Age | Commit message (Collapse) | Author | Files | Lines |
|
The Pdp Policy Status is not updated during the pdp heartbeat
handling/registration flow, or during State change.
When a PDP goes down the pdppolicystatus gets cleared. But when the PDP
comes back online, PAP deploys the right policies into it, but the
deployment status was not getting updated.
Similarly, when the State is changed to PASSIVE, the status should be
undeployed, and when the state is changed back to ACTIVE, then the
status should be deployed for any active policies in any PDP.
These issues are fixed here.
Updating postman collection.
Also updating the csit test to check for 202 instead of 200 for the
deployment api response.
Change-Id: I1e05f22df23daaa8da8be1376413e995be2376f3
Issue-ID: POLICY-2526
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
Change-Id: I5281937fe5a742916c0c86983f71e280bbe126b2
Issue-ID: POLICY-2526
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
The dependency upgrades in policy-parent caused some new sonars. In
particular, initMocks() has been deprecated - replaced with calls to the
Mockito Runner and one call to openMocks().
Issue-ID: POLICY-2911
Change-Id: I6f28af9af41a1e5c83dead2eabf580731ffb9526
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Addressed the following issues:
- don't use eq() with verify()
Issue-ID: POLICY-2911
Change-Id: Idf08195e4e2fb74e5e0d0d30f31b9e630720f48f
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
The classes PolicyTypeFilter and PolicyFilter only work on Policy types
and policies. However they just filter on name/version/type/typeversion
so there is no reason that the filters could not be applied to other
entities.
This commit adapts the PolicyTypeFilter to EntityFilter and PolicyFilter
to TypedEntityFilter, so the filters can be applied to collections of
any ToscaEntity class.
Issue-ID: POLICY-2995
Change-Id: I7cc0ffa23fb12ab5b6e7617491dccd6f3289fa22
Signed-off-by: liamfallon <liam.fallon@est.tech>
|
|
Removed old code that tracked the status using a cache. Modified the
code to use the status found in the DB instead.
Issue-ID: POLICY-2648
Change-Id: I4f1f5201eff7b3352631f50d67c40349ac7f2014
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Issue-ID: POLICY-2648
Change-Id: Icbb9545e3df6942e6f9cf98689607c461f6c8cdd
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
When PAP is made stateless, a new mechanism will be needed to track and
generate notifications. Added some classes to facilitate that.
Issue-ID: POLICY-2648
Change-Id: Ib7b707f68a557e7b306dfdd1c6e6e9abd4671ec1
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
The policy models tosca classes ToscaPolicyIdentifier and
ToscaPolicyIdentifierOptVersion can be used to identify any TOSCA
concept, not just TOSCA policies so they are renamed to
ToscaConceptIdentifier and ToscaCinceptIdentifierOptVersion
respectively.
The class ToscaPolicyTypeIdentifier is redundant and is replaced by
ToscaConceptIdentifier.
Issue-ID: POLICY-2900
Change-Id: I22bb3fdbaf2e237221a5c89afbf7898410603835
Signed-off-by: liamfallon <liam.fallon@est.tech>
|
|
Issue-ID: POLICY-2900
Change-Id: I986074f69bef24ca894adc38eaf8b3ca4b4dc5a2
Signed-off-by: liamfallon <liam.fallon@est.tech>
|
|
Change-Id: I0f81906098c310aa5c437a2c06b708ae27560af8
Issue-ID: POLICY-2911
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
When we upgrade to oparent, these will fail as oparent upgrades
to 8.37 of puppycrawl which fixes a few bugs that should have
flagged these previously.
Issue-ID: POLICY-2887
Change-Id: I7c04f58c683d6eb0322232790e0561b464c1696a
Signed-off-by: Pamela Dragosh <pdragosh@research.att.com>
|
|
Issue-ID: POLICY-2875
Change-Id: Ibad95f08cb39adc285f9326dd1d590163ede5803
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Issue-ID: POLICY-2869
Change-Id: I3696c9c6cae59124f0269f1d2619b22b134b882c
Signed-off-by: Pamela Dragosh <pdragosh@research.att.com>
|
|
Issue-ID: POLICY-2839
Change-Id: I37704935946a8c249e814bf4e515178528705cd6
Signed-off-by: Pamela Dragosh <pdragosh@research.att.com>
|
|
The legacy operational policy type for apex has been removed in Guilin release.
