HDR related problem

[antzzz]

I have HDR running on an IDS 7.31 on SCO Unix 5.0.5. I took a Level 0 backup of my Primary, did some transactions which were replicated to my Secondary. I then shutdown my Primary using onmode -ky. I restored my previous Level 0 backup on the Primary, started replication using onmode -d primary <sec_name> and logs were sent from Secondary to Primary.

I finally get the following status when I execute onstat -:

On-line(Prim)(CKPT REQ)
Blocked: CKPT

I tried to wait for a couple of hours, but unfortunately, it stays that way forever.

What seems to be the problem?

Pls help.

Thanks!

 

[vpshriyan]

Hi,

You said, you have restored your previous level 0 backup on the primary, after you did some transactions; in a period when the HDR was active and functional. When you did this you have effectively rolled your primary server to a backdated status, and incidentally the secondary server is still in a latest status. Obviously this has caused the break in hookup.

To solve this, I'm afraid, you have to re-implement the the HDR.

Regards,
Shriyan

&quot;The learning and knowledge that we have, is, at the most, but little compared with that of which we are ignorant.&quot;

 

[antzzz]

Thanks for the reply.

I was trying to test whether the transactions will be sent from the Secondary, which happened automatically by viewing the log of the Secondary. Unfortunately, the Primary stayed in a blocked Checkpoint state.

Actually, I, sort of, realized what the problem was because my log backups were still running as scheduled during the DBSpace restore and it reported an I/O chunk error in the log. This may have affected the status of my DBSpaces which may have caused the &quot;eternal&quot; hanging.

To try to solve the problem, I stopped the scheduled log backups, and did a restore again. This time it worked.

P.S. During restoration, I noticed an Invalid Dbspacetemp message in the log but afterwards, my temp dbspace seemed ok.