Start HADR cannot complete...
rakeysh
I was trying to setup a HADR for the sample database on two servers
under the windows environment.
I am using IBM DB2 9.5 FP "0" and windows server 2003 SP2.
I set up the configuration parameters correctly on both the primary
server and standby server.
First I tried with GUI which run fune till the point it reaches
"update services file".
and then it threw an error which was like javaw.exe : application
error
and then the control center exits.
Then I tried to set up manually .and It run fine till I start primary
server
DB2 start HADR on db SAMPLE as primary
the error it gives is different when I run it different time with RC=1
and RC=99 and RC=8
SQL1767N Start HADR cannot complete. Reason code = "1".
Standby server is running successfully.
can somebody please tell me what I am missing?
Thanks ,
Rakesh
Larry
http://www-01.ibm.com/support/docview.wss?rs=71&context=SSEPGG&context=SSEPDU&context=SSVGXH&context=SSVGZB&context=SSFHEG&context=SSYK8P&context=SSTLZ9&q1=sql1767n&uid=swg21225805&loc=en_US&cs=utf-8&lang=en
Or search here
http://www-01.ibm.com/software/data/db2/support/db2_9/troubleshoot.html
for "SQL1767N"
Larry Edelstein
Steve Pearson (news only)
particular RC 8 does not seem to be defined for SQL1767N.
It may be from the nearby SQL1768N instead.
Check these links, or get help from db2; e.g. from clp try "?
SQL1767N"
You may also want to check your diagnostic log (db2diag.log) on both
primary and standby to look for messages at the time Start HADR was
attempted that may indicate what went wrong.
If you continue to have trouble, contact service.
> See http://www-01.ibm.com/support/docview.wss?rs=71&context=SSEPGG&contex...
Hmm, I'll have to send feedback on this support note, because it looks
strangely wrong to me.
SQL1767N rc 1 means that the standby (not primary) database is not in
the correct state when you try to start HADR, and in particular, it is
not marked as being ready for a "rollforward" type operation to begin
or resume. If you just performed a Restore DB (or otherwise properly
initialized the HADR standby), it would be marked as "rollforward
pending". If you have interrupted an already started HADR standby,
the db there would be marked as "rollforward in progress". It should
be one of those.
The text from the doc is:
"The database was not in roll forward-pending or roll forward-in-
progress state when the START HADR AS STANDBY command was issued."
which seems to have been misconstrued in the support note.
Regards,
- Steve P.
--
Steve Pearson, DB2 for Linux, UNIX, and Windows, IBM Software Group
"Portland" Development Team, IBM Beaverton Lab, Beaverton, OR, USA
rakesh
wrote:
> I don't think all of those RC are from the same error, since in
> particular RC 8 does not seem to be defined for SQL1767N.
> It may be from the nearby SQL1768N instead.
>
> Check these links, or get help from db2; e.g. from clp try "?
> SQL1767N"
>
> You may also want to check your diagnostic log (db2diag.log) on both
> primary and standby to look for messages at the time Start HADR was
> attempted that may indicate what went wrong.
>
>
> http://publib.boulder.ibm.com/infocenter/db2luw/v9r5/index.jsp?topic=...
>
> If you continue to have trouble, contact service.
>
>
> Hmm, I'll have to send feedback on this support note, because it looks
> strangely wrong to me.
>
> SQL1767N rc 1 means that the standby (not primary) database is not in
> the correct state when you try to start HADR, and in particular, it is
> not marked as being ready for a "rollforward" type operation to begin
> or resume. If you just performed a Restore DB (or otherwise properly
> initialized the HADR standby), it would be marked as "rollforward
> pending". If you have interrupted an already started HADR standby,
> the db there would be marked as "rollforward in progress". It should
> be one of those.
>
> The text from the doc is:
>
> "The database was not in roll forward-pending or roll forward-in-
> progress state when the START HADR AS STANDBY command was issued."
>
> which seems to have been misconstrued in the support note.
>
> Regards,
> - Steve P.
> --
> Steve Pearson, DB2 for Linux, UNIX, and Windows, IBM Software Group
> "Portland" Development Team, IBM Beaverton Lab, Beaverton, OR, USA
Thanks Steve!
Its been quite some time when I posted this problem....and in the
meanwhile there has been some development in setting up HADR on
windows.
The good news is that I have successfully done the HADR configuration
on windows XP machine and I did exactly what the redbooks suggested.
The one thing I was missing was setting up the LOGARCHMETH1 param as
when I took the backup from primary server it was D:\Db2logs and when
I restored it on standby there was no D drive on that machine so I
fixed it first as the servers must have the same architecture.
The other thing I missed was not removing the firewall set up.
It was ON when I first attempted.
and then it was success and both the servers were in peer state.
I heaved a sigh of relief after so many tribulations I finally got it
work.
Now was the time for implementing this on two Windows server 2003
machine.
Unfortunately I am still getting RC 99 problem on primary server
after starting the HADR on db as standby successfully.
I have taken care of the two problems that I faced earlier but this
time its not working.
I wish GUI had worked , Life would have been easier.
Any help would be appreciated.
Thanks,
Rakesh
DB2 DBA