International Business Machines Corporation

Kishor S Kulkarni

Madhu B Ananthapadmanabh

Sreejith Nalamvathukkal

Hariharan N Venkitachalam

Gopalakrishnan P

Anubhuti Kaushik

Matthew John Comer

Reshmi George

Dipesh Rawat

Debabrata Sarkar

Intelligent container management is provided. An issue that caused a failed state of a container is identified based on an analysis of exit code information, container log information, host node agent log information, and resource utilization data corresponding to the container using a group of solution agents. It is determined whether the issue that caused the failed state of the container is a resolvable issue with a fix. The fix is applied to the container using a runtime mutating webhook controller prior to restart of the container in response to determining that the issue that caused the failed state of the container is the resolvable issue with the fix. The restart of the container is performed on a host node.

Intelligent container management utilizing runtime mutating webhook

Embodiments include methods, systems and computer program products for performing a two-phase commit conformance test for a cloud based online transaction processing system (OLTP). Aspects include receiving, by a test case manager of the OLTP from a transaction manager of the OLTP, a transaction event including metadata regarding a transaction and determining a state of the transaction. Aspects also include identifying a test case based on the metadata and the state of the transaction and issuing state events to the transaction manager, wherein the state events are determined based on the test case. Aspects further include obtaining log information from the transaction manager and determining compliance by the transaction manager with the two-phase commit conformance test based at least in part on the log information.

Testing a two-phase commit protocol conformance of a cloud based online transaction processing platform

A method, computer program product, and a system to globally serialize transactions where a processor(s) establishes a communications connection a (serialization) resource and a resource manager for a distributed computing system. The processor(s) obtains a first request from an application executing on the resource for access to a global resource managed by the resource manager, for executing a transaction. The processor(s) implements a lock for the global resource in an object store of the resource manager over the communications connection. The processor(s) communicates the lock to the application, which executes the transaction and the processor(s) updates a memory with a record comprising attributes of the lock. The processor(s) obtains a second request from the application to terminate the lock, obtains, identifies the lock for the transaction, in the object store, and updates the object store to delete the lock.

Global transaction serialization

A method, system and computer program is provided for enabling varying computer applications that are running in Enterprise Information System to send request to a cluster of Java Enterprise Server (J2EE server) cluster. One or more computers provide an application server environment including one or more servers operating within the application server environment to run an application. The computers create a cluster of resource adapters and create a cluster thread for each of the resource adapters to read a cluster port from configured properties, wherein the cluster thread for each of the resource adapters enables an enterprise information system to participate in a cluster of J2EE runtime servers.

Enabling a traditional language platform to participate in a Java enterprise computing environment

The present disclosure relates to techniques for discovering and exposing middleware applications as service endpoints through a universal connector in a middleware-agnostic fashion that does not require additional coding. In one embodiment, a back-end middleware system identifies a set of middleware applications that operate on an input/output (I/O) data area. Next, the back-end middleware system generates a set of discovery artifacts describing the middleware applications and the I/O data area. A discovery client sends a discovery request to a the back-end middleware system, which sends structured information from the discovery artifacts in response. Based on the structured information, the discovery client creates a REST model for the set of middleware applications, mapper artifacts that map REST operations to the middleware applications, and transformer artifacts that indicate how to convert data between Javascript-object-notation (JSON) format and format that is compatible with the I/O data area.

Growing community of inventors

Karnataka, India

Kishor S Kulkarni