Log file Viewer
#mongodb logs for Thursday the 21st of July, 2016
[01:42:23] <warp0x00> Jonno_FTW, https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkArEQFggdMAA&url=http%3A%2F%2Fstackoverflow.com%2Fquestions%2F12036941%2Fmongodb-consumes-a-lot-of-memory&usg=AFQjCNGrJGZ6ZtlV0RoMndf4gX_NQ-M6nA&sig2=LY9S6d-DuY6k_baY3I1bnQ https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkAr
[01:42:29] <warp0x00> EQFggkMAE&url=http%3A%2F%2Fstackoverflow.com%2Fquestions%2F9142304%2Fmongodb-is-using-too-much-memory&usg=AFQjCNG1kV0CsyOX4xMxGUjj7qe_PvSKEQ&sig2=J8IhTjQPDt1oJ-ezuZtvvg https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&cad=rja&uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkArEQFggrMAI&url=http%3A%2F%2Fstackoverflow.com%2Fquestions%2F6861184%2Fis-there-any-option-to-limit-mongodb-memory-usage&usg=AF
[01:42:35] <warp0x00> QjCNFK4iOcnt0QZdjAOAJRC-_1MGahKg&sig2=2qPsjEcrCqxoBQ57qBFi2w https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&cad=rja&uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkArEQFggyMAM&url=https%3A%2F%2Fwww.quora.com%2FMongoDB-Why-my-mongod-uses-too-much-virtual-memory&usg=AFQjCNGKa5CYjp53JDid4QgfMNMEfmFSAA&sig2=Gky1G5rXG17zpTZMXTWgwA https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=6&cad=rja&
[01:42:41] <warp0x00> uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkArEQFgg8MAU&url=http%3A%2F%2Fdba.stackexchange.com%2Fquestions%2F64653%2Fmongo-takes-too-much-memory-when-insert-data&usg=AFQjCNFP4zK-Pg8V_Z-RhKhCp7EMxUQu0g&sig2=6uRPyXhIQiC6OO2u93Vs0A https://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&uact=8&ved=0ahUKEwjWl_LUtIPOAhUC7YMKHUDkArEQFghDMAY&url=http%3A%2F%2Fdba.stackexchange.com%2Fquestions%2F58436%
[01:42:47] <warp0x00> 2Fwhy-does-mongodb-virtual-memory-get-so-large&usg=AFQjCNHe7rfr6Bpnbz2lOvhCkvaDPVZOxA&sig2=hMG6vlkGNgok_hvJ_FfXVw
[01:42:57] <warp0x00> Jonno_FTW, https://dba.stackexchange.com/questions/58436/why-does-mongodb-virtual-memory-get-so-large
[01:43:07] <warp0x00> Jonno_FTW, https://dba.stackexchange.com/questions/64653/mongo-takes-too-much-memory-when-insert-data
[01:43:16] <warp0x00> Jonno_FTW, https://www.quora.com/MongoDB-Why-my-mongod-uses-too-much-virtual-memory
[01:43:31] <warp0x00> Jonno_FTW, https://stackoverflow.com/questions/6861184/is-there-any-option-to-limit-mongodb-memory-usage
[01:43:39] <warp0x00> Jonno_FTW, https://stackoverflow.com/questions/9142304/mongodb-is-using-too-much-memory
[01:43:49] <warp0x00> Jonno_FTW, https://stackoverflow.com/questions/9142304/mongodb-is-using-too-much-memory
[03:13:47] <trq> Hi people. We currently have a replica set with 1 primary and one secondary, both running 2.6.something. We want to move to 3.2. I have looked at the docs and the apparent upgrade path is 2.6 -> 3.0 then 3.0 to 3.2. What I am wondering however is this. Could we create two entirely new secondaries have the data synced accross and then when we are ready step down the old secondary, then step down the old
[03:14:42] <trq> Failing that, I would probably do the same process but create these new instances using 3.0, sync them, then upgrade them to 3.2 and sync them.
[03:36:42] <joannac> trq: A mixed-mode replica set with 2.6 and 3.2 has never been tested, so I can't validate it would work
[03:38:11] <joannac> In addition, there is a change in replication protocol in 3.2, and you won't be able to upgrade to that until all nodes are on 3.2
[03:39:22] <trq> Probably just as easy to keep the machines we have them and upgrade them using the "rolling" upgrade docs.
[03:41:04] <joannac> you can upgrade the arbiter to 3.0.x, upgrade b to 3.0.x, then upgrade c to 3.2.x, then upgrade b and a to 3.2.x
[03:41:46] <joannac> I would also go out on a limb and say I highly doubt there would be any issue going straight to 3.2 with your arbiter
[03:42:38] <joannac> don't forget to upgrade the repl protocol https://docs.mongodb.com/manual/release-notes/3.2-upgrade/#upgrade-the-replication-protocol
[13:13:40] <obiwahn> i have a file with data and i want to import every line so that the linenumber is the key and the data is addressable as data. Can I do that with somehow with mongoimort?
[13:17:09] <obiwahn> so i need to write the line number into the file and do a cvs import with --fields _id, data?
[13:50:28] <jayjo> Can I pass my ssl cert in a connection string? Trying to set up the bi connector and have ssl enabled on my db
[13:57:53] <jayjo> ok thanks - I got that to go. As the root user of either the db or the server, can I reset the password of the biuser I just created?
[14:25:36] <jayjo> I've gone through all of the steps of setting up the bi connector and the biuser. If my mongo instance has ssl, does this "postgres" instance have the same ssl creds?
[14:41:43] <LiranM> would appreciate it very much if anyone can help me solve this issue. many thanks in advance.
