08-20-2013 12:10 AM - edited 03-19-2019 07:09 AM
Hi All
We have recently ran into CUPS publisher reinstall and restore DB from subscriber due to crashed HDD.
I have followed instruction from doc for restore procedure and pub is restore successful.
In terms of dbreplication, we ran into below:
Pub:
*************************************************************************************************
Cluster Replication State: Replication status command started at: 2013-08-20-15-36
There no active nodes in the replication network
DB Version: ccm8_6_3_10000_20
Number of replicated tables: 0
Cluster Detailed View from PUB (2 Servers):
PING REPLICATION REPL. DBver& REPL. REPLICATION SETUP
SERVER-NAME IP ADDRESS (msec) RPC? STATUS QUEUE TABLES LOOP? (RTMT) & details
----------- ------------ ------ ---- ----------- ----- ------- ----- -----------------
POL-CUPS1 10.10.35.37 0.027 Yes Off-Line N/A match N/A (0) PUB Setup Completed
PIB-CUPS1 10.2.35.37 1.02 Yes Off-Line N/A 269 No (3) Not Setup
***************************************************************************************************
Sub:
*************************************************************************************************
DB and Replication Services: ALL RUNNING
Cluster Replication State: Only available on the PUB
DB Version: ccm8_6_3_10000_20
Number of replicated tables: 269
Cluster Detailed View from SUB (2 Servers):
PING REPLICATION REPL. DBver& REPL. REPLICATION SETUP
SERVER-NAME IP ADDRESS (msec) RPC? STATUS QUEUE TABLES LOOP? (RTMT)
----------- ------------ ------ ---- ----------- ----- ------- ----- -----------------
POL-CUPS1 10.10.35.37 1.01 Yes Active-Dropped 1483407 0 No (0)
PIB-CUPS1 10.2.35.37 0.024 Yes Connected 0 match Yes (3)
*************************************************************************************************
2 question stands:
1. whether we use correct procedure to restore Pub, refere to doc above.
2. if the restore procedure is correct, how do you rectify dbreplication here.
Regards
Fei
08-20-2013 12:45 AM
Hi Fei,
Judging by the document you referred to, it looks like you did the following steps:
1> Rebuilt the Pub with the same hostname, IP address, Subnet, CUPS version, etc.
2> Did a DRS Restore using the most recent backup file that had been taken prior to the Publisher going down.
3> Landed up with the dbreplication broken.
Also, check to make sure that the cmhosts, rhosts and sqlhosts tables have the correct entries.
Question: Did you do a cluster reboot after the restore? If not, please go ahead and do so.
If the above is correct, then you can proceed with the following steps (in order and one at a time) to correct the issue:
1> "utils dbreplication stop" from Subscriber
2> "utils dbreplication stop" from Publisher
3> "utils dbreplication dropadmindb" from Publisher
4> "utils dbreplication dropadmindb" from Subscriber
5> "utils dbreplication reset all" from Publisher
6> Monitor the progress through "utils dbreplication runtimestate" from Publisher to see how far along the replication setup has gotten.
From the outputs, it seems like the syscdr DB is not getting formed correctly which is why the Pub shows offline. Once the Pub does a cdr define and realizes the db template, the Sub will do a cdr define, realize the template and do a cdr sync. Also, please refer to the following docs:
https://supportforums.cisco.com/docs/DOC-13672
Hope this helps.
--
Regards,
Harmit.
08-20-2013 01:07 AM
Hi Harmit
You are correct, we rebuild CUPS Pub with correct all info but in terms of restore, the most recent backup files are 3 month old.
As you can see, only rhosts and sqlhosts show error - does not match publisher.
CUP hosts have equivalent host files.
We have not schedule whole cluster reboot after restore but we do have Pub reboot afterwards.
If whole cluster reboot is suggested, what order should we take? reboot Pub then Sub?
Regards
Fei
08-20-2013 08:08 PM
Hi Harmit
the DB replication still incorrect after reboot cluster and issue commands as you suggested. I have engaged TAC, TAC advised the restore process we took does not realy work.
most likely we need to rebuild CUPS again and TAC will assist on restore procedure.
but really appreciated the reading provied, it helps a lot.
Regards
Fei
08-20-2013 09:56 PM
Hi Fei,
Thank you for the update, just saw both your responses. I'm surprised with what TAC had to say about the DRS restore process. Given the fact that you said you have a DRS backup taken from the Pub (quite sometime ago), that backup should be and can be restored onto the Pub. It's possible that I may have misunderstood and they saw something which suggested it isn't possible. Having access and taking a look first hand is always more reassuring when suggesting next steps for dbreplication :-) I'm glad my explanation and links came in handy for you. Good luck with the rebuild.
--
Regards,
Harmit.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: