sad.PNG
[MXS-2255] Match clause matching incorrect schema Created: 2019-01-10 Updated: 2020-12-08 Resolved: 2019-01-29 |
|
| Status: | Closed |
| Project: | MariaDB MaxScale |
| Component/s: | Filter |
| Affects Version/s: | 2.3.2 |
| Fix Version/s: | 2.3.3 |
| Type: | Bug | Priority: | Major |
| Reporter: | Adam Erickson (Inactive) | Assignee: | markus makela |
| Resolution: | Fixed | Votes: | 6 |
| Labels: | None | ||
| Environment: |
CentOS 7, MariaDB 10.1 Master, MariaDB 10.3 Slave |
||
| Attachments: |
sad.PNG
|
| Description |
|
Using MX 2.3 for binlog replication & filtering. Where primary master (MDB 10.1) DB has multiple schemas and MX match defined to replicate only one of them to a downstream (MDB 10.3 slave.) |
| Comments |
| Comment by markus makela [ 2019-01-28 ] |
|
Please try this with the 2.3.3 release of MaxScale. |
| Comment by Adam Erickson (Inactive) [ 2019-01-29 ] |
|
The slave instance connected to maxscale is not requesting a binlog file / position which maxscale reports as missing or unavailable. I've attached a SS from the client with the exact failure messages. If a binlog entry is skipped due to the binlogfile would that cause such an error if the slave from MX requested a position that MX itself filtered out already? |
| Comment by Adam Erickson (Inactive) [ 2019-01-29 ] |
|
I'm attaching the output of mysqlbinlog -vv for the relevant schema from the original master to help diagnose the binlog events. The slave instance was somehow given an invalid or corrupt log position to read from MaxScale with. |