| 1 |
I was planning on updating a client from Mysql 3.23 to Mysql 4.0, but |
| 2 |
since 4.1 has stabilized I figured I'd give it a shot. And I'd also be |
| 3 |
able to move another client off their crappy db server on the new shared |
| 4 |
server if it had 4.1. However my setup has a few quirks. |
| 5 |
|
| 6 |
The 3.23 install is acting for the replication master for the 4.0 |
| 7 |
install on the new server. The client is testing with 4.0 and everything |
| 8 |
looks good. Ideally I'd like to stop replication, dump the db, upgrade |
| 9 |
to 4.1, import the db, and start replication again. 4.1 can replicate |
| 10 |
from 3.23, but I'm unsure if it can pick up where 4.0 left off. Has |
| 11 |
anyone tried this? It's not the end of the world if it explodes in my |
| 12 |
face, but I would keep up with my sleep by skipping another late night |
| 13 |
maintenance to do a fresh dump from the master to restart replication |
| 14 |
correctly. |
| 15 |
|
| 16 |
The other question is about libs. I'm looking at a project that will |
| 17 |
likely use Mysql 5.0 on the server side. At what point do I need to |
| 18 |
update the client libs... or can I just use Mysql 4.0 libs in PHP and |
| 19 |
whatnot till Redhat updates their builds? Some sort of checklist, Google |
| 20 |
has been not helped so far, explaining when a client build would need to |
| 21 |
be updated would be keen. Again not the end of the world, but I'd rather |
| 22 |
not be building a bunch of custom packages if I can help it. |
| 23 |
|
| 24 |
kashani |
| 25 |
-- |
| 26 |
gentoo-user@g.o mailing list |
Archives