Red Hat, Inc.

Jeffrey Thomas Layton

Eric L Paris

Neil R T Horman

Patrick Joseph Donnelly

Zheng Yan

Josef Michael Bacik

Ramana Krisna Venkatesh Raja

Varsha Rao

Gregory Allan Farnum

James Bruce