[patch 0/6] Bind patches for telco tree (take 2)

(sent to @mysql instead of @lists.mysql by mistake, reposting for benefit of all)

Hi all,

I've updated the Master_Bind patches based on feedback.

New things here:
- Master_Bind in SHOW SLAVE STATUS moved to the end
Mark, Chip, Guilhem, Sinisa: you guys have more real world

	experinece with these changes and the resulting potential chaos.
	Is this a much better solution? If anyone can test some of the
	random scripts around, would be great. Do you forsee any other
	problems?

• All the test results updated due to the creation/move of the Master_Bind column
• mysql-test-run.pl is passing with stmt and row based rpl. Possibly some other test will "fail" because of the new column that isn't run in my build... hoping pushbuild will pick this up (is trivial fixes).
• Fixed a (at least potential) bug when sending the Master_Bind column out for SHOW SLAVE STATUS.. turns out previously i was sending twice.

My primary concerns:
- Client side bind: does this break any binary compatibility?
  I'm not familiar enough with the client library to judge this.

• Kay's patch for replication delay will need some merging love with this. Hopefully the test suite will help in picking up if this is merged wrongly..
• behaviour when you set Master_Bind to something silly (e.g. '42').

Where this is heading:
- Currently only to the mysql-5.1-telco tree (MySQL Cluster Carrier Grade
  Edition).

• Don't know of any plans for it to hit the "main" tree apart from when we merge CGE in (which could be 5.2 or later).
• Others are welcome to convince the right people that this should go into something else as well. Arguably, this is quite a useful feature.

Feedback much appreciated.

--
Stewart Smith

-- 
MySQL Internals Mailing List
For list archives: 
http://lists.mysql.com/internals
To unsubscribe:    
http://lists.mysql.com/internals?unsub=lists@pantek.com