Bug #91163 | MGR member cannnot be joined, which has more executed transactions than present | ||
---|---|---|---|
Submitted: | 7 Jun 2018 1:10 | Modified: | 22 Jun 2018 15:31 |
Reporter: | Yue Wang Wang | Email Updates: | |
Status: | Can't repeat | Impact on me: | |
Category: | MySQL Server: Group Replication | Severity: | S1 (Critical) |
Version: | 8.0.11 | OS: | CentOS (7.4) |
Assigned to: | MySQL Verification Team | CPU Architecture: | Any |
[7 Jun 2018 1:10]
Yue Wang Wang
[7 Jun 2018 1:13]
Yue Wang Wang
My username as well as my e-mail 'wanyue74@126.com' is hacked. Contact me with 'wanyue76@126.com' if necessarily.
[7 Jun 2018 2:17]
Yue Wang Wang
There is an alternative way to build an MGR within version 8.0.11. I copy all the datafile and archfile to a new cluser member, without initialize a new DB. Thus, there is no more executed transactions than those present in the cluster, which bypass the problem.
[22 Jun 2018 15:31]
MySQL Verification Team
Hi, I'm not sure what exactly are you doing but I can't reproduce this the way you described. Are you upgrading the existing setup or what? Setting up cluster with 8.0.11 worked ok for me, don't see what your issue is. Can you write step by step how are you reproducing this issue? With all the config files attached please. kind regards Bogdan