json.UnmarshalError happens only when we pass a non-nil pointer, which
is not the case for us. Usage of go vet helps us make sure it also
won't be happening in the future.
- use kraft instead of zookeeper
- add handlers in case of cfg changes
- create a separate user for the kafka service
- bump kafka version
- make the role more configurable
Previously, msg.Ack(true) was used, which is mostly used for batch
processing. It mistakenly acknowledged all previously unacknowledged
messages, causing errors from the AMQP server. Now, messages are
acknowledged individually after each one is processed.
Messages that ERs failed to process are now rejected and requeued
for future processing attempts.
The reader is now closed immediately if the message delivery
channel closes. Therefore, it prevents an endless loop by avoiding
continuous consumption from empty or closed channels.
Addresses: #4146
HierarchyPath parser now returns nil when
the path is empty (instead of a string slice with one
EmptyString element).
If the prefix is set to true, when calling the AsString
method on a nil HierarchyPath, only the separator will
be returned. This avoids a nil expr error coming from
the xmlquery library.
Use the Query and QueryAll functions from the xmlquery
package to be able to handle the errors ourselves and
avoid panics.
Remove config default value for xml_root_path. The field
will remain commented in config_defaults for reference.
Add tests for HierarchyPath.AsString function.
Add comments for XmlProvider and xml_root_path opt.
Added in-depth comments to the constructor.
Added test cases for most edge cases (if not all).
Updated the test to verify not only if an error occurs, but to also
check whether it matches the one we expect.
Added 3 more sanity checks that were missed.
Replaced the 'nightly' distribution configuration with 'master'. This
change may trigger an unused database error. Resolve it by either
running `reprepro clearvanished` (removing 'nightly' completely) or
re-adding and maintaining the 'nightly' configuration alongside
'master'.
Updated ees/ers implementation to use the jetstream package which
separates the jetstream context from Core NATS.
Removed the jsOpts fields from the NatsEE struct. We are now using the
jetStreamMaxWait option directly through a timeout context.
Added streamName option for NATS reader since it is now required to be
specified when creating a consumer (it is not inferred based on subject
anymore).
Updated nats ers integration test.
Removed deprecated birpc unit test.
Updated tests to also use the new jetstream package.
Updated tests to start the nats-server using their official driver
instead of using the std go exec package. time.Sleeps are now not
required anymore to wait for the server.
In test configurations for nats readers, made sure that natsStreamName
option is populated. It is now required for consumers to know where to
subscribe.
Fixed potential panic that happened when jetstreamMaxWait option would
not be set.
Upgraded NATS server version in ansible role.
upgraded go.mod nats version due to an issue caused by version mismatch
between driver and server (uncertain).
Increased amount of sleep in tests after starting a NATS server with
jetstream enabled from 50ms to 100ms.
## *NatsER.Serve
- Replaced ChanQueueSubscribe with QueueSubscribe for Core NATS consumer
to handle the message processing directly.
- Since QueueSubscribe is now used regardless of jetstream status, the
message handler has been assigned to a separate variable that can be
reused.
- The message handler is now dealing with the message processing
directly, therefore the select case listening for the channel which is
feeding NATS messages can be removed together with the channel itself
and the select. Currently, the goroutine within Serve only has to block
until the rdrExit chan is closed.
- Moved the resource check inside the handler right before starting the
message processing goroutine.
## ers.getProcessedOptions
- Renamed function from getProcessOptions to getProcessedOptions.
- Initially, the EventExporterOpts struct was always initialized to be
non-nil which meant exporting processed messages by the reader was
always enabled by force. Function has been updated to initialize it
only once when the first Processed option that was set is found.
- Created init function for all types of opts structs to avoid
repetition.
## *NatsEE.parseOpts
- Renamed function from parseOpt to parseOpts.
- Updated function to return early in case of nil opts struct to reduce
nesting.
- The nested jetstream status and maxwait conditions nil verification
have been merged into one condition.
- Handled the error coming from GetNatsOpts function.
- NATS Subject assignment has been removed. It was redundant, since it
had already been set from before this function was called.
## *NatsEE.Connect
- Updated function to return early in case of non-nil nats.Conn value
to reduce nesting.
## *NatsEE.ExportEvent
- Use defer to release resources and RUnlock.
## *NatsEE.Close
- Use defer to Unlock.
- Update function to return early in case of nil nats.Conn value to
reduce nesting.
## ees.GetNatsOpts
- Chose switch over if else when parsing client certificate and keys
opts.
- Updated function to return the errors directly instead of assigning
them to a separate variable right before returning.
## ers.GetNatsOpts
- Passed the NATSROpts struct directly to the function.
- Chose switch over if else when parsing client certificate and keys
opts.
- Updated function to return the errors directly instead of assigning
them to a separate variable right before returning.
Removed tab from commented natsJetStreamMaxWaitProcessed option
value in config_defaults.go under ers section.
Removed redundant NATS prefix from config.NATSROpts type field names.
Added function in ers/lib_test.go to create a config file in /tmp
based on a config string.
Added integration test for ERs NATS.
