* logical/aws: Harden WAL entry creation
If AWS IAM user creation failed in any way, the WAL corresponding to the
IAM user would get left around and Vault would try to roll it back.
However, because the user never existed, the rollback failed. Thus, the
WAL would essentially get "stuck" and Vault would continually attempt to
roll it back, failing every time. A similar situation could arise if the
IAM user that Vault created got deleted out of band, or if Vault deleted
it but was unable to write the lease revocation back to storage (e.g., a
storage failure).
This attempts to harden it in two ways. One is by deleting the WAL log
entry if the IAM user creation fails. However, the WAL deletion could
still fail, and this wouldn't help where the user is deleted out of
band, so second, consider the user rolled back if the user just doesn't
exist, under certain circumstances.
Fixes#5190
* Fix segfault in expiration unit tests
TestExpiration_Tidy was passing in a leaseEntry that had a nil Secret,
which then caused a segfault as the changes to revokeEntry didn't check
whether Secret was nil; this is probably unlikely to occur in real life,
but good to be extra cautious.
* Fix potential segfault
Missed the else...
* Respond to PR feedback
* plumbing request context to expiration manager
* moar context
* address feedback
* only using active context for revoke prefix
* using active context for revoke commands
* cancel tidy on active context
* address feedback
* Add request timeouts in normal request path and to expirations
* Add ability to adjust default max request duration
* Some test fixes
* Ensure tests have defaults set for max request duration
* Add context cancel checking to inmem/file
* Fix tests
* Fix tests
* Set default max request duration to basically infinity for this release for BC
* Address feedback
This change makes it so that if a lease is revoked through user action,
we set the expiration time to now and update pending, just as we do with
tokens. This allows the normal retry logic to apply in these cases as
well, instead of just erroring out immediately. The idea being that once
you tell Vault to revoke something it should keep doing its darndest to
actually make that happen.
* Add an idle timeout for the server
Because tidy operations can be long-running, this also changes all tidy
operations to behave the same operationally (kick off the process, get a
warning back, log errors to server log) and makes them all run in a
goroutine.
This could mean a sort of hard stop if Vault gets sealed because the
function won't have the read lock. This should generally be okay
(running tidy again should pick back up where it left off), but future
work could use cleanup funcs to trigger the functions to stop.
* Fix up tidy test
* Add deadline to cluster connections and an idle timeout to the cluster server, plus add readheader/read timeout to api server
* Store lease times suitable for export in pending
This essentially caches lease information for token lookups, preventing
going to disk over and over.
* Simplify logic
Taking inspiration from
https://github.com/golang/go/issues/17604#issuecomment-256384471
suggests that taking the address of a stack variable for use in atomics
works (at least, the race detector doesn't complain) but is doing it
wrong.
The only other change is a change in Leader() detecting if HA is enabled
to fast-path out. This value never changes after NewCore, so we don't
need to grab the read lock to check it.
* Hand off lease expiration to expiration manager via timers
* Use sync.Map as the cache to track token deletion state
* Add CreateOrFetchRevocationLeaseByToken to hand off token revocation to exp manager
* Update revoke and revoke-self handlers
* Fix tests
* revokeSalted: Move token entry deletion into the deferred func
* Fix test race
* Add blocking lease revocation test
* Remove test log
* Add HandlerFunc on NoopBackend, adjust locks, and add test
* Add sleep to allow for revocations to settle
* Various updates
* Rename some functions and variables to be more clear
* Change step-down and seal to use expmgr for revoke functionality like
during request handling
* Attempt to WAL the token as being invalid as soon as possible so that
further usage will fail even if revocation does not fully complete
* Address feedback
* Return invalid lease on negative TTL
* Revert "Return invalid lease on negative TTL"
This reverts commit a39597ecdc23cf7fc69fe003eef9f10d533551d8.
* Extend sleep on tests
* govet cleanup in token store
* adding general ttl handling to login requests
* consolidating TTL calculation to system view
* deprecate LeaseExtend
* deprecate LeaseExtend
* set the increment to the correct value
* move calculateTTL out of SystemView
* remove unused value
* add back clearing of lease id
* implement core ttl in some backends
* removing increment and issue time from lease options
* adding ttl tests, fixing some compile issue
* adding ttl tests
* fixing some explicit max TTL logic
* fixing up some tests
* removing unneeded test
* off by one errors...
* adding back some logic for bc
* adding period to return on renewal
* tweaking max ttl capping slightly
* use the appropriate precision for ttl calculation
* deprecate proto fields instead of delete
* addressing feedback
* moving TTL handling for backends to core
* mongo is a secret backend not auth
* adding estimated ttl for backends that also manage the expiration time
* set the estimate values before calling the renew request
* moving calculate TTL to framework, revert removal of increment and issue time from logical
* minor edits
* addressing feedback
* address more feedback
* logbridge with hclog and identical output
* Initial search & replace
This compiles, but there is a fair amount of TODO
and commented out code, especially around the
plugin logclient/logserver code.
* strip logbridge
* fix majority of tests
* update logxi aliases
* WIP fixing tests
* more test fixes
* Update test to hclog
* Fix format
* Rename hclog -> log
* WIP making hclog and logxi love each other
* update logger_test.go
* clean up merged comments
* Replace RawLogger interface with a Logger
* Add some logger names
* Replace Trace with Debug
* update builtin logical logging patterns
* Fix build errors
* More log updates
* update log approach in command and builtin
* More log updates
* update helper, http, and logical directories
* Update loggers
* Log updates
* Update logging
* Update logging
* Update logging
* Update logging
* update logging in physical
* prefixing and lowercase
* Update logging
* Move phyisical logging name to server command
* Fix som tests
* address jims feedback so far
* incorporate brians feedback so far
* strip comments
* move vault.go to logging package
* update Debug to Trace
* Update go-plugin deps
* Update logging based on review comments
* Updates from review
* Unvendor logxi
* Remove null_logger.go
* Start work on passing context to backends
* More work on passing context
* Unindent logical system
* Unindent token store
* Unindent passthrough
* Unindent cubbyhole
* Fix tests
* use requestContext in rollback and expiration managers
* Add logic for using Auth.Period when handling auth login/renew requests
* Set auth.TTL if not set in handleLoginRequest
* Always set auth.TTL = te.TTL on handleLoginRequest, check TTL and period against sys values on RenewToken
* Get sysView from le.Path, revert tests
* Add back auth.Policies
* Fix TokenStore tests, add resp warning when capping values
* Use switch for ttl/period check on RenewToken
* Move comments around
* Start work on context aware backends
* Start work on moving the database plugins to gRPC in order to pass context
* Add context to builtin database plugins
* use byte slice instead of string
* Context all the things
* Move proto messages to the dbplugin package
* Add a grpc mechanism for running backend plugins
* Serve the GRPC plugin
* Add backwards compatibility to the database plugins
* Remove backend plugin changes
* Remove backend plugin changes
* Cleanup the transport implementations
* If grpc connection is in an unexpected state restart the plugin
* Fix tests
* Fix tests
* Remove context from the request object, replace it with context.TODO
* Add a test to verify netRPC plugins still work
* Remove unused mapstructure call
* Code review fixes
* Code review fixes
* Code review fixes
* Move location of quit channel closing in exp manager
If it happens after stopping timers any timers firing before all timers
are stopped will still run the revocation function. With plugin
auto-crash-recovery this could end up instantiating a plugin that could
then try to unwrap a token from a nil token store.
This also plumbs in core so that we can grab a read lock during the
operation and check standby/sealed status before running it (after
grabbing the lock).
* Use context instead of checking core values directly
* Use official Go context in a few key places