[14:41:47] <LiranM> http://stackoverflow.com/questions/38507003/mongodb-cannot-query-a-sharded-collection-from-mongos-after-filesystem-snapsho
[14:46:16] <jayjo> Via pgadmin3, I can connect to the 'postgres' instance, which is a lie of an instance, but when I try to run queries that retrieve data (I can do anything that doesn't actually access the data right now) I get back the error (from psql): ERROR: Error in python: ConfigurationError Detail: `ssl_cert_reqs` is not ssl.CERT_NONE and no system CA certificates could be loaded. `ssl_ca_certs` is required.
[14:51:42] <jayjo> I don't know the solution to this - security is important to me, but in order to make this work do I need to disable ssl for the connector to do its work?
[15:10:22] <jayjo> I think I know the error- too ... I got it to work by changing the order of the connection string. I needed the paramater ssl_ca_certs in the string to be the first. The rest are ignored (and the command hangs after running it)
[17:37:39] <DigitalSpeed> so im getting started with a database for the first time and decided to try out mongo...i just created my /data/db directory...how do i claim ownership over this directory so that i can change the read/write permissions?
[18:27:41] <`jpg> I'm seeing strange behaviour when trying to serialise the replSetConfig using mgo BSON.
[18:28:39] <`jpg> Basically I want to be able to set the version field to a specific integer for each version. This works for 1 but anytime I try to do a replSetReconfig with a version other than 1 specified I see the server applying a completely different versoin.
[19:06:48] <DigitalSpeed> okay im having crazy amounts of problems with mongo and looking for some help in simply getting the correct version up and running. been attempting to debug for a while and having no luck.
[19:07:56] <DigitalSpeed> ive ran that in an attempt to remove mongodb completely from my machine yet i still have mongo somewhere
[19:08:25] <DigitalSpeed> if i type just mongo in my shell it attempts to run but then gives me an error
[19:08:44] <DigitalSpeed> Error: couldn't connect to server 127.0.0.1:27017, connection attempt failed :
[19:09:31] <DigitalSpeed> ive ran sudo dpkg -l | grep mongodb and it shows up i believe in the amd64 folder though im not sure where that is located
[19:10:18] <DigitalSpeed> my thinking is that the purge would remove it but obviously that isnt working
[19:10:46] <DigitalSpeed> ii mongodb-org 3.2.8 amd64 MongoDB open source document-oriented database system (metapackage)
[19:52:06] <`jpg> Is it even possible to set the version for a replica set configuration? (aside from the initial replica set configuration at version 0)
[20:05:54] <`jpg> cheeser to enforce that updates to the configuration are only applied in the order I want them to be
[20:06:15] <`jpg> I have a state machine managed by a consensus store that manages the replica set topology
[20:06:56] <`jpg> I need to be able to use the "force" replSetReconfig because at times I may need to recover the replica set when a quorum is not available
[20:07:09] <`jpg> inspecting the code I find that the version is randomly incremented if and only if you are using the force flag
[20:07:17] <DigitalSpeed> ive gotten further along now by specifying each individual package that needed to be uninstalled such as sudo apt-get remove mongodb-org-shell --purge etc but still having issues with getting things running in general. not really sure what the problem is but i think it may have to do with permissions...ive already set the permissions for /data/db and a few other folders/files but i have no idea which ones exactly i need to change
[20:07:57] <cheeser> DigitalSpeed: https://docs.mongodb.com/manual/tutorial/install-mongodb-on-debian/
[20:09:07] <`jpg> I am trying to understand the motivation but am at a loss as to why it would ignore the operator provided version number
[20:11:03] <DigitalSpeed> that's essentially the tutorial i followed EXACTLY and it still isnt working. ive changed the permissions already on those two files/directories as well /var/lib/mongodb and /var/log/mongodb
[20:11:04] <cheeser> i would ask on the mailing list first. chances are your understanding might be off a bit. could be a bug, though.
[20:11:34] <cheeser> DigitalSpeed: you shouldn't need to change any permissions if you're doing that as root or sudo'ing appropriately
[20:16:00] <DigitalSpeed> if i just type mongod into my shell everything seems to be operating okay...with the last line of the code as waiting for connections on port 27017. but then when i attempt to start it as a service then open another shell and type sudo systemctl status mongod i get an error telling me how it exited with an invalid argument.
[20:16:18] <DigitalSpeed> is it possible that i messed up the installation with attempting to change the permissions?
[20:37:59] <dsardari> right on. well i suppose it's not morphia specifically, but using the @Embedded tag, creating that object then adding it to two separate lists on two separate entities should be fine, right?
[20:45:18] <dsardari> does is it seem typical for a tagging/label system to have a list of embedded tags for a user and embedded tags for each taggable object? deletes/updates on the user tags would have be reflected on the tagged objects and filtering on collections based on the tags would be required. i wasn't sure if references or embedding would make more sense.
[20:48:11] <cheeser> depends on the nature of the tags but usually they're just dumb strings and can be embedded in an array on the taggable item.
[20:50:20] <dsardari> essentially, yeah. a string and an associated color, possibly additional fields in the future. i'm mostly worried about the propagation of delete/update operations across multiple collections and a potentially large amount of items in each.
[21:03:01] <cheeser> if you find yourself updating these labels, it might make sense to make the references. though that has certain obvious performance updates on querying.
[21:03:42] <cheeser> it *might* be better to propagate such updates to all the various embedded tags instead. that's a cost you'd pay once for an update (which is hopefully rare) vs every query.
[22:02:37] <DigitalSpeed> so i completely uninstalled then reinstalled using that exact page you linked (for probably the third time today) and still no luck.
[23:15:02] <tambu> Anyone open to a query question. http://pastebin.com/E66mKqdE .. assuming the objects under tasks are random and there could be many. I'd like to query for documents with a task that has a status: "running".. also is there a way to filter out the rest of the Task ID that do not have status running from a returned document?