If a new job is started as the test successfully compeltes and begins
the scheduler shutdown process. The new job will deadlock waiting on
channel reads.
Instead of requiring the each proxy handler to handle the webSocket
connection closure. On error the worker handler will appropriately
close the connection and event any errors.
This adds new dimension table to record credential information. Since a
session can have multiple credentials, the wh_session_accumulating_fact
is is linked to the credential_dimensions via bridge tables,
wh_credential_group and wh_credential_group_membership. When a session
is inserted, the credental group and credential dimensions will be
upserted. If the group changed in any way, a new group is inserted along
with all of the corresponding credentials.
This adds interfaces for the following credential types:
- Username and password
- Username and private key
- Certificate and private key
It also adds the following types for redacting secret data:
- Password
- PrivateKey
This also affected credential libraries and managed groups. It seems
that any end-to-end tests of these grants currently had them included in
wildcards instead of individually, but these types were never added to
the valid list for resources. Additionally, credential store wasn't
added as a valid top level type for pinning purposes.
The docker package and its StartDbInDocker function should only be used
by the `boundary dev` command now. So moving it into an internal package
of `base` will prevent it from being use for tests accidentally, while
still allowing it to be used for `boundary dev`.
This swaps the test initialization to use the dbtest package so each
test will get a fresh database created from a template database, instead
of a completely new docker container.
* feat(host): Add host plugin related schemas and tests
This adds the schema files for host plugins, plugin based catalogs and
plugin based sets. This also adds the protos and tests which utilize
gorm to ensure the schemas enforce the constraints we expect.
Along with this change was a migration which moved catalog names to the
base type table so we can ensure all catalogs in a single scope,
regardless of subtype, have a unique name.
* Updated plugin schemas to do the following:
- Add plugin_executable table
- plugin table now also has scope_id and name w/ uniqueness constraints
- host_plugin no longer has id_prefix column
- host_plugin separates name and plugin_name
- plugin_version table renamed version to symantic_version column.
* Fix typo in delete_plugin_subtype trigger. Add update test for host plugins.
* Update id prefixes. Fix public_id typos in protos and plugin_executable name in sql diagrams.
* Created internal/plugin and moved base plugin structs to it and moved host plugin subdirectory to it.
* Created internal/plugin and moved base plugin structs to it and moved host plugin subdirectory to it.
* Add first pass at enums for OS and Arch supported by plugin executables.
* Capitalizing the const enums for OS and Architectures.
* Add comments to sql files.
* Reviewer comments.
Remove unsupported OSes and Archs from SQL and go constants.
Add missing tests.
* Be more permissive in the values the db accepts.
The go logic can restrict which values it will allow getting populated into the db.
* Removing constraints on which values can be placed inside operating_system and architecture columns.
* Updating test helper functions to only create a single catalog or set.
* Move files to new migration version.
* Fixing job/plugin test.
* Update table names, named a number of constraints, remove plugin version and executable tables.
* Add constraints for plugin name and id prefix.
* Rename constraint, remove unhelpful comment block.
* Adding semantic version to plugin_host and set attributes to not null.
* Remove semantic versioning from all locations.
* Remove semantic_version from plugin_host table comment.
* Fix op copy/paste error. Don't use deprecated errors methods. Update constraint names. Add docs.
* Apply suggestions from code review
Fixing constraint names.
Co-authored-by: Michael Gaffney <mgaffney@users.noreply.github.com>
* `recovery` is only used by the API (and the main module, for decryption)
* `wrapper` is only used by the CLI right now
Moving these brings them closer to the code that uses them, and makes
the SDK package a bit more clear in terms of functionality.
When a session is terminated by the user, the cancel session path is
triggered, in this path the websocket connection is closed. If the
connection is already closed, an EOF is returned. In this situation an
EOF is expected and should not be logged to the user.
Example log line:
{
"id": "ezqk6w9dOQ",
"source": "https://hashicorp.com/boundary/dev-controller/boundary-dev",
"specversion": "1.0",
"type": "error",
"data": {
"error": "failed to close WebSocket: failed to read frame header: EOF",
"error_fields": {},
"id": "e_n60YPsDkaj",
"version": "v0.1",
"op": "worker.(Worker).handleProxy",
"info": {
"msg": "error closing client connection"
}
},
"datacontentype": "text/plain",
"time": "2021-09-02T10:05:46.587513-07:00"
}
The existing logic was to ignore Forbidden errors when recursively
listing in favor of looking for downstream scopes with appropriate
permissions. The problem is that when there is actually no
authentication token at all, what comes back is an Unauthenticated
error. Therefore if you didn't have permissions on the ingress scope
(e.g. the default of `global`) but did downstream for the anonymous
user, you'd get back an error.
I don't think there's any security issue with doing this; the user could
always have simply started the search at a downstream scope to begin
with so nothing more is being revealed, this just fixes behavior to
match expectations in this case.
A test was added for one of the services; the code is exactly the same
across them. The reason that the existing tests that were added with the
output fields testing didn't catch this is that even though they were
specifying `u_anon` as the user, the disabled test context skipped the
logic that would return a 401 in this case. It turns out that fixing
this is very complicated because we don't have a great way in most tests
of defining grants to put onto the generated scopes, so even if a true
anonymous context was added each test would require the anonymous user
to have list permissions on whatever resource was under test, requiring
a lot of role and grant and principal finagling. This is probably a good
longer-term TODO but not one that seemed worth it for this fix.
Fixes ICU-2301
Jeff Mitchell
Todd
Louis Ruch
Jim
Timothy Messier
Michael Gaffney
Randall MoreyandSusmitha Girumala
Chris Marchesi