When upstream has gap in GTID sets, switching enable-gtid from OFF to ON will fail the synchronization

[glkappe]

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. What did you do? If possible, provide a recipe for reproducing the error.
   dm worker error log：
   
   mysql binlog info：
   

query-status 03 syncerbinloggtid info：
"unresolvedDDLLockID": "", "sync": { "totalEvents": "0", "totalTps": "0", "recentTps": "0", "masterBinlog": "(mysql-bin.074545, 417140060)", "masterBinlogGtid": "148da50a-935f-11ea-9456-0ecb2cc2c6f2:1-26025771647,5ba91a4e-2b16-11eb-8406-0af8ee284d60:1-43991284520,903b8ce1-9357-11ea-a7af-0eac1754e2d4:1-7,f5cdfc8b-2b15-11eb-ba43-0a342fdb5c6c:1-21093139382", "syncerBinlog": "(mysql-bin.074534, 999293514)", "syncerBinlogGtid": "148da50a-935f-11ea-9456-0ecb2cc2c6f2:23174871189-26025771647,5ba91a4e-2b16-11eb-8406-0af8ee284d60:1-43991284520,f5cdfc8b-2b15-11eb-ba43-0a342fdb5c6c:1-21082962932",
mysql binlog 74534 :

mysql master status & purged info：

2. What did you expect to see?
   replication is stop, and error is 'gtid is purged'

3. What did you see instead?
   task is running

4. Versions of the cluster

   • DM version (run dmctl -V or dm-worker -V or dm-master -V):
     v2.0.4

     (paste DM version here, and you must ensure versions of dmctl, DM-worker and DM-master are same)
     

   • Upstream MySQL/MariaDB server version:
     5.7-percona

     (paste upstream MySQL/MariaDB server version here)
     

   • Downstream TiDB cluster version (execute SELECT tidb_version(); in a MySQL client):
     v4.0.10

     (paste TiDB cluster version here)
     

   • How did you deploy DM: DM-Ansible or manually?
     tiup

     (leave DM-Ansible or manually here)
     

   • Other interesting information (system version, hardware config, etc):

     centos 7

5. current status of DM cluster (execute query-status in dmctl)
   task 3 is paused

6. Operation logs

   • Please upload dm-worker.log for every DM-worker instance if possible
   • Please upload dm-master.log if possible
   • Other interesting logs
   • Output of dmctl's commands with problems

7. Configuration of the cluster and the task

   • dm-worker.toml for every DM-worker instance if possible
   • dm-master.toml for DM-master if possible
   • task config, like task.yaml if possible
   • inventory.ini if deployed by DM-Ansible
     add enable-gtid: true to 3source file

8. Screenshot/exported-PDF of Grafana dashboard or metrics' graph in Prometheus for DM if possible
   no

[lichunzhu]

This problem is caused by previous gtid sets may not include all purged gtid sets. MySQL reported an error when we request a purged GTID.

[lance6716]

maybe we could use select @@global.gtid_executed; to get a no gap GTID, will test later