mongoDB - replica section

[paf1]

Hello,
would you add new section to mongo.pm monitoring ?

add replica section

replica can have these states
PRIMARY / SECONDARY / STARTUP2 / RECOVERY / ......
https://docs.mongodb.com/manual/reference/replica-states/

@ show replica count of nodes ( 2+1 in our case, but can be more )
@ show replica state for current node

_HA0:STARTUP2> rs.status()
{
"set" : "HA0",
"date" : ISODate("2017-11-23T12:41:45.249Z"),
"myState" : 5,
"term" : NumberLong(23),
"syncingTo" : "mongo1:27017",
"heartbeatIntervalMillis" : NumberLong(2000),
"optimes" : {
"lastCommittedOpTime" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"appliedOpTime" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"durableOpTime" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
}
},
"members" : [
{
"_id" : 1,
"name" : "mongo1:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 5148,
"optime" : {
"ts" : Timestamp(1511440905, 132),
"t" : NumberLong(23)
},
"optimeDurable" : {
"ts" : Timestamp(1511440905, 95),
"t" : NumberLong(23)
},
"optimeDate" : ISODate("2017-11-23T12:41:45Z"),
"optimeDurableDate" : ISODate("2017-11-23T12:41:45Z"),
"lastHeartbeat" : ISODate("2017-11-23T12:41:45.124Z"),
"lastHeartbeatRecv" : ISODate("2017-11-23T12:41:45.125Z"),
"pingMs" : NumberLong(0),
"electionTime" : Timestamp(1511211655, 1),
"electionDate" : ISODate("2017-11-20T21:00:55Z"),
"configVersion" : 565069
},
{
"_id" : 2,
"name" : "mongo3:27017",
"health" : 1,
"state" : 5,
"stateStr" : "STARTUP2",
"uptime" : 5150,
"optime" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"optimeDate" : ISODate("1970-01-01T00:00:00Z"),
"syncingTo" : "mongo1:27017",
"configVersion" : 565069,
"self" : true
},
{
"id" : 3,
"name" : "mongo2:27017",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 5148,
"lastHeartbeat" : ISODate("2017-11-23T12:41:44.121Z"),
"lastHeartbeatRecv" : ISODate("2017-11-23T12:41:41.449Z"),
"pingMs" : NumberLong(0),
"configVersion" : 565069
}
],
"ok" : 1
}

• when replication is running for the first time we get these outputs, but it's not possible to get db.stat() at this time ( exception is needed ) :

HA0:PRIMARY> db.printReplicationInfo()
configured oplog size: 1024MB
log length start to end:8041secs (2.23hrs)
oplog first event time: Thu Nov 23 2017 11:32:46 GMT+0100 (CET)
oplog last event time: Thu Nov 23 2017 13:46:47 GMT+0100 (CET)
now: Thu Nov 23 2017 13:46:47 GMT+0100 (CET)
HA0:PRIMARY>

HA0:STARTUP2> db.printReplicationInfo()
this is a slave, printing slave replication info.
source: mongo3:27017
syncedTo: Thu Jan 01 1970 01:00:00 GMT+0100 (CET)
1511441113 secs (419844.75 hrs) behind the primary
HA0:STARTUP2>

When finished, then replica STARTUP2 switch to SECONDARY state and the output is the same like PRIMARY ( with different values )

@ show OPLOG size
@ show "log length start to end" value sec / hours -very usefull for maintenance window. If overtimed, then full replica needed, not diff. only.
@ show OPLOG first/last timestamp too

OPLOG - not needed for ARBITER ( no data, just only for cluster voting )

More info for ARBITER node when replication finished ( a few days later )

regs.
Pavel