Removing it fron pdpGroup config in PAP.
Issue-ID: POLICY-2857
Change-Id: Iddf2e6b6301dbab3addcbff4c806142f4cf8e6c6
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Addressed the following sonar issues:
- too many assertions in test method
- extract common test method
- replace this null literal by an "Optional" object
Issue-ID: POLICY-2679
Change-Id: Ie80e5c5fdb94230a2773ea2931758cf54204ae86
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Issue-ID: POLICY-2827
Signed-off-by: jhh <jorge.hernandez-herrero@att.com>
Change-Id: I95db673538947cccc39bb5a389ab3f544fa91af8
|
|
Issue-ID: POLICY-2789
Change-Id: Ie4d4df890c1faac0e09d78d3b0d0157d8ef95410
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Default group will now have match Policy Type.
Issue-ID: POLICY-2596
Change-Id: I53c07b417097dc8250191e6ee8d043488b02562d
Signed-off-by: Pamela Dragosh <pdragosh@research.att.com>
|
|
New guard filter should be advertised by pap.
Issue-ID: POLICY-2590
Change-Id: I7652057d4d759e60c358ee9fad84a8c6a9af6898
Signed-off-by: Pamela Dragosh <pdragosh@research.att.com>
|
|
Change-Id: If4613c0fc6f940bc82cf616aba2ba9cc03e1b599
Issue-ID: POLICY-2754
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
|
|
Make the application to return exitCode as 1 when the main process
terminates due to issues such as:
1) Any basic issue in the startup config file causing error and
terminates the main process, for e.g. a wrong name in parameterGroup.
2) Wrong info specified for database connectivity, say wrong username.
This doesn't impact a case where database isn't ready yet and PAP
container comes up first, such waiting is already handled by
initContainer in the oom chart. PAP pod waits for policy-mariadb pod to
be ready first, then only the main process is started and comes to this
point.
3) An invalid port number is specified and the rest server is not able
to start.
These are the some of the scenarios which already makes the main process
of PAP to terminate, and in these cases, an exception is thrown now which
returns the exitCode as 1.
Change-Id: Ie48726bcb36d7304eb60807fbaabbee8245e9c70
Issue-ID: POLICY-2754
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
|
|
Changing the policy component health check to fail if the required
PDP instances are down or not registered with PAP.
The check is performed by comparing the "currentInstanceCount"
& "desiredInstanceCount" fields of PdpSubGroup.
Issue-ID: POLICY-2718
Change-Id: Iad1075d0812f4ef166dfed4f110eff4d9e5f995e
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Issue-ID: POLICY-2763
Change-Id: I9701a80cca10b9fd49d8060ecb6601097ca5824d
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Issue-ID: POLICY-2695
Change-Id: I7f30fa9c89d7ad9ddc00644c170d8659f7ee3230
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Replaced try/catch blocks in policy/pap test cases with assertj
assertions
Issue-ID: POLICY-2451
Change-Id: Ic963558e812b59d321e12fa87e7285732d6761e5
Signed-off-by: waynedunican <wayne.dunican@est.tech>
|
|
Somehow missed this sonar on one line of code:
- remove useless assignments
Apparently, sonar only complains about some useless assignments (perhaps
those that do more than copy a constant). Fixed one about which it did
not complain, which appeared immediately before the one about which it
DID complain.
At one point, sonar complained about creating ArrayLists without
specifying the initial size (when it was known). It didn't complain
about this one, but went ahead and fixed it anyway.
