Ensure the new create_time and update_time columns
are marked as not null after populating the data.
Also ensure that the data is populated safely with an
updated query pattern inspired by the auth method
pattern.
* wip
* implement list pagination for auth methods
* remove unused code
* update tests to fix sorting
* address PR comments
* simplify query, add token tests
* change test name
* wip sql testing
* add boolean column to indicate whether auth methods should be returned to unauthenticated users, remove secrets from api response
* is_active_public not null
* remove secrets and unwrapping
* move not null to immediately after data migration
* order public_id descending
* fix(db): Fix auth method migrations (#4193)
* wip
* fix db tests
* fix failing api tests
* reverse order of create time and update time in test
---------
Co-authored-by: Michael Gaffney <mike@gaffney.cc>
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids this issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The previous sort would run the risk of missing items which
collide on update or create times because of the compound
comparison using strictly less than for both the timestamp
and the public_id. Sorting both descending avoids thi issue.
The commands added are:
boundary daemon start
boundary daemon stop
boundary daemon add-token
boundary daemon status
boundary search
If these actions are performed on a platform that doesn't support sqlite
natively in go an error is returned.
Besides some hooks into the cmd package all the significant logic
resides in internal/clientcache
feat (cmds/authenticate): output the token name when storing token
feat: update go-dbw which now uses pure go sqlite driver
refact (base.Command): export FlagOutputCurlString
Exporting FlagOutputCurlString allows other commands that embed
the base.Command to access the flag as needed
feat: add a cache Repository to manage cache storage
feat: add a "cache" CLI command
This new command will support both managing and searching the
local CLI cache daemon.
feat: start cache daemon as needed when issue CLI commands
Move files to new directories and add persona table (#3526)
* Add persona table and move files to new directories
Client daemon now uses unix sockets and update pathing (#3535)
* Add unix socket listen and dialers with file permissions applied
* Update the pathing to match RFC
Cache refreshing now uses all stored personas. (#3545)
Updating headers for LLB files
Decouple daemon and cache server (#3566)
* Separate daemon from server
* Add persona handler to daemon server (#3579)
Moved search to use mql and added tests for the search command (#3589)
* search uses hashicorp/mql
Windows support for client side daemon (#3618)
* exec.Command for background process
* Log to stderr when in foreground and file always
* Use pid files across platform for start and stop
Requests with version info must match the daemon's version (#3640)
* requests with version info must match daemons
Update dependencies and breakout cgo/pure go sqlite drivers (#3657)
upgrade go-dbw deps and add cgo build tags
Enable graceful shutdown in windows (#3695)
* Enable graceful shutdown in windows
Co-authored-by: Jim <jlambert@hashicorp.com>
Targets delete when a persona changes address or user id (#3673)
* Handle cache refresh in repo
Add sessions to the client side cache (#3682)
* Add sessions to repository
* Search command supports sessions
* adding session related tests to db and refresh
Change cache schema from personas to tokens+users (#3737)
* Change from personas to tokens+users
* Separate keyring tokens from auth tokens
Add support for keyringless auth tokens (#3765)
* Add support for keyringless auth tokens
Remove expired auth tokens from cache (#3768)
* Remove expired auth tokens from cache
Complete the TODO of checking for inflight tx (#3779)
Delete tokens that the controller says are invalid (#3782)
fix search test with boundary token check (#3799)
Collect resource info in repository_<resource>s.go (#3769)
* Collect resource info in repository_<resource>s.go
* Move search logic into the domain (#3794)
Daemon and search restricted to supported platforms (#3801)
* Daemon and search restricted to supported platforms
Reorganize the clientcache packages (#3814)
* Refactor for self contained domain logic
* Put clientcache stuff under internal
* Add pagination support for auth tokens
* change analyze count estimate comment
* adjust queries for efficiency and security
* even more efficient queries
The new design uses the base table for querying, so the base
table needs to have both create_time and update_time. In this
commit we add the update_time to the target base table. We
also add a number of new indexes to help with looking up targets.
Creates a new foo_deleted table for each subtype of each
resource that needs to support pagination. The deleted
tables are used to track deletions of resources and return
them to users of the new pagination API. We track only the
ID and delete_time of the resource, so there's no way to
filter out resources that the user could not see before it
was deleted. This is intentional to keep the design simple
and is not considered a security issue.
Boundary OIDC method does not currently support passing in prompts during authentication. This change adds the capability of passing OIDC prompts. Prompts are optional OIDC parameters that determine the behaviour of the authentication server: https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
## Changes
- New `auth_oidc_prompt` table which contains all the prompts for OIDC auth method
- New `auth_oidc_prompt_enm` table which contains possible enum values for a prompt
- Currently supported:
- `none`:
The Authorization Server MUST NOT display any authentication or consent user interface pages
- `login`: The Authorization Server SHOULD prompt the End-User for reauthentication
- `consent`: The Authorization Server SHOULD prompt the End-User for consent before returning information to the Client
- `select_account`: The Authorization Server SHOULD prompt the End-User to select a user account
- `oidc_auth_method_with_value_obj` view has been updated to return `prompt` value
- Add `prompt` option for OIDC auth method CLI create and update
- Pass configured prompt during OIDC authentication
- Add `prompt` API validation for create and update
We inadvertently used
kms.GetWrapper(ctx, am.ScopeId, kms.KeyPurposeDatabase)
which resulted in the use of a database key for an oplog entry.
See:
https://github.com/hashicorp/boundary/pull/3665/files#diff-bcf83b5283886c98fc45b6435fdc94d87d545b2768c6beac8acd14eb9d6b13f1
This fixes the issue with the missing "Database" purpose key when
we converted the oplog keys into their own tables.
We chosen to truncate the existing oplog entries to eliminate any
references to a Database kms key in existing oplog entries.
* fix (events): convert multierror to stdlib error
multierror errors will be converted to a stdlib error (because
multierror doesn't support json marshaling).
* refact (errors): convert from multierror to stdlib errors.Join(...)
* refact: convert from pkg/errors to stdlib error wrapping
* fix (auth/ldap): fix which DEK is used for oplog during authenticate
* chore: update go-kms-wrapping dep for new extras/kms changes
The latest version supports setting a table name prefix for a kms
which is needed to fix our oplog DEK problem
* chore: update extras/kms dep
This latest version supports an optional TableNamePrefix which is
required to fix our existing oplog DEK issue.
* fix (oplog): schema oplog DEK fix
These changes will allow use to store the oplog DEK in it's own
set of tables, so they won't get deleted when a scope is deleted
* fix (kms): move oplog DEK into its own internal kms
* tests: update tests to be compatible with oplog fix
* tests: remove migration tests which are incompatible with oplog fix
These tests no longer work because they are missing the required
schema changes for the oplog fix
Johan Brandhorst-Satzkorn
Michael Milton
Todd
Jim
Damian Debkowski
Elim Tsiagbey
dani
Irena Rindos