MongoDB chunk migration steps in changelog collection - mongodb

Question

Can someone clarify all the steps for "moveChunk.from" and "moveChunk.to". I want to know at these steps, which operations are performed (I guess the value of the steps represents the time ms, it took for the step). This will help me to derive any slowest step that is occurring during chunk migration.
{
"_id" : "bdvlpabhishekk-2013-07-20T17:46:28-51eaccf40c5c5c12e0e451d5",
"server" : "bdvlpabhishekk",
"clientAddr" : "127.0.0.1:50933",
"time" : ISODate("2013-07-20T17:46:28.589Z"),
"what" : "moveChunk.from",
"ns" : "test.test",
"details" : {
"min" : {
"key1" : 151110
},
"max" : {
"key1" : 171315
},
"step1 of 6" : 0,
"step2 of 6" : 1,
"step3 of 6" : 60,
"step4 of 6" : 2067,
"step5 of 6" : 7,
"step6 of 6" : 0
}
}
{
"_id" : "bdvlpabhishekk-2013-07-20T17:46:31-51eaccf7d6a98a5663942b06",
"server" : "bdvlpabhishekk",
"clientAddr" : ":27017",
"time" : ISODate("2013-07-20T17:46:31.671Z"),
"what" : "moveChunk.to",
"ns" : "test.test",
"details" : {
"min" : {
"key1" : 171315
},
"max" : {
"key1" : 192199
},
"step1 of 5" : 0,
"step2 of 5" : 0,
"step3 of 5" : 1712,
"step4 of 5" : 0,
"step5 of 5" : 344
}
}

Answer 1

All these steps are explained in the "M202: MONGODB ADVANCED DEPLOYMENT AND OPERATIONS" course, which is available online for free (I can't post this link here because of the limitation on stackoverflow on a number of posted URLs, just try to find the course in google)
Related videos from this course are: Anatomy of a migration overview and Anatomy of a migration deep dive.
The explanation follows.
All time values are in milliseconds.
Let's say F is a "moveChunk.from" and T is a "moveChunk.to". The steps are F1..F6 and T1..T5.The steps are performed sequentially F1, F2, F3, F4: {T1, T2, T3, T4, T5}, F5, F6. Step F4 includes {T1..T5} and timing of F4 is the sum of T1..T5 (but there is no exact match).
F1 - mongos sends 'moveChunk' command to F (primary of the shard to migrate from)
F2 - sanity checking of the command
F3 - F sends the command to T (read this chunk from me)
F4,T1..T3 - transfer starts, T performs sanity checking, index checking etc
F4,T4 - catch up on subsequent operations (if there were inserts into the chunk while the transfer was going, send updates from F to T)
F4,T5 - steady state (changes are written to the primary log)
F5 - about to commit to config server changes about the new chunk location (critical secition)
F6 - clean up

Answer 2

All chunk migrations use the following procedure:
The balancer process sends the moveChunk command to the source shard.
The source starts the move with an internal moveChunk command. During the migration process, operations to the chunk route to the source shard. The source shard is responsible for incoming write operations for the chunk.
The destination shard begins requesting documents in the chunk and starts receiving copies of the data.
After receiving the final document in the chunk, the destination shard starts a synchronization process to ensure that it has the changes to the migrated documents that occurred during the migration.
When fully synchronized, the destination shard connects to the config database and updates the cluster metadata with the new location for the chunk.
After the destination shard completes the update of the metadata, and once there are no open cursors on the chunk, the source shard deletes its copy of the documents.
Taken from this.