Issue-ID: POLICY-2679
Change-Id: I270341195634416db1f8152a95a5822c8d505803
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
|
|
Addressed issues reported due to updates to the sonar rules:
- invoke only one method in a junit lambda
Issue-ID: POLICY-2679
Change-Id: I422856527b191a45b30a78bb8520f679ca2e7896
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Issue-ID: POLICY-2390
Signed-off-by: puthuparambil.aditya <aditya.puthuparambil@bell.ca>
Change-Id: I42f443c64bcf6652adb9795ee8e71e37d8fa8c71
|
|
Change-Id: I995515659a4b7837bb15120f9c50fa0fd1f5434c
Issue-ID: POLICY-2188
Signed-off-by: a.sreekumar <ajith.sreekumar@bell.ca>
|
|
replace by assertNotSame
Issue-ID: POLICY-2616
Change-Id: I927dbd8c57654cd2d2ef6262d13d63c79ba9035f
Signed-off-by: Taka Cho <takamune.cho@att.com>
|
|
Modified PAP to discard any messages old than 10 minutes so that
if PAP is unable to communicate with DMaaP and then suddently gets
flooded, it will discard old messages.
Updates per review comments
- made the "max message age" configurable, with a default of 10 minutes
Issue-ID: POLICY-2170
Change-Id: I541c03aabf29482af12a07f51eeb5888bbafbf31
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Issue-ID: POLICY-2078
Change-Id: I035eceb45e76ef11df0e903b584fcf7f8eb63208
Signed-off-by: RossC <ross.carter@est.tech>
|
|
Issue-ID: POLICY-2514
Change-Id: Id780f9bc672cde633cf3d7d42b7c9f98fad0396f
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Issue-ID: POLICY-2514
Change-Id: Iea5b6375975b0225d5b803cc2500adc46fc83804
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Tracked it down to the following sequence:
- policies were deployed to the pdp
- received a request to undeploy a policy
- generated a notification for the formerly deployed policy
- this should not have been generated
- undeployed the policy
- received a response from the pdp indicating the policy had
been undeployed
- generated a notification indicating the policy was undeployed
Removed the code that generated a notification when a policy is
initially moved from deployed to undeployed (or vice versa).
Issue-ID: POLICY-2539
Signed-off-by: Jim Hahn <jrh3@att.com>
Change-Id: If175974b5fa5ccda6a1e1ab8fa1326b263bb8005
|
|
This is a more robust solution to the race condition previously
identified with back-to-back deployment requests. The old fix has
been rolled back and replaced with this fix.
Issue-ID: POLICY-2527
Signed-off-by: Jim Hahn <jrh3@att.com>
Change-Id: I2ea93f3a5eaac822abecf5d0745429b95712c861
|
|
|
|
The problem appears to be a multi-threading type of issue:
- PAP sends request to XACML
- JMeter pushes the next deployment request into PAP
- PAP receives response from XACML
- based on request ID, the response is routed to the handler,
but blocks while PAP processes the deployment request
- PAP replaces old list of policies with new, and changes the
request ID
- PAP responds to JMeter
- PAP handles XACML response, even though it's request ID no
longer matches
Modified PAP to no longer replace any messages that are at the
head of the queue.
Fixed the license.
Issue-ID: POLICY-2527
Change-Id: I43218a7d0591649d761e17282189e51d5dbd26ed
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
|
|
Issue-ID: POLICY-2525
Signed-off-by: jhh <jorge.hernandez-herrero@att.com>
Change-Id: Iea676911eaa7222949959ed0a7074b1260d8ce87
|
|
Issue-ID: POLICY-2510
Change-Id: I7dae085491e10082f222404179554436beb81bf1
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Fixed the code so that if a request is made to activate a PDP Group
that is already active, the group remains active.
Issue-ID: POLICY-2522
Change-Id: I106332b1ce15fde8dbf0b3e72034fd11c250dc69
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
Issue-ID: POLICY-2378
Change-Id: I91483bc2ab270a6cce251a1b8e80b9767d956d89
Signed-off-by: Ram Krishna Verma <ram_krishna.verma@bell.ca>
|
|
Addressed these remaining sonar issues:
- main arg validation
- TODO item regarding subgroup updates with no PDPs
- after discussion, it was decided to not do this
Issue-ID: POLICY-2305
Change-Id: Ib2b340d029679d6cd2b2a558318b25d38b830833
Signed-off-by: Jim Hahn <jrh3@att.com>
|
|
|