between these servers. The issue I am facing is that I need to first remove the existing CDR database and all its associated baggage (cdr delete tables in all databases) on the server that I just restored . However I cannot even get ER to start. cdr start complains ER is already active and cdr stop says its not active. All this is I presume the result of the restore etc. Any quick way to drop the syscdr database without just dropping it and combing thru all the databases looking for cdr_delete tables that would also need to be cleaned up ? I tried cdr delete server xxx -force but get the same "er is not active" error. I will be re-doing the restore again so if there is way to do this more cleanly I surely will appreciate it. Thanks, Mark ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

my colleague wrote this that works on 11.5 # # EMERGENCY kill of replication by direct deletions of tables, database and views. # to be run as user informix - 9.40 version. # this is necessary sometimes when image has been restored to foreign host and replication # database is also restored but has wrong host and other params # this is the SHELL version if [ `whoami` != "informix" ] then echo " Please run this script as informix user" exit 1 fi # open sysmaster database and get databases to file db1 echo "select name from sysdatabases" | dbaccess sysmaster > /tmp/db1 2>/dev/null # cat /tmp/db1 # get table names from each database (make command file db2, ouput in db3) echo "scanning databases for cdr_deltab tables" awk ' BEGIN { OFS=""; FS=" "; } /name/ { print ("echo DATABASE " , $2); print ("echo \\\\"select tabname from systables\\\\" | dbaccess ", $2, " 2>/dev/null" ); } ' /tmp/db1 > /tmp/db2 2>/dev/null . /tmp/db2 > /tmp/db3 # cat /tmp/db2 # finally ! create drop commands for each cdrdeltab table echo "Dropping cdr_deltab tables" awk ' BEGIN { OFS=""; FS=" "; } /DATABASE/ { db=$2; } /cdr_deltab/ { print ("echo \\\\"DB ", db, " Dropping Table ", $2 "\\\\""); print ("echo \\\\"drop table ", $2 ,"\\\\" | dbaccess ", db, " 2>/dev/null" ); } ' /tmp/db3 > /tmp/db2 # cat /tmp/db2 . /tmp/db2 # Now drop the cdr extra tables and views; echo "Cleaning out sysmaster replication tables" echo " drop view syscdrgrp" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrserver" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrerror" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrgrppart" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrrepl" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrpart" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrqueued" | dbaccess sysmaster 2>/dev/null echo " drop view syscdrtxproc" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrack_buf where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrack_txn where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrctrl_buf where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrctrl_txn where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrprog where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrq where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrrecv_buf where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrrecv_txn where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrs where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrsend_buf where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrsend_txn where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrsync_buf where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrsync_txn where 1=1" | dbaccess sysmaster 2>/dev/null echo "delete from syscdrtx where 1=1" | dbaccess sysmaster 2>/dev/null #echo "Dropping syscdr" #echo "drop database syscdr" | dbaccess sysmaster 2>/dev/null rm /tmp/db1 rm /tmp/db2 rm /tmp/db3 echo "clean up is done !!" /home/informix/replication [informix wldbp654 dbp654_ecms_live]