We write their own by directing a API Java program for single thread insert, insert the article 40 w data, each insert 2000 returns a time consuming, during crash 3 subdivision of 3 primary (different fragmentation), still continue to insert, finally found less total data 6 w,
Suppose 1 because I use a kill 9 mongod pid makes its data,

CodePudding user response:

But I opened the journal log? Should not be lost data?

CodePudding user response:

We can see the Java log

CodePudding user response:

refer to the second floor rucypli response:

can see the Java log

What is the log? Spring data mongo's log?