* better check for granule-ification
* Handling blob granule initial split too large
* Re-evaluating split size if too large, even if read doesn't get transaction_too_old
* reworked to have blob worker propose split key
* New GranuleStatusReply to avoid seqno issue stream side effects
* Handling retries on reevaluateInitialSplit properly
* Waiting for stream to be initialized
* Checking reevaluate split for additional split points beyond proposed
* Fixing more races in reevaluate initial split
* properly handling cleaning up old change feed after split re-evaluate
* fixing granule conversion bug with hard boundaries
* fixing clear and merge check race with cycle test
* refactor missed knob check for clearAndMerge
* Fixing formatting
* review comments and improving large range conversion
* fixing typo
* more formatting
* Fixed ChangeServerKeysContext name issue.
* Update fdbserver/storageserver.actor.cpp
Co-authored-by: Andrew Noyes <andrew.noyes@snowflake.com>
Co-authored-by: He Liu <heliu@apple.com>
Co-authored-by: Andrew Noyes <andrew.noyes@snowflake.com>
Issue #7258
The ratekeeper is recording the busiest write tag for *all* storage
servers, which throttles the traceevent. Distribute the busiest write
tag to corresponding storage servers should reduces this throttling
issue.
* Disabled tests for ShardedRocks.
Cleaned up ShardedRocks TraceEvent.
Added assertion in ShardManager::validate().
* Added test trace.
* Make sure TraceEvent contains `ShardedRocks`.
* Exclude ShardedRocksDB when SHARD_ENCODE_LOCATION_METADATA is disabled.
Co-authored-by: He Liu <heliu@apple.com>
* 'main' of github.com:sfc-gh-nwijetunga/foundationdb: (42 commits)
Get ShardedRocks ready for simulation test. (#7679)
fixing specific unit test
formatting
addressing review comments
Fix incorrect deserialization of FdbClientLogEvents::Event (#7707)
Fix a crash bug during CC shutdown process (#7705)
addressing review comments
remove runAfter
Add comments explaining the use of the TransactionState tenant() and hasTenant() functions
Make sure resumeFromDataMoves() starts after resumeFromShards().
Fix: during recovery, it was possible for tenant operations to briefly fail because the tenant mode is not known
formatting
Fixed granule purging bug and improved debugging for purging
making purge failures fail test
cleanup and polish
Bug fix and cleanup
First version of key-sorted delta files
Added full granule read unit test
Completed delta format unit test
delta file test and delta generation
...
fdbserver/storageserver.actor.cpp:654:24: error: cannot pass object of non-trivial type 'const std::string' (aka 'const basic_string<char, char_traits<char>, allocator<char>>') through variadic function; call will abort at runtime [-Wnon-pod-varargs]
Just use fmt instead
* Storage server shard management with physical shards.
* Cleanup.
* Resolved comments.
* Added `UnlimintedCommitBytes`.
Co-authored-by: He Liu <heliu@apple.com>
* 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
Josh Slocum