Why mongodb sharded cluster does not have any chunk

Ops and Admin Sharding
,
1

I had a mongodb instance which was up as a replicaset and was a single node in the replicaset. Now I have added to other replicasets each with one node and connected them to clutser through mongos using addShard command, enabled sharding on a db using enableSharding command and sharded a collection using shardCollection . However when I run sh.status the output says that I have no chunk and also my data did not get balanced. Is there anything wrong?
mongodb version: 6.0.2
mongodb sh.status() results:

[direct: mongos]> sh.status()
shardingVersion
{
  _id: 1,
  minCompatibleVersion: 5,
  currentVersion: 6,
  clusterId: ObjectId("64d62ba7e9d199311e7be6b5")
}
---
shards
[
  {
    _id: 'myshard1',
    host: 'myshard1/node3:27018',
    state: 1,
    topologyTime: Timestamp(6, 1691758010)
  },
  {
    _id: 'myshard2',
    host: 'myshard2/node2:27018',
    state: 1,
    topologyTime: Timestamp(206, 1691773056)
  },
  {
    _id: 'myshard3',
    host: 'myshard3/node1:27018',
    state: 1,
    topologyTime: Timestamp(51, 1691773082)
  }
]
---
active mongoses
[ { '6.0.4': 1 } ]
---
autosplit
{ 'Currently enabled': 'yes' }
---
balancer
{
  'Currently enabled': 'yes',
  'Currently running': 'no',
  'Failed balancer rounds in last 5 attempts': 0,
  'Migration Results for the last 24 hours': {
    '3': "Failed with error 'aborted', from myshard3 to myshard1",
    '30': "Failed with error 'aborted', from myshard1 to myshard2",
    '157': "Failed with error 'aborted', from myshard1 to myshard3",
    '2669': 'Success'
  }
}
---
databases
[
  {
    database: {
      _id: 'mydb',
      primary: 'myshard1',
      partitioned: false,
      version: {
        uuid: UUID("2dcac9aa-595c-4515-8033-27298507f20c"),
        timestamp: Timestamp(1, 1691758436),
        lastMod: 1
      }
    },
    collections: {
      'mydb.user': {
        shardKey: { _id: 1 },
        unique: false,
        balancing: true,
        chunks: [],
        tags: []
      }
    }
  },
  {
    database: { _id: 'config', primary: 'config', partitioned: true },
    collections: {
      'config.system.sessions': {
        shardKey: { _id: 1 },
        unique: false,
        balancing: true,
        chunks: [],
        tags: []
      }
    }
  }
]

getShardDistribution on sharded collection:

[direct: mongos]> db.user.getShardDistribution()
Shard myshard1 at myshard1/node3:27018
{
  data: '248.01GiB',
  docs: 210646,
  chunks: 0,
  'estimated data per chunk': '0B',
  'estimated docs per chunk': 0
}
---
Shard myshard3 at myshard3/node1:27018
{
  data: '249.23GiB',
  docs: 44758,
  chunks: 0,
  'estimated data per chunk': '0B',
  'estimated docs per chunk': 0
}
---
Shard myshard2 at myshard2/node2:27018
{
  data: '248.63GiB',
  docs: 65768,
  chunks: 0,
  'estimated data per chunk': '0B',
  'estimated docs per chunk': 0
}
---
Totals
{
  data: '745.88GiB',
  docs: 321172,
  chunks: 0,
  'Shard myshard1': [
    '33.25 % data',
    '65.58 % docs in cluster',
    '1.2MiB avg obj size on shard'
  ],
  'Shard myshard3': [
    '33.41 % data',
    '13.93 % docs in cluster',
    '5.7MiB avg obj size on shard'
  ],
  'Shard myshard2': [
    '33.33 % data',
    '20.47 % docs in cluster',
    '3.87MiB avg obj size on shard'
  ]
}

also my data size is the following which is not balanced:

14G	node1
15G	node2
33G	node3

Can anyone please help me with this?

3

Hi @Sobhan_Safdarian and welcome to MongoDB community forums!!

As per my understanding from the above statements, you are trying to connect multiple single node shard servers together. Please correct me if my understanding is wrong here.
However, could you confirm, if the config server is also a single node replica set ?
If my understanding is correct, as mentioned in the official documentation for Sharded cluster, for the production environment, you should use a replica set with three members.
It would be helpful, if you could convert all the replica set with three node replica set and connect to form a sharded cluster and let us know if you are still facing the similar issue.

Also, it would be helpful to assist you further if you could help me with the below mentioned info:

  1. The MongoDB version you have mentioned is 6.0.2 but the mongos version is 6.0.4 which shows a conflict in the version. For a production environment, it is recommended to have same version for all the shards and config servers and mongos of the sharded deployment.
  2. Was the collection sharded post inserting of the document of before inserting the documents. Because this error might occur because of the timeout that occurs when balancer tried to divide the data in between the chunks.
    I tried to replicate both the scenarios and I could see the same error when I tried to insert the data first before i sharded the collection but I still see data in the chunks which in your case seems to 0B. In the later case, when i inserted documents in a. sharded collection, I could see the error and the chunks are evenly distributed.
  3. Could you help me with the exact steps you followed which made you result in this error?
  4. was the database/collection user previously existed and was dropped, or is this a brand new collection?

Let me know if you have any further questions.

Warm Regards
Aasawari