[MDEV-16146] MariaDB slave stops with incompatible heartbeat Created: 2018-05-11 Updated: 2021-05-04 Resolved: 2021-05-01 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | Replication |
| Affects Version/s: | 5.5.56 |
| Fix Version/s: | 10.2.38, 10.3.29, 10.4.19, 10.5.10 |
| Type: | Bug | Priority: | Major |
| Reporter: | Diptesh Patel | Assignee: | Sujatha Sivakumar (Inactive) |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | replication | ||
| Environment: |
Centos 7.4.1708 |
||
| Attachments: |
|
||||
| Issue Links: |
|
||||
| Description |
|
MariaDB slave stops with following errors. 180511 11:07:58 [ERROR] Slave I/O: Unexpected master's heartbeat data: heartbeat is not compatible with local info;the event's data: log_file_name mysql-bin.000009 log_pos 1054262041, Error_code: 1623 I am using Master-Master and Master-Slave Replication using MariaDB servers. I am replicating all Databases on Master-Master Replication but there is only one databse replicating on slave servers.Slave servers ignores the query from another database which is not replicating there. All other Binlogs which from not replicating skips by slave servers. We have heavy traffic on Master server. Slaves only for select purpose. Master databse server's system load goes upto 10%. NOTE: All the slaves are connected using Gigabit Ethernet(Private Network). We are using 24 cores CPU and 32GB RAM on master and 16 cores CPU and 24GB RAM on each slave. I am using following version of MariaDB server on my 3 servers Operating system Following is diagram of replication |
| Comments |
| Comment by Sujatha Sivakumar (Inactive) [ 2021-03-19 ] |
|
Hello Andrei, Can you please review the fix for Patch: https://github.com/MariaDB/server/commit/7c2ea6b963f8ee3894e25e1f479f9d09b5b766b9 Buildbot: http://buildbot.askmonty.org/buildbot/grid?category=main&branch=bb-10.5-sujatha At present patch is implemented for 10.5 it needs to be implemented for all GA versions. Thank you. |
| Comment by Andrei Elkin [ 2021-04-22 ] |
|
Waiting for update to complete review. |
| Comment by Sujatha Sivakumar (Inactive) [ 2021-04-26 ] |
|
Hello Andrei, Thank you for the review comments. I have addressed them in the following patch. https://github.com/MariaDB/server/commit/724862441d5e1a889a974b5fdb32fd27f409cf6f BuildBot: http://buildbot.askmonty.org/buildbot/grid?category=main&branch=bb-10.5-sujatha Please review them. Thank you. |
| Comment by Andrei Elkin [ 2021-04-26 ] |
|
Comments were made. Waiting for a new patch. |
| Comment by Sujatha Sivakumar (Inactive) [ 2021-04-28 ] |
|
Hello Andrei, Please find the new patch which addresses your latest review comments.
https://github.com/MariaDB/server/commit/f2b685c491e0ebe12ac7d5475f33ef65146ca602 Thank you. |
| Comment by Andrei Elkin [ 2021-04-29 ] |
|
Thanks! The patch is free to go. Github suggests a small change to the commit message. |
| Comment by Sujatha Sivakumar (Inactive) [ 2021-05-01 ] |
|
Fix is implemented in 10.2.38. No merge conflicts were observed upto 10.4. Fix was cherry-picked to higher versions and tested. 10.3 patch: https://github.com/MariaDB/server/commit/4be2c8baae5cdff607e48fbfb3f76be87aa9fdbd In 10.5, the changes in 'log_event.cc' have to be moved to 'log_event_server.cc'. 10.5 patch: https://github.com/MariaDB/server/commit/1d9dbd895e3f358979df0b9c5f6f75a2a4c35c12 |