Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
$documents is returning no documents when connected to a 6.1.0 mongos anna.henningsen@mongodb.com discovered that this only happens if the database is empty. Enterprise [direct: mongos] test> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) [ { X: 10 }, { X: 2 }, { X: 5 } ] Enterprise [direct: mongos] test> db.dropDatabase() { ok: 1, dropped: 'test' } Enterprise [direct: mongos] test> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) Enterprise [direct: mongos] test>
JIRAUSER1271286 commented on Tue, 24 Jan 2023 21:20:03 +0000: This was fixed on 6.0.3, not 6.0.2. Similarly, this was an issue in 6.1.0 but was fixed in a later 6.1 patch release. Since 6.0.2 is a patch release, the fix will not be backported. 6.0.2: MongoDB Enterprise mongos> db.version() 6.0.2 MongoDB Enterprise mongos> use test switched to db test MongoDB Enterprise mongos> db.dropDatabase() { "ok" : 1, "$clusterTime" : { "clusterTime" : Timestamp(1674592837, 1), "signature" : { "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="), "keyId" : NumberLong(0) } }, "operationTime" : Timestamp(1674592837, 1) } MongoDB Enterprise mongos> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) MongoDB Enterprise mongos> 6.0.3: MongoDB Enterprise mongos> db.version() 6.0.3 MongoDB Enterprise mongos> use test switched to db test MongoDB Enterprise mongos> db.dropDatabase() { "ok" : 1, "$clusterTime" : { "clusterTime" : Timestamp(1674594149, 1), "signature" : { "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="), "keyId" : NumberLong(0) } }, "operationTime" : Timestamp(1674594149, 1) } MongoDB Enterprise mongos> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) { "X" : 10 } { "X" : 2 } { "X" : 5 } MongoDB Enterprise mongos> dmitry.lukyanov commented on Tue, 24 Jan 2023 20:01:25 +0000: We see this issue again (with server 6.0.2): Enterprise [direct: mongos] test> use test already on db test Enterprise [direct: mongos] test> db.dropDatabase() { ok: 1, dropped: 'test' } Enterprise [direct: mongos] test> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) Enterprise [direct: mongos] test> db.test.insertOne({ _id : 1 }) { acknowledged: true, insertedId: 1 } Enterprise [direct: mongos] test> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) [ { X: 10 }, { X: 2 }, { X: 5 } ] Enterprise [direct: mongos] test> Server details: Enterprise [direct: mongos] test> db.version() 6.0.2 Enterprise [direct: mongos] test> sh.status() Warning: MongoshWarning: [SHAPI-10003] You are not connected to a mongos. This command may not work as expected. shardingVersion { _id: 1, minCompatibleVersion: 5, currentVersion: 6, clusterId: ObjectId("63d032d0ab474b70b1cd5aec") } --- shards [ { _id: 'shard01', host: 'shard01/localhost:27018,localhost:27019,localhost:27020', state: 1, topologyTime: Timestamp({ t: 1674588891, i: 11 }) }, { _id: 'shard02', host: 'shard02/localhost:27021,localhost:27022,localhost:27023', state: 1, topologyTime: Timestamp({ t: 1674588891, i: 17 }) } ] --- active mongoses [ { '6.0.2': 1 } ] --- autosplit { 'Currently enabled': 'yes' } --- balancer { 'Currently running': 'no', 'Currently enabled': 'yes', 'Failed balancer rounds in last 5 attempts': 0, 'Migration Results for the last 24 hours': { '31': 'Success' } } --- databases [ { database: { _id: 'config', primary: 'config', partitioned: true }, collections: { 'config.system.sessions': { shardKey: { _id: 1 }, unique: false, balancing: true, chunkMetadata: [ { shard: 'shard01', nChunks: 993 }, { shard: 'shard02', nChunks: 31 } ], chunks: [ 'too many chunks to print, use verbose if you want to force print' ], tags: [] } } }, { database: { _id: 'test', primary: 'shard01', partitioned: false, version: { uuid: new UUID("e3a8ab91-fa67-4695-8732-de189a5d2a84"), timestamp: Timestamp({ t: 1674588968, i: 1 }), lastMod: 1 } }, collections: {} } ] Enterprise [direct: mongos] test> Has the fix been backported? The same behavior we see with rapid env. The server version 6.1.0: db version v6.1.0 Build Info: { "version": "6.1.0", "gitVersion": "0ca11aca38c75d3c8fb5bac5bd103b950718a896", "modules": [ "enterprise" ], "allocator": "tcmalloc", "environment": { "distmod": "windows", "distarch": "x86_64", "target_arch": "x86_64" } } dmitry.lukyanov commented on Fri, 23 Dec 2022 00:36:37 +0000: cc: alyssa.wagenmaker@mongodb.com dmitry.lukyanov commented on Fri, 23 Dec 2022 00:30:34 +0000: Recently we've noticed this issue once again in this test. See this patch for details. However I wasn't able to reproduce this on my windows machine. MongoDB Enterprise mongos> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) { "X" : 10 } { "X" : 2 } { "X" : 5 } MongoDB Enterprise mongos> db.dropDatabase() { "ok" : 1, "$clusterTime" : { "clusterTime" : Timestamp(1671755091, 1), "signature" : { "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="), "keyId" : NumberLong(0) } }, "operationTime" : Timestamp(1671755091, 1) } MongoDB Enterprise mongos> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) { "X" : 10 } { "X" : 2 } { "X" : 5 } MongoDB Enterprise mongos> db.version() 6.3.0-alpha-743-g5732e9c cc: robert@mongodb.com JIRAUSER1271286 commented on Thu, 8 Dec 2022 21:47:35 +0000: I reproduced the issue on 6.1.0, but it appears to have been fixed since then on the v6.1 branch: [direct: mongos] test> db.version() 6.1.0-rc3-100-g3b45a1f [direct: mongos] test> db.dropDatabase() { ok: 1, dropped: 'test' } [direct: mongos] test> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) [ { X: 10 }, { X: 2 }, { X: 5 } ] [direct: mongos] test> It also is working on earlier versions (I checked on the current v6.0 branch) as well as on master. Given that, I'll close this ticket.
Enterprise [direct: mongos] test> use Tests11301056 switched to db Tests11301056 Enterprise [direct: mongos] Tests11301056> db.aggregate([{ $documents : [{ X : 10 }, { X : 2 }, { X : 5 }] }]) Enterprise [direct: mongos]