joiner failed to reconnect cluster

[tomatolog]

Bug Description:

if one node of the cluster stops as usual then new table got added into cluster joiner failed to reconnect cluster due to missed table

node1

start daemon
mysql>create cluster test;

node2

start daemon
mysql>join cluster test at 'node1:9312';

check that cluster is ok at both nodes then stop node2

node2

stop daemon

node1

create table t (tmp int);
alter cluster test add t;

node2

start daemon

after daemon started node2 show that cluster at that node got destroyed. That is why node2 joins cluster and Galera issue IST and send TNX stream to node2 and one of the command in that stream is add local table t but the table t is missed at the node2 as it does not participate in cluster wide table t send during alter cluster test add t. That is why in this case node2 can not join via IST.

After that node2 can join the cluster from scratch via full SST.

Manticore Search Version:

any

Operating System Version:

linux

Have you tried the latest development version?

None

Internal Checklist:

^{To be completed by the assignee. Check off tasks that have been completed or are not applicable.}

• Implementation completed
• Tests developed
• Documentation updated
• Documentation reviewed
• Changelog updated