* Throttle the cluster if the blob manager cannot assign ranges
* fixed a number of different bugs which caused ratekeeper to throttle to zero because of blob worker lag
* fix: do not mark an assignment as block if it is cancelled
* remove asserts to merge bug fixes
* fix formatting
* restored old control flow to storage updater
* storage updater did not throw errors
* disable buggify to see if it fixes CI
* Cleaned up BlobGranule TODO + FIXMEs and addressed some
* popping feed at correct version
* blob worker taking over a granule will pop from where previous worker left off
* addressed fixme of blob worker not re-snapshotting from old change feed
* formatting
* more change feed popped fixes after pop updates
* Getting rid of change feed parallelism lock since it can cause deadlocks in fetching, and relying on full fetch lock
* New blob worker metric and fixing old one
* server-side popped checking still doesn't work because of pops at non-mutation versions
* format
* Add the verify option for \xff\xff/worker_interfaces
* Remove unused code
* update documentations
* update documentations
* solve comments from review
* update some of the comments to be more clear
* throttle the cluster when blob workers fall behind
* do not throttle on blob workers if they are not enabled
* remove an unnecessary actor
* fixed a compile error
* fetch blob worker metrics at the same interval as the rate is updated, avoid fetching the complete blob worker list too frequently
* fixed another compilation bug
* added a 5 second delay before bw throttling to prevent false positives caused by the 100e6 version jump during recovery. Lower the throttling thresholds to react much quicker to bw lag.
* fixed a number of problems
* changed the minBlobVersionRequest to look at storage server versions since this will be a lot more efficient
* fix: do not let desired go backwards
* fix: track the version of notAtLatest changefeeds for throttling
* ratekeeper now throttled blob workers by estimating the transaction per second throughput of the blob workers
* added metrics for blob worker change feeds
* added a knob to disable bw throttling
* fixed the transaction options in blob manager
Currently GRV is reporting proxy_memory_limit_exceeded error which has
error message claiming Commit proxy failing. This split should remove
such confusion.
* Granule purge cannot delete history entry for fully deleting granule until all children are completely done splitting
* Several purging fixes related to granule history
* Fixed typo in refactor
* fixing memory model for purgeRange
* formatting
* weakening granule purge test for now
* cleanup
* First version of force purging granules
* fixing issue in BW range assignment reporting
* Fixing incorrect assert with force purging
* Error handling when checking force purged state
* fixed force purging and recover/reassign range races and check
* Handling force purge + boundary change race
* more places to check for force purged status
* fixed manager restart in the middle of force purge bug
* fixing same-BM purge and assignment races in all cases
* weakening orphaned granule history check a bit because of difficult to solve races
* fixing txn options on retry
* loading force purged ranges at start to avoid resuming a merge that is being force purged
* cleanup
* Enabling purging in granule tests, and adding check for leaked change feeds in force purge
* formatting
* missed parameter in merge conflicts
* Fixing leaked change feed race with merge and force purge
* adding change feed cleanup when new blob manager recovers in-progress merge that raced with force purge
* added forcepurge fdbcli command
* fixing another actor cancelled callback issue in change feed stream
* remove wait for on move, to ensure stream can't read from a shard the SS no longer owns
* fixing key alignment after downsampling split
* Fixing out-of-date granule reporting itself as a merge candidate
* fix formatting
* Enable configuring the next future protocol version as the current protocol version in FDB client, fdbserver, and fdbcli
* Auto format python files used in upgrade tests
* Add a test for upgrading to a future FDB version
* Emphasize that the options for using future protocol version are intended for test purposes only
* Make the global variable for current protocol version visible only locally
* Refactirng to avoid using currentProtocolVersion() in static intialization
* Update go bindings
* Track when version not found in version vector.
* Respond to comments
* include vv in event
* Respond to review comments.
* Fix bug where we logged too many lines
Dummy transaction commits are made when earlier transaction returned
commit_unknown_result or request_maybe_delivered, to affirm the status
of earlier transaction. Without tenant-awareness, these transactions
will fail due to lack of authorization in case of untrusted clients.
* proof of concept
* use code-probe instead of test
* code probe working on gcc
* code probe implemented
* renamed TestProbe to CodeProbe
* fixed refactoring typo
* support filtered output
* print probes at end of simulation
* fix missed probes print
* fix deduplication
* Fix refactoring issues
* revert bad refactor
* make sure file paths are relative
* fix more wrong refactor changes
+ Make Authz tokens survive Transaction resets
+ Make tokens explicitly resettable by assigning empty value to authz token TR option
+ Fill out missing trace for some failed token verifications
+ Trace non-sensitive token parts upon failed verification
* Shard based move.
* Clean up.
* Clear results on retry in getInitialDataDistribution.
* Remove assertion on SHARD_ENCODE_LOCATION_METADATA for compatibility.
* Resolved comments.
Co-authored-by: He Liu <heliu@apple.com>
* Add an internal C API to support memory connection records
* Track shared state in the client using a unique and immutable cluster ID from the cluster
* Add missing code to store the clusterId in the database state object
* Update some arguments to pass by const&
* Fix a few places we weren't doing exponential backoff
We re-create the transaction every iteration of each of these retry
loops, so we need to manage exponential backoff here ourselves.
Closes#7301
* Remove former Backoff definition
also in order to continue tracing the pending network thread activity.
Poll event throttler only in the network thread in order to avoid a race condition.
* Fixing simulation validation assert that was tripping incorrectly
* Commenting out debugging prints
* Fixed multiple error propagation issues in blob worker
Only fetch requested worker interfaces in rebootWorkerActor
Add killall test to multiprocess_fdbcli_tests
Make the rebootWorker only sends requests if all the requested workers are valid
Add comments; Organize print texts; Fix the bug clear the map every time run kill/suspend/expensive_data_check
* Enable MATCHED and UNMATCHED mode for index prefetch
MATCHED mode returns index entries whose secondary KVs are present,
UNMATCHED mode returns index entries whose secondary KVs are absent.
Note that the conflict read range of this txn is set in 2 steps:
* Set the conflict range for primary query according to request
* Set the conflict ranges for secondary queries according to responses.
As a result, conflicts of different match_index mode are taken care of.
* Fix c binding
Currently, GlobalConfig is a singleton that means for each process there is only
one GlobalConfig object. This is bug from clients perspective as a client can
keep connections to several databases. This patch tracks GlobalConfig for each
database using an unordered_map in flowGlobals.
We discovered this bug while testing multi-version client, where the client got
stuck. This was lucky, as normally it'd just write down config to the wrong
database.
* Fixing leaked stream with explicit notify failed before destructor
* better logic to prevent races in change feed fetching
* Found new race that makes assert incorrect
* handle server overloaded in initial read from fdb
* Handling more blob error types in granule retry
* Fixing rollback metadata problem, added better debugging
* Fixing version race when fetching change feed metadata
* Better racing split request handling
* fixing assert
* Handle change feed popped check in the blob worker
* fix: do not use a RYW transaction for a versionstamp because of randomize API version (#6768)
* more merge conflict issues
* Change feed destroy fixes
* Fixing change feed destroy and move race
* Check error condition in BG file req
* Using relative endpoints for blob worker interface
* Fixing bug in previous fix
* More destroy and move race fixes
* Don't update empty version on destroy in case it gets rolled back. moved() and removing will take care of ensuring it is not read
* Bug fix (#6796)
* fix: do not use a RYW transaction for a versionstamp because of randomize API version
* fix: if the initialSnapshotVersion was pruned, granule history was incorrect
* added a way to compress null bytes in printable()
* Fixing durability issue with moving and destroying change feeds
* Adding fix for not fully deleting files for a granule that child granules need to re-snapshot
* More destroy and move races
* Fixing change feed destroy and pop races
* Renaming bg prune to purge, and adding a C api and unit test for it
* more cleanup
* review comments
* Observability for granule purging
* better handling for change feed not registered
* Fixed purging bugs (#6815)
* fix: do not use a RYW transaction for a versionstamp because of randomize API version
* fix: if the initialSnapshotVersion was pruned, granule history was incorrect
* added a way to compress null bytes in printable()
* fixed a few purging bugs
Co-authored-by: Evan Tschannen <evan.tschannen@snowflake.com>
* Initialize cluster version at wall-clock time
Previously, new clusters would begin at version 0. After this change,
clusters will initialize at a version matching wall-clock time. Instead
of using the Unix epoch (or Windows epoch), FDB clusters will use a new
epoch, defaulting to January 1, 2010, 01:00:00+00:00. In the future,
this base epoch will be modifiable through fdbcli, allowing
administrators to advance the cluster version.
Basing the version off of time allows different FDB clusters to share
data without running into version issues.
* Send version epoch to master
* Cleanup
* Update fdbserver/storageserver.actor.cpp
Co-authored-by: A.J. Beamon <aj.beamon@snowflake.com>
* Jump directly to expected version if possible
* Fix initial version issue on storage servers
* Add random recovery offset to start version in simulation
* Type fixes
* Disable reference time by default
Enable on a cluster using the fdbcli command `versionepoch add 0`.
* Use correct recoveryTransactionVersion when recovering
* Allow version epoch to be adjusted forwards (to decrease the version)
* Set version epoch in simulation
* Add quiet database check to ensure small version offset
* Fix initial version issue on storage servers
* Disable reference time by default
Enable on a cluster using the fdbcli command `versionepoch add 0`.
* Add fdbcli command to read/write version epoch
* Cause recovery when version epoch is set
* Handle optional version epoch key
* Add ability to clear the version epoch
This causes version advancement to revert to the old methodology whereas
versions attempt to advance by about a million versions per second,
instead of trying to match the clock.
* Update transaction access
* Modify version epoch to use microseconds instead of seconds
* Modify fdbcli version target API
Move commands from `versionepoch` to `targetversion` top level command.
* Add fdbcli tests for
* Temporarily disable targetversion cli tests
* Fix version epoch fetch issue
* Fix Arena issue
* Reduce max version jump in simulation to 1,000,000
* Rework fdbcli API
It now requires two commands to fully switch a cluster to using the
version epoch. First, enable the version epoch with `versionepoch
enable` or `versionepoch set <versionepoch>`. At this point, versions
will be given out at a faster or slower rate in an attempt to reach the
expected version. Then, run `versionepoch commit` to perform a one time
jump to the expected version. This is essentially irreversible.
* Temporarily disable old targetversion tests
* Cleanup
* Move version epoch buggify to sequencer
This will cause some issues with the QuietDatabase check for the version
offset - namely, it won't do anything, since the version epoch is not
being written to the txnStateStore in simulation. This will get fixed in
the future.
Co-authored-by: A.J. Beamon <aj.beamon@snowflake.com>
Each GetReadVersionReply now has the GRV proxy ID that sent it so that the
client can detect old GRV proxy's replies and avoid apply stale version vector
changes.
The events that led to the assertion failure are:
Many txn started, some get ReadVersion1, some get ReadVersion2.
Recovery happened.
GRV proxy change detected and the client version cache is cleared.
Old GRV responded with ReadVersion1, and client version cache is updated to ReadVersion1.
A txn starts reading with ReadVersion2, triggering the assertion.
The fix is to throw an error to trigger transaction restart so that a new GRV
will update the version cache.
Evan Tschannen