"the master has purged binary logs containing GTIDs that the slave requires"

[abh]

When adding a new replica I now get this error:

"Got fatal error 1236 from master when reading data from binary log:
'The slave is connecting using CHANGE MASTER TO MASTER_AUTO_POSITION = 1,
but the master has purged binary logs containing GTIDs that the slave requires.'"

How does this happen?

[abh]

The active master is db-1 and the new (starting from blank) replica is db-0.

I just noticed that db-0 didn't start with a snapshot of db-1, as it has worked otherwise when a new replica entered the system. db-0 just has a default "blank" database.

[AMecea]

Hi @abh,

Is this issue related to #412 and #326?

Thank you for your contribution!

[abh]

Yes, I think this issue is a duplicate of those. Thank you!

(That explains why it came up now after upgrading to 0.3.x since I'd seen the system automatically take care of this scenario before).