WiredTiger storage structure

[megha garg]

Hi,

I was trying new wiredTiger storage in 3.0.0 with some of my production data. But I am not able to understand its storage structure. I restored one collection from 2.6.7 to 3.0. the data directory created below files/directories. I am not able to find any description for this:-

collection-0--7880187485072602249.wt

collection-2--7880187485072602249.wt

index-1--7880187485072602249.wt

index-3--7880187485072602249.wt

index-4--7880187485072602249.wt

index-5--7880187485072602249.wt

index-6--7880187485072602249.wt

index-7--7880187485072602249.wt

index-8--7880187485072602249.wt

journal

_mdb_catalog.wt

mongod.lock

mongo.log

sizeStorer.wt

storage.bson

_tmp

WiredTiger

WiredTiger.basecfg

WiredTiger.lock

WiredTiger.turtle

WiredTiger.wt

Can somebody help me with this?

[Tim Callaghan]

This is the current state of not just WiredTiger but also TokuMXse, and it's unfortunate. There is no way that I can see to map files to indexes/collections.

-Tim

[Asya Kamsky]

The metadata mapping the collection and index names to files is in the
_mdb_catalog.wt for WiredTiger.

What's the primary reason for wanting to match up files to objects in
the DB? You can probably get the mapping by comparing
db.collection.stats().storageSize and matching it with sizes of
collection-xxx files and same for index sizes...

We can also decode the contents of _mdb_catalog.wt but I guess I'd
like to understand why that's important to have available...

Asya

> --
> You received this message because you are subscribed to the Google Groups
> "mongodb-user"
> group.
>
> For other MongoDB technical support options, see:
> http://www.mongodb.org/about/support/.
> ---
> You received this message because you are subscribed to the Google Groups
> "mongodb-user" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to mongodb-user...@googlegroups.com.
> To post to this group, send email to mongod...@googlegroups.com.
> Visit this group at http://groups.google.com/group/mongodb-user.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/mongodb-user/687cdddd-3fde-40aa-b996-18ba8a4daf13%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[megha garg]

I just wanted to understand the basic storage structure of wiredtiger as it has compressed the data. 

How can we get the mapping for indexes?

[megha garg]

And how can we read this _mdb_catalog.wt file

[Asya Kamsky]

The same way - each file has one index, its size on disk and in RAM is the same as it's prefix-compressed in both.

Asya

To view this discussion on the web visit https://groups.google.com/d/msgid/mongodb-user/5b174826-6c10-4249-90b7-1c4d6cdbb7b2%40googlegroups.com.

[Deepjyoti Saha]

Hi Megha

Use directoryperdb=1 option in config file while running mongod process.
It will create all the database folders and map all the database collection and index files with that particular database folder.