Age | Commit message (Collapse) | Author | Files | Lines |
|
Convert python2 scripts to python3
Issue-ID: CCSDK-3906
Signed-off-by: Dan Timoney <dtimoney@att.com>
Change-Id: I68b82108c5cd51a99735f65f964b8ebc16181c1b
|
|
Issue-ID: CCSDK-3698
Signed-off-by: Lukasz Rajewski <lukasz.rajewski@t-mobile.pl>
Change-Id: Ie5c78a9a347373b5faf588627e42138806d69c0c
|
|
Issue-ID: CCSDK-3684
prepareEnv (mistakenly) had a recursive call, but no decement.
Simplified it in favour of the inner/outter flag instead of retry counter as in Frankfurt release.
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: I5d7f623d6e206a844d33d95a15f85758da3a6681
|
|
Issue-ID: CCSDK-3531
Improve end-user CBA's Python scripts ability to return
error messages on failues.
See JIRA story for examples: https://jira.onap.org/browse/CCSDK-3531
adds functions send_response_err_msg(err_message)
and send_response_err_msg_and_exit(ret_err_msg, code=1)
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ideda64dd108f987c3c4515c70daf9f7b6d2b1f2f
|
|
Earlier patch introducing MeterRegistry to AbstractComponentFunction
was not initializing the lateinit property and causing NPE.
Also add additional code to handle compatibility issues after the
introduction of PV/PVC elminiation for CommandExecutor. This allows
blueprintsprocessor to communicate with earlier versions of command-
executor which still use the shared pvc.
Issue-ID: CCSDK-3471
Change-Id: I84a04601c4fe09c5f3a06664ce877800a30531f1
Signed-off-by: Jozsef Csongvai <jozsef.csongvai@bell.ca>
|
|
Issue-ID: CCSDK-3471
Signed-off-by: Michael Pruss <michprus@amdocs.com>
Change-Id: I871a032ff5c9ebfd408aaa28a7d24185dee7c953
|
|
The renaming in CCSDK-3098 caused breaking changes to the grpc api and
compile issues for kotlin scripts.
Issue-ID: CCSDK-3385
Change-Id: I0d745cb858371678eabcb2284671c1fd76a1ab6d
Signed-off-by: Jozsef Csongvai <jozsef.csongvai@bell.ca>
|
|
executeCommand was called from a GlobalScope.async coroutine to handle executeCommand timeout.
Execution timeout is already handled on the gRPC side, there's no need to handle this timeout by wrapping the gRPC call with a coroutine timeout.
When a lot of long running processes in executeCommand were running, it was causing process to queue on BP side to get their executeCommand executed because all the capacity of couroutines were being used by some long running process.
This was causing a delay on the execution of prepareEnv and executeCommand and would even some time make those process timeout because of component timeout.
Issue-ID: CCSDK-3386
Signed-off-by: Julien Fontaine <julien.fontaine@bell.ca>
Change-Id: If3aef4b6b65006e874525436bf3f6d1b1a3d8c9a
|
|
|
|
Issue-ID: CCSDK-3086
commit 1: orig commit
commit 2: update retry counter
commit 3: bug fix for packages being null should still trigger creation of python environment
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Id9e7024df8c604c719828211cce31d6f0452330b
|
|
BlueprintError needs to associate errors with the steps in which they occurred
in order for imperative workflow to handle on_failure properly.
Made stepName more accessible and corrected places where stepName was assigned to nodeTemplateName.
Issue-ID: CCSDK-3219
Change-Id: I7e5805745c63558cff6be533e1b99c32ad06c3db
Signed-off-by: Jozsef Csongvai <jozsef.csongvai@bell.ca>
|
|
Issue-ID: CCSDK-3086
1. initial commit
2. fix accidental paste / rebased
cleaned up unneeded validation call in cmd-exec upload
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ife5460c5be59aa8d8592d82099b27c507b08c6c6
|
|
Replacing BluePrint with Blueprint throughout
Issue-ID: CCSDK-3098
Signed-off-by: KAPIL SINGAL <ks220y@att.com>
Change-Id: Ibee8bad07ae7d9287073db2d4f2f2cd730fa8b96
|
|
Code Formatter was turned off due to java 11 migation
Issue-ID: CCSDK-2852
Signed-off-by: Singal, Kapil (ks220y) <ks220y@att.com>
Change-Id: I3d02ed3cc7a93d7551fe25356512cfe8db1517d8
|
|
Adding parent pom to cba-assembly-descriptor as it needs to inherit repository from parent so maven-deploy-plugin works
Issue-ID: CCSDK-2720
Signed-off-by: Singal, Kapil (ks220y) <ks220y@att.com>
Change-Id: I44ce3721f280555b81475580ca47539ae83d07ca
|
|
* Update gRPC interface to handle SubRequestID and Originator ID
* Update command-executor log format to handle request IDs in the log statement
* Add SubRequest ID and Originator ID in BP log statement
Issue-ID: CCSDK-2687
Change-Id: If261bb0912df5800ed2fc65371c5ee784d6a45fb
Signed-off-by: Julien Fontaine <julien.fontaine@bell.ca>
|
|
Issue-ID: CCSDK-2603
* execute-command-logs" should always be a list (in some cases,
we had a string causing possible parsing problems)
* response-data should be JSON and "N/A" not acceptible.
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ie6fceb6e1f4c86bfcdeb7671bacd74b909628096
|
|
BlueprintJythonServiceTest started failing in Jenkins for unknown
reasons. Seeing as jython is deprecated and should be removed, its
not worth tracing the cause.
Issue-ID: CCSDK-2602
Signed-off-by: Jozsef Csongvai <jozsef.csongvai@bell.ca>
Change-Id: I495409a18fc324e00c373c99bcc6da803666240a
|
|
Issue-ID: CCSDK-2549
fixes:
1. CMD-exec returns payload as JSON object.
There was some regression where we overwrite returned JSON with errMsg.
2. adds 100ms delta to executionTimeout, as was in our downstream version
(want to guarantee that we get grpc timeout exception on BP side
rather than coroutine w/o waiting extra)
3. setNodeOutputErrors should not have ability to set execution status
to success.
rev1. initial import
rev2. reworked the calls to setNodeOutputErrors: addError was brought out of it
as we can set the node execution error based on prepare.env or execution.
rev3. did not mean to add BlueprintRuntimeService change
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: I6d509df5ae51598f33ab9f0ea53806d653cf79c0
|
|
Issue-ID: CCSDK-2535
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: I897678a5a8a23503a878f2d3eb836ba4597a6e6e
|
|
* Modified command exec returned value in case of failure during execution. It now prints the response_data defined by the user
* Modified truncation method of the gRPC returned object to use ByteSize() to get the exact sizxe consumed within the buffer
Issue-ID: CCSDK-2501
Signed-off-by: Julien Fontaine <julien.fontaine@bell.ca>
Change-Id: Ie1db8db265623b5137ab3946ff4e3abda1c54a78
|
|
Issue-ID: CCSDK-2445
In certain cases, status message is returned on error instead of status=FAILURE.
rev 2: reuse same msg
rev 3: past elalto, compoment level timeout is implementation.timeout
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: I2651e5706b7e62894c5e071dc1563fb1b9dd67e1
|
|
rev1. initial commit
rev2. spacing (silly klint!)
Issue-ID: CCSDK-2445
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ie8cc62c89e212efbf5240304856b79c173c88abb
|
|
Refactoring of cmd-exec component
- Improve display of error messages within the response
- Fix gRPC buffer limitation (4Mb) by truncating error messages and cmd-exec logs if too heavy (>3Mb)
Truncation of BP responses (<4Kb) before sending them in kafka audit topics.
- Truncation if needed of error messages for every response
- Truncation of cmd-exec logs in cmd-exec responses
(Spike) Add a flag in the application.properties to enable/disable the display of cmd-exec responses on the BP side
(Fix) Correction of BP processing with kafka regression
(Fix) Changed default SSL Endpoint Algo
Issue-ID: CCSDK-2326
Change-Id: If4d0e661117d1dd156cf19c95774824e754d870a
Signed-off-by: Julien Fontaine <julien.fontaine@bell.ca>
|
|
|
|
Issue-ID: CCSDK-2049
rev1. initial commit
rev2. added better logging/passing errors.
rev3. bugfixes
rev4. bugfix for CommandExecutorServer.prepareEnv return status was still returning bool.
(was caught only during a defective mariadb pip package install!)
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ifbddaeb0e15857f42b121af10e30e29dec5c20b2
|
|
|
|
Issue-ID: CCSDK-2124
Signed-off-by: Steve Siani <alphonse.steve.siani.djissitchi@ibm.com>
Change-Id: Ief468a56f9e7b3ef86c357965aa7b15f0a4cfa22
|
|
Issue-ID: CCSDK-2012
rev1: initial commit
rev2: reworked + added grpc deadline
rev3: wrong err msg..
rev4: timeouts in millis
rev5: timeout defaulting was after logging...
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: I02eb414904cc5dfcf51ff7b7552dafe54857ed1e
(cherry picked from commit 3f56fcfec7d14e7351280b9e2a1d220638626693)
|
|
Add "babel/compat-data" package to Designer Client devDependencies
Remove package-lock.json from .gitignore
Issue-ID: CCSDK-2192
Signed-off-by: Michal Jagiello <michal.jagiello@t-mobile.pl>
Change-Id: I2676412dc4994fa5314cd411ce64d19df2790e5a
|
|
Added and improved blueprint runtime service property resolution api.
Relationship template CBA sample
Performance improvement on Artifact resolutions.
Added connection config DSL for DB, GRPC connections.
Added starter relationship connection config models.
Issue-ID: CCSDK-1054
Signed-off-by: Brinda Santh <bs2796@att.com>
Change-Id: I73e1deed708dc924352867ddd1d1b2271c8ea106
|
|
Mandate Tosca.meta template name, version, type and tags.
Auto copy metadata from Tosca.meta to ServiceTemplate definitions.
Optimize Blueprint context and runtime creation from file path.
Removed attached CBA zip file in test repository dirs
Issue-ID: CCSDK-1992
Signed-off-by: Brinda Santh <bs2796@att.com>
Change-Id: I5d9d7a4599234a38d431328dbd9b74bd831e0115
|
|
Refactored cds python capabilities to python-executor function, so that it will easy to disable and deprecation
This change is not jython depreciation, just separating out
Issue-ID: CCSDK-1865
Signed-off-by: Brinda Santh <bs2796@att.com>
Change-Id: I3727d72d7195c49d670c30a6fc4245f5a8599e3a
|
|
No Business logic change, just the code format.
Competible with IntelliJ: https://github.com/pinterest/ktlint#option-3
To format run: mvn process-sources -P format
Issue-ID: CCSDK-1947
Signed-off-by: Singal, Kapil (ks220y) <ks220y@att.com>
Change-Id: Ic9e9209fb7023d77f434693ad5a01229f8d09331
|
|
Also ensure that if environment preparation does not succeed the processing fails.
Issue-ID: CCSDK-1855
Signed-off-by: Serge Simard <serge@agilitae.com>
Change-Id: Iab2ec8b44a0e9e7edef68d0d5c13860eedc5be67
Signed-off-by: Serge Simard <serge@agilitae.com>
|
|
Issue-ID: CCSDK-1855
Signed-off-by: Serge Simard <serge@agilitae.com>
Change-Id: I924918fb06a5a0f118772dedaa41c536c7eca9e7
Signed-off-by: Serge Simard <serge@agilitae.com>
|
|
No Business logic change, no code change.. just optimizing imports and reformatting code
Issue-ID: CCSDK-1744
Signed-off-by: Singal, Kapil (ks220y) <ks220y@att.com>
Change-Id: I1fb2ed10525a6f392f3c0d963c2ed03cfe8dec7e
|
|
Appending New Line before and after request body within doWrapRpc and removing new line is child API's which forms the request body
Issue-ID: CCSDK-1743
Signed-off-by: Singal, Kapil (ks220y) <ks220y@att.com>
Change-Id: Ic58566e24a21c71d545c45fb95a667c17de2fe66
|
|
Changed Line Separator for all files using find . -type f -not -name .* -exec dos2unix {} \;
Issue-ID: CCSDK-1733
Signed-off-by: SINGAL, KAPIL (ks220y) <ks220y@att.com>
Change-Id: I2efa3276c35c0a0159c6d4debe21c0bd0b4a7b85
|
|
Change-Id: If68ce2e944700cb19380d34ff3b466a9c8633472
Issue-ID: CCSDK-1577
Signed-off-by: Brinda Santh <brindasanth@in.ibm.com>
|
|
Change-Id: I83a3d1391c60cd426236b55d0d399e273e7c15dc
Issue-ID: CCSDK-1577
Signed-off-by: Brinda Santh <brindasanth@in.ibm.com>
|
|
Change-Id: I7888ac302da2a51ca34c531b7553d04af13786ac
Issue-ID: CCSDK-1380
Signed-off-by: Brinda Santh <brindasanth@in.ibm.com>
|
|
Change-Id: I193e2a092ff8cefe48419031096e1336e5c84fc1
Issue-ID: CCSDK-1046
Signed-off-by: Brinda Santh <brindasanth@in.ibm.com>
|
|
Change-Id: I77a60de87acec862ce47256557a1223fefe82a12
Issue-ID: CCSDK-1497
Signed-off-by: Alexis de Talhouët <adetalhouet89@gmail.com>
|
|
Issue-ID: CCSDK-1468
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: If8eb50fc2e8dfa084e486b02ace163eee1bfac77
|
|
|
|
|
|
There are three existing ways to perform the resolution:
either we don't store the results at all, whether for resource or template
either we store using the resolution key. The combination of blueprintName,
blueprintVersion, artifactName and resolutionKey has to be unique. If it
is re-used, it is considered as a new attempt for that specific resolution
request, and process will only try to resolve resources not marked at SUCCESS
in the database.
either we store using the resourceId and resourceType. As previous point, the
combination of blueprintName, blueprintVersion, artifactName and resolutionKey
has to be unique. If it is re-used, it is considered as a new attempt for that
specific resolution request, and process will only try to resolve resources not
marked at SUCCESS in the database.
TBD: add uni tests
Issue-ID: CCSDK-1423
Change-Id: I6b7198453cf0002edfa7a0c9ea3179555211b5dc
Signed-off-by: Alexis de Talhouët <adetalhouet89@gmail.com>
|
|
* python-executor STDOUT contains cleaner logs
* return status has been reworked to return a list
of lines
Issue-ID: CCSDK-1404
Signed-off-by: Oleg Mitsura <oleg.mitsura@amdocs.com>
Change-Id: Ib17495e92a136ff4f950bb6e049246a9e100eb7d
|
|
Issue-ID: CCSDK-1426
Signed-off-by: Serge Simard <serge@agilitae.com>
Change-Id: I919e30b76bb472b4e0aeeb2294f0b3087a655efd
|