Glean stores facts in databases.
A Glean database has:
- A name. This is often (but not always) the name of the source code repository from which the facts in the database were collected. Indeed, for historical reasons we sometimes say "repository" or "repo" when we mean "database".
- A hash. This is just an arbitrary string, used to distinguish different versions of databases with the same name. For source code repositories it can be the revision of the repository that was indexed.
- Some properties. A database has a set of key-value pairs associated with it. Some of these are created by Glean itself, and others can be set by a client when the DB is created, or while writing facts. Properties can be used to store arbitrary metadata about the DB, and can be accessed cheaply.
The name and hash together uniquely identify a database. This is written
<name>/<hash>, and it is how you refer to a database in most cases when working with Glean. For example, in the shell's
:db command, or the
--repo argument to the command-line tools.
Working with local databases
Most Glean tools (in particular the shell and the CLI tool) can work either by talking to a Glean server or accessing databases on the local filesystem directly. The access method is chosen by these command-line flags:
--service <host>:<port>Connect to a remote Glean server.
--db-root <dir>Use databases stored locally in the directory
--db-tmpCreate a temporary directory to store DBs and delete it when the program exits.
These flags are accepted by all the Glean command-line tools,
To use the shell with local databases, you can do:
glean shell --db-root /tmp/glean
To run a server, see Running the Glean Server.
Lifecycle of a database
A database will typically be created by an automatic periodic job to index the current state of a source repository. The process works like this:
The job invokes
glean kickoff --service <write-server> <args>to create the database.
At this point the database is in the Incomplete state. Queries are supported in this state, and always reflect the current contents.
Facts are written to the database using the methods described in Writing data to Glean, and finally the database is closed by invoking
glean finish --service <write-server> <args>or the appropriate Thrift method.
The database is now in the Complete state.
If backups are allowed for this database, then:
- the write server uploads the database to backup storage.
- servers that are configured to restore databases automatically can download the DB from backup storage, and use it to serve queries from clients.
There are currently no backup backends implemented for open-source Glean.