Log file Viewer
#mongodb logs for Wednesday the 9th of September, 2015
[00:01:32] <curlhelp> In MongoDB is it better to have a parent record with an array of child IDs? Or for every child to have a "parent" column with the parent ID? Which is more efficient/cost effective memory wise?
[00:03:55] <cheeser> it does depend on the case, but i generally prefer children pointing to parents.
[00:05:22] <curlhelp> to expand on your answer, however (for records); "The children should have parent IDs so that the number of children can grow without bound. Putting children IDs in the parent would cause the parent document to grow over time leading to document moves as the doc needs more space on disk and ultimately the 16MB limit on doc sizes. Though if you get to that 16MB limit you probably have other issues..."
[00:06:55] <cheeser> a single document can only be 16mb large. you can have a trillion of those documents.
[00:06:58] <StephenLynx> if you expect a large number of children, I suggest you put the children in a separate collection.
[00:07:21] <curlhelp> interesting... yeah so i think i'll definitely go with the children with parent id approach
[00:07:55] <curlhelp> if i want to track something like "facebook" likes (dw i'm not building another social media thing lol), would a "post" have "liked_by":[iD, ID, ID]
[00:08:14] <StephenLynx> embedded children are good if you expect the children to have multiple parents, to be complex cases or to have a limited number of children.
[00:08:33] <cheeser> curlhelp: in that case, yes. separate collection because those are different entities.
[00:09:11] <curlhelp> thanks for your help everyone sorry for xposting and hope i was able to mend that by planting question here
[03:49:21] <energizer> How do I do a find({}, {}) that shows all of the data for every matching document?
[05:47:22] <Boomtime> @Jonno_FTW: there are complete self-contained examples here: https://mongodb.github.io/node-mongodb-native/api-articles/nodekoarticle1.html#time-to-query
[05:57:45] <Boomtime> Jonno_FTW: what's wrong with the regular mongo shell? you need to test the query first
[05:59:53] <Boomtime> if you provide the output from each of these things, and pastebin it - the call you make in node, and the query line from the mongo shell, someone here might be able to determine what is wrong
[09:09:10] <mapc> Specifically we're having the problem, that we have some very expensive, long running queries and we'd like to give them a lower priority/nice value
[09:09:39] <mapc> so those queries are just using any leftover resources and not claiming any resources for their own.
[09:18:18] <gcfhvjbkn> i get "could not verify that config servers are in sync" error on my mongos instances; after googling a little bit i see that i should find the problem on my own and fix it manually; my problem is that i don't really know what to look for, it doesn't seem to be any documentation about this
[09:26:11] <gcfhvjbkn> btw is it ok if some of the configservers only have two tables in "config" db? ("lock" and "lockpings") out of all these http://docs.mongodb.org/manual/reference/config-database/
[12:51:43] <taspat> with underscore I can do that but maybe there is some built in fucntion in mongo
[12:59:28] <nekyian> I am building a website in PHP & MongoDB. I have this weird error while running an import script. Can you help?
[12:59:37] <nekyian> localhost:27017: Read timed out after reading 0 bytes, waited for 30.000000 seconds
[13:30:31] <christo_m> i dont want to log into the remote and copy from there and have to open ports etc to access my machine here
[14:03:03] <EricL> I have a sharded Mongo 2.4 cluster with 3 shards, 3 replicas per shard. I am trying to reduce it down to 2 shards. I followed these instructions: http://docs.mongodb.org/v2.4/tutorial/remove-shards-from-cluster/ and have made it to the point where I am just checking the status. However the remaining chunks hasn't changed since I started it an hour ago and the status is "draining ongoing". How do I know if something is happening?
[14:04:10] <CustosL1men> is there something like dbegines ranking which was maintained before 2011 ?
[14:37:45] <preyalone> how do i select certain fields, without matching on any particular values for those fields?
[14:42:15] <preyalone> odd, db.config.find({}, {submission:1}) returns documents with _id's, but no submissions
[14:44:05] <preyalone> my documents don't appear to have a submission field, but that query is returning results anyway. without submission fields.
[14:49:22] <EricL> I have a sharded Mongo 2.4 cluster with 3 shards, 3 replicas per shard. I am trying to reduce it down to 2 shards. I followed these instructions: http://docs.mongodb.org/v2.4/tutorial/remove-shards-from-cluster/ and have made it to the point where I am just checking the status. However the remaining chunks hasn't changed since I started it two hours ago and the status is "draining ongoing". How do I know if something is happening? I
[16:02:42] <EricL> I have a sharded Mongo 2.4 cluster with 3 shards, 3 replicas per shard. I am trying to reduce it down to 2 shards. I followed these instructions: http://docs.mongodb.org/v2.4/tutorial/remove-shards-from-cluster/ and have made it to the point where I am just checking the status. However the remaining chunks hasn't changed since I started it two hours ago and the status is "draining ongoing". How do I know if something is happening? I
[16:32:07] <StephenLynx> if you are going to use ubuntu (which is not recommended for a server by a considerable amount of people) you would be better using 14.
[16:36:35] <kashike> is there really no fix for this? running the mongod process manually works, just the init script doesn't work
[16:38:34] <kashike> I'm using 'eb http://repo.mongodb.org/apt/ubuntu trusty/mongodb-org/3.0 multiverse' as the repo currently
[16:39:32] <BadCodSmell> so even sourcing their packages and manually building an update is a nightmare
[16:39:53] <BadCodSmell> at least with things like centos you know all those patches are almost always going to be to make it build properly or backports andway you can ignore
[16:39:56] <StephenLynx> for a server you either want a rolling distro or something derivate from RHEL
[16:40:39] <BadCodSmell> one package in debian I tried to manually update but gave up after it had several patches including new features including a 100k patch to add ipv6 and a massive patch to add authentication
[16:41:02] <BadCodSmell> And where they get those patches from *shrugs* unofficial pull requests in git?
[16:42:19] <BadCodSmell> and if you look at the complexity of their packaging and how much work they give themselves compared to a simple spec file, OMG, and the awful quality of the huge helper bash scripts (vomit)
[19:04:33] <mprelude> http://stackoverflow.com/questions/32486579/implementing-mongodb-sasl-authentication-in-cl-mongo In case anyone knows how this works? Have got SCRAM implemented, but having issues making initial request.
[19:55:37] <saml> i don't want clients to be connected to the hot standby member with mongodb://mongo1,mongo2,mongo3/test where mongo4 is hot standby
[20:03:32] <deathanchor> oh I just stop mongod on the hidden, start a backup of the disk, and then start it back up.
[20:21:29] <Doyle> Does anyone know if the response to this ticket is accurate? https://jira.mongodb.org/browse/SERVER-19261 "local.oplog.rs Assertion failure a() != -1 src/mongo/db/storage/record.cpp 538"
[20:35:05] <Doyle> Interesting. I hadn't noticed before, but the lgoSizeMB reports the default 51200, while the local db is 120GB, as set when I adjusted the oplog size initially.
[20:37:20] <Doyle> db.local.stast() gives ... { "ok" : 0, "errmsg" : "Collection [test.local] not found." }
[20:39:11] <Doyle> Immediately before the assertion failure, "replSet initial sync cloning db: test"
[22:15:06] <EricL> I have a sharded Mongo 2.4 cluster with 3 shards, 3 replicas per shard. I am trying to reduce it down to 2 shards. I followed these instructions: http://docs.mongodb.org/v2.4/tutorial/remove-shards-from-cluster/ and have made it to the point where I am just checking the status. However the remaining chunks hasn't changed since I started it two hours ago and the status is "draining ongoing". How do I know if something is happening?
[22:28:24] <Fernandos> I added a new array field to my schema. How can I extend existing objects in a collection that don't have this field? Creating new objects is no problem, but editing objects created without that field doesn't work.
[22:37:41] <tejasmanohar> db.getCollection('pres').update({}, {$set:{new_field:'source'}}, false, true)
[23:16:19] <kashike> StephenLynx: just thought I'd let you know I have mongodb 3.0.6 running on ubuntu 15.04 successfully