Hey Dayle,

2.3.1 you have 'In the Server Removing_Cloud_Providers area, uncheck Database Synchronization.'

I guess 'Removing_Cloud_Providers' was a typo, it should read as follows:

'In the Server Roles area, uncheck Database Synchronization.'

Other than that everything else looks good to me.

Hi Dayle,

We are missing # vmdb before # bin/rake evm:dbsync:uninstall 1.3.1 4 and 2.3.1 4

Also I have just noticed we are missing some more steps for 5.5 - 5.7 replication. We also need to run 'rake evm:db:check_schema' and fix the table columns that do not match, using rails r tools/fix_column_ordering.rb <table name>.

Now I am willing to bet this is not part of the replication document (if we have that yet) as its only required after migrating from 5.5. We can probably just add the following after step 6 in 2.4.1:

7. rake evm:db:check_schema

^ This will give you a list of tables most likely these three (computer_systems, event_streams and security_contexts)

8. rails r tools/fix_column_ordering.rb <table_name> (replace <table_name> with table from list, this will rearrange the table columns to match the new schema)

Cheers,
Luke

Hi Dayle,

After further investigation it seems that the rails r tools/fix_column_ordering.rb task mentioned above needs to be run after postgres upgrade so after 2.4.1 11 should work.

Cheers,
Luke

Hi Luke,

Thanks so much for the comments, and for testing this so thoroughly.
Apologies also for my slow reply -- I've been sidetracked on another doc until now.

Changes made:

From Comment 3 - fixed typo.

From Comments 4 & 5:
- Added # vmdb before # bin/rake evm:dbsync:uninstall (2 places)

-  You're right -- we don't have this documented in any replication setup procedures. I think since it's only required right aftre migrating from 5.5, we're safe to just keep it in this guide. 

So, only for the 5.5 -> 5.7 migration, I've added the two steps for checking & rearranging table columns, but after step 11 ( Enable and start PostgreSQL)

Other edits:
--> I spotted a sentence in the 5.5 -> 5.7 section (step 5 in 2.4.1.) reading: "Run the following to migrate everything in the database from 5.6 to work with the latest 5.7 configuration:" I believe this was meant to be "everything in the database from 5.5", so I've updated this. Please let me know if that's wrong and should be changed back. :)

Cheers,
Dayle

Hey Dayle,

So the edits you have made all look good to me, the combining of 1.3. 'Preparing the Appliances for Migration' works fine, I suggest doing the same thing to 2.3 for consistency throughout the doc. 

I have since noticed a few other points, 1.3 step 1 b says:

In the Server Roles area, unchec ization* role has stopped by going to Settings → Configuration → Diagnostics → Region. 

Another typo perhaps? Although this section seems a little jumbled too. I think it should read like so:

A. Using the CloudForms user interface, navigate to the specific server’s page under Settings → Configuration → settings → server.

B. Within the Server tab under Server Control, uncheck Database Synchronization Role.

C. Click Save.

D. Verify that the Database Synchronization role has stopped by going to Settings → Configuration → Diagnostics → Region.
    
E. Click the Replication tab. The number for Current Backlog should be increasing. 



The section 2.3.1 looks good just change 1a and b to the following:

A. Using the CloudForms user interface, navigate to the specific server’s page under Configure → Configuration → settings → server. 

B. Within the Server tab under Server Control, uncheck Database Synchronization Role.



The only other thing I would suggest changing is 'a table' in 2.4.1 step 13 with 'each table' or similar as there is more than likely to be multiple tables to rearrange.

We should be pretty good after those last few edits.

Thanks,
Luke

Hi Luke,

Thanks for the feedback, and sorry for the headaches! I think the mouse I was using was selecting and somehow deleting things (so weird - I noticed it doing it once or twice but thought I caught all the unintentional changes).

> Hey Dayle,
> 
> So the edits you have made all look good to me, the combining of 1.3.
> 'Preparing the Appliances for Migration' works fine, I suggest doing the
> same thing to 2.3 for consistency throughout the doc. 

All done.

> 
> I have since noticed a few other points, 1.3 step 1 b says:
> 
> In the Server Roles area, unchec ization* role has stopped by going to
> Settings → Configuration → Diagnostics → Region. 
> 
> Another typo perhaps? Although this section seems a little jumbled too. I
> think it should read like so:
> 
> A. Using the CloudForms user interface, navigate to the specific server’s
> page under Settings → Configuration → settings → server.
> 
> B. Within the Server tab under Server Control, uncheck Database
> Synchronization Role.
> 
> C. Click Save.
> 
> D. Verify that the Database Synchronization role has stopped by going to
> Settings → Configuration → Diagnostics → Region.
>     
> E. Click the Replication tab. The number for Current Backlog should be
> increasing. 
> 

Fixed.


> The section 2.3.1 looks good just change 1a and b to the following:
> 
> A. Using the CloudForms user interface, navigate to the specific server’s
> page under Configure → Configuration → settings → server. 
> 
> B. Within the Server tab under Server Control, uncheck Database
> Synchronization Role.
> 

Done. 

> 
> The only other thing I would suggest changing is 'a table' in 2.4.1 step 13
> with 'each table' or similar as there is more than likely to be multiple
> tables to rearrange.

Agreed, 'each' is more specific. Thanks Luke!

Dayle,

Not seeing the update on 2.3.1 from:

 
> a. Using the CloudForms user interface, navigate to the specific server’s
> page under Configure → Configuration → settings → server. 

still missing '→ settings → server'



Also 2.4.1 step 8 is missing the new edits seen in 1.4.1 step 8, we may as well combine the edits and remove the line numbers to simplify. If we just change section 2.4.1 step 8 to as follows that should do it:

-checkpoint_segments = 15   # MIQ Value;
-#checkpoint_segments = 3    # in logfile segments, min 1, 16MB each
+#max_wal_size = 1GB
+#min_wal_size = 80MB

-#shared_preload_libraries = ''        # (change requires restart)        
+shared_preload_libraries = 'pglogical,repmgr_funcs'   # MIQ Value (change requires restart)

+wal_log_hints = on

-#wal_level = minimal            # minimal, archive, or hot_standby  
+wal_level = 'logical' # MIQ Value (pglogical)

-#max_wal_senders = 0        # max number of walsender processes   
+max_wal_senders = 10        # MIQ Value (pglogical) max number of walsender processes

+max_worker_processes = 10  # MIQ Value (pglogical)    
+max_replication_slots = 10 # MIQ Value (pglogical)

Thanks again,
Luke

Thanks again for your reviews, Luke. These issues have now been fixed.

Looks good to me; moving to VERIFIED.

The section on migrating non-VDMB appliances, and other changes from this bug, are now published in:

https://access.redhat.com/documentation/en/red-hat-cloudforms/4.2/single/migrating-to-red-hat-cloudforms-42/