1

Yup - my db02 server lost a disk, and the array itself is corrupt (RAID5, yet corrupt data.) I was able to get the OS back online, but the corruption is severe. I have zero documentation about the mysql configuration, other than a copy of the /etc/my.cnf file from the failed host. (one small joy in this.)

1) How would I determine if it was in a m-s or m-m configuration?

2) Since I am lucky enough to have the /etc/my.cnf file, is there a clean way of restoring from the master?

3) Anyone have any other ideas as to what I should look for, in order to make sure what else needs to be put back in place? Backups are managed via scripting, from yet another system.

0

For M-M, both need to be masters and both need to be slaves. Those need

log_slave_updates
log_bin

These should show something, even if not actively replicating at the moment:

SHOW SLAVE STATUS;
SHOW MASTER STATUS;

If you can't run things on the broken machine, then I suggest you do

  • Clone the good machine's disk to the bad
  • Change my.cnf to have a different server_id
  • Turn on the two flags above
  • CHANGE MASTER ... on both -- to establish being a Slave, and provide the pwd for the replication 'user'.

Now you will have Master-Master. (There is not a lot of differences between M-S and M-M, so you may as well go for M-M.)

| improve this answer | |
  • If the my.cnf does not contain "log_slave_updates", then they are not in a M-M setup, correct? – Bee Kay Jan 24 '17 at 21:37
  • @RickJames would these 2 commands (show master and slave status) reveal the setup if they are run on the first, intact server? – ypercubeᵀᴹ Jan 24 '17 at 21:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.