command failed -- Enterprise Replication not activ
Posted in 2010
User on IDS 11.50.FC6X3 (SUSE) tried to set up Enterprise Replication between two dev instances and kept getting 'command failed -- Enterprise Replication not active' from cdr define server. Madison pointed out the command must run against the node being defined (e.g. cdr define server --connect=g_cdrdev1 --init g_cdrdev1, then --connect=g_cdrdev2 --init -S g_cdrdev1 g_cdrdev2) and asked to see sqlhosts. After posting sqlhosts, the poster found a conflict: removing the ids_cluster group entries from both sqlhosts files (plus engine restarts) let both CDR servers be defined successfully. The reason for the conflict was never explained.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: High Availability & Replication, Networking & sqlhosts Configuration
IDS: 11.50.FC6X3
OS: SUSE 10.2
A long time ago we had ER setup between our dev server, psdev1, and an
evaluation instance. The eval instance is long gone and now I want to set up
ER between two dev servers, psdev1 and psdev2. But what ever I do I get the
error:
command failed -- Enterprise Replication not active
psdev1 had syscdr db so I brought the engine up with oninit -D and dropped
that db. I then bounced the psdev1 engine using oninit -v.
psdev2 is a new instance so has never had ER running.
Here is the command I'm using to create a new ER server on psdev1:
cdr define server --connect=cdrdev1 --init g_cdrdev2
I've tried defining a server in the other direction also:
cdr define server --connect=cdrdev2 --init g_cdrdev1
But it is always the same error.
Any ideas?
These lines are in the psdev1 sqlhosts file:
g_cdrdev2 group - - i=65
cdrdev1 onsoctcp xxx.xx.xx.9 1550
cdrdev2 onsoctcp xxx.xx.xx.10 1550 g=g_cdrdev2
The define server command must be executed on the server which you want=
to
define.
If you do "cdr define server --connect=3Dx --init y", then you are exec=
uting
on node 'x' the creation of 'y'. That isn't correct.
Try issuing "cdr define server --connect=3Dg_cdrdev1 --init g_cdrdev1"
followed by
"cdr define server --connect=3Dg_cdrdev2 --init -S g_cdrdev1 g_cdrdev2"=
M.Pruet
=
From: "BEVIS KENNEDY" <bkennedy@utah.gov> =
=
To: ids@iiug.org =
=
Date: 03/23/2010 08:33 AM =
=
Subject: command failed -- Enterprise Replication not activ [1938=
7]
=
Sent by: ids-bounces@iiug.org =
=
IDS: 11.50.FC6X3
OS: SUSE 10.2
A long time ago we had ER setup between our dev server, psdev1, and an
evaluation instance. The eval instance is long gone and now I want to s=
et
up
ER between two dev servers, psdev1 and psdev2. But what ever I do I get=
the
error:
command failed -- Enterprise Replication not active
psdev1 had syscdr db so I brought the engine up with oninit -D and drop=
ped
that db. I then bounced the psdev1 engine using oninit -v.
psdev2 is a new instance so has never had ER running.
Here is the command I'm using to create a new ER server on psdev1:
cdr define server --connect=3Dcdrdev1 --init g_cdrdev2
I've tried defining a server in the other direction also:
cdr define server --connect=3Dcdrdev2 --init g_cdrdev1
But it is always the same error.
Any ideas?
These lines are in the psdev1 sqlhosts file:
g_cdrdev2 group - - i=3D65
cdrdev1 onsoctcp xxx.xx.xx.9 1550
cdrdev2 onsoctcp xxx.xx.xx.10 1550 g=3Dg_cdrdev2
***********************************************************************=
********
Forum Note: Use "Reply" to post a response in the discussion forum.
=
Thanks for the response.
I changed the sqlhosts files so g_cdrdev1 and g_cdrdev2 groups are defined on
both servers. I ran the command:
cdr define server --connect=cdrdev1 --init g_cdrdev1
the first time I got:
connect to cdrdev1 failed
INFORMIXSERVER does not match either DBSERVERNAME or DBSERVERALIASES.
(-761)
command failed -- unable to connect to server specified (5)
(it is defined as DBSERVERALIASES in the onconfig)
so I ran the command again and got the original error:
command failed -- Enterprise Replication not active (62)
Interestingly enough this time it created the 'syscdr' db. This was in the
syslog and I can see the db in dbaccess:
09:06:22 Building 'syscdr' database ...
09:06:22 'syscdr' database built successfully.
I tried cdr list server and the error remains the same.
arrggghhhh!
post your sqlhost file.
=
From: "BEVIS KENNEDY" <bkennedy@utah.gov> =
=
To: ids@iiug.org =
=
Date: 03/23/2010 10:20 AM =
=
Subject: Re: command failed -- Enterprise Replication not a [1939=
1]
=
Sent by: ids-bounces@iiug.org =
=
Thanks for the response.
I changed the sqlhosts files so g_cdrdev1 and g_cdrdev2 groups are defi=
ned
on
both servers. I ran the command:
cdr define server --connect=3Dcdrdev1 --init g_cdrdev1
the first time I got:
connect to cdrdev1 failed
INFORMIXSERVER does not match either DBSERVERNAME or DBSERVERALIASES.
(-761)
command failed -- unable to connect to server specified (5)
(it is defined as DBSERVERALIASES in the onconfig)
so I ran the command again and got the original error:
command failed -- Enterprise Replication not active (62)
Interestingly enough this time it created the 'syscdr' db. This was in =
the
syslog and I can see the db in dbaccess:
09:06:22 Building 'syscdr' database ...
09:06:22 'syscdr' database built successfully.
I tried cdr list server and the error remains the same.
arrggghhhh!
***********************************************************************=
********
Forum Note: Use "Reply" to post a response in the discussion forum.
=
psdev1:
g_cdrdev1 group - - i=63
cdrdev1 onsoctcp XXX.XX.XX.9 1550 g=g_cdrdev1
g_cdrdev2 group - - i=65
cdrdev2 onsoctcp XXX.XX.XX.10 1550 g=g_cdrdev2
ids_cluster group - - i=64
psdev1 onsoctcp XXX.XX.XX.9 1546 g=ids_cluster
psdev2 onsoctcp XXX.XX.XX.10 1546 g=ids_cluster
cm_cluster group - - i=66,c=1
psdev1_cm onsoctcp XXX.XX.XX.9 1559 g=cm_cluster
psdev2_cm onsoctcp XXX.XX.XX.10 1559 g=cm_cluster
# Connection Manager
psdev_soc onsoctcp XXX.XX.XX.9 1547
psdev_main onsoctcp XXX.XX.XX.9 1548
psdev_rpt onsoctcp XXX.XX.XX.9 1549
psdev2:
g_cdrdev1 group - - i=63
cdrdev1 onsoctcp XXX.XX.XX.9 1550 g=g_cdrdev1
g_cdrdev2 group - - i=65
cdrdev2 onsoctcp XXX.XX.XX.10 1550 g=g_cdrdev2
# psdev server cluster
ids_cluster group - - i=64
psdev1 onsoctcp XXX.XX.XX.9 1546 g=ids_cluster
psdev2 onsoctcp XXX.XX.XX.10 1546 g=ids_cluster
# cm cluster
ids_cm group - - i=66,c=1
psdev1_cm onsoctcp XXX.XX.XX.9 1559 g=ids_cm
psdev2_cm onsoctcp XXX.XX.XX.10 1559 g=ids_cm
# Connection manager stuff
psdev_soc onsoctcp XXX.XX.XX.10 1547
psdev_main onsoctcp XXX.XX.XX.10 1548
psdev_rpt onsoctcp XXX.XX.XX.10 1549
Yahoo, it's working. It looks like there was a conflict in the sqlhosts file. By removing the ids_cluster group on from both groups (and multiple engine bounces) I was able to create cdr servers on both engines. I don't know why that made a difference but that mystery can wait for another day. Thanks for your help.