For batch operations, collect and annotate the associated span with the set of all operations
contained in the batch.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
For batch operations, collect and annotate the associated span with the set of all operations
contained in the batch.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
The code starting at `ZKUtil.dump(ZKWatcher)` is a small mess – it has cyclic dependencies woven
through itself, `ZKWatcher` and `RecoverableZooKeeper`. It also initializes a static variable in
`ZKUtil` through the factory for `RecoverableZooKeeper` instances. Let's decouple and clean it
up.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Signed-off-by: Josh Elser <elserj@apache.org>
Add support for `db.system`, `db.connection_string`, `db.user`.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Signed-off-by: Huaxiang Sun <huaxiangsun@apache.org>
Co-authored-by: Josh Elser <josh.elser@gmail.com>
* refactors how we handle running the passed in initialization script to make use of IRB sessions
directly instead of reimplementing things ourselves
* simplify how we initialize our IRB config
* insert a shim for capturing exit codes passed via user calls to exit
* make use of user provided exit code unless we're reading stdin in interactive mode
This changes the exit code of the shell
* a 0 return code, or no return code, passed to a call to exit from stdin in non-interactive mode
will now exit cleanly. in prior versions this would have exitted with an error and non-zero exit
code.
* for other combinations of passing in an initilization script or reading from stdin with using the
non-interactive flag, the exit code being 0 or non-0 should now line up with releases prior to
2.4.z, which is a change in behavior compared to 2.4.z.
Signed-off-by: Peter Somogyi <psomogyi@apache.org>
(cherry picked from commit 821e6a36cc)
This refactor reduces the size and scope of the `ZKUtil` class. The core of this refactor is
moving the `login*` methods from `ZKUtil` into their own class, `ZKAuthentication`. The class
`JaasConfiguration` is also moved along with them.
Signed-off-by: Andrew Purtell <apurtell@apache.org>
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Change scan caching back to 1 to avoid extra work at Region Server side. For most cases, the extra 4 results
fetched are wasted at the client side.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
`CallDroppedException` can be thrown with `CallRunner.drop()` by queue implementations that decide to drop calls to groom the RPC call backlog. The LifoCoDel queue does this I believe and with Pluggable queue it's possible for 3rd party queue implementations to be using `drop()` for similar reasons. It would be nice for the server to be tracking these exceptions in metrics since otherwise you might have to do some extra lifting on the client side.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Reviewed-by: Bryan Beaudreault <bbeaudreault@hubspot.com>
RpcServer identifies the services it hosts by unqualified service name. Thus, use `getName()`
instead of `getFullName()`. See also HBASE-26589.
Signed-off-by: Peter Somogyi <psomogyi@apache.org>
This is needed in a couple places in order to test that traces over the IPC layer carry correct
span names, and it's good hygiene anyway.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Follows the guidance outlined in https://github.com/open-telemetry/opentelemetry-specification/blob/3e380e2/specification/trace/semantic_conventions/database.dm
* all table data operations are assumed to be of type CLIENT
* populate `db.name` and `db.operation` attributes
* name table data operation spans as `db.operation` `db.name`:`db.hbase.table`
note: this implementation deviates from the recommended `db.name`.`db.sql.table` and instead
uses HBase's native String representation of namespace:tablename.
Signed-off-by: Duo Zhang <zhangduo@apache.org>
Signed-off-by: Tak Lon (Stephen) Wu <taklwu@apache.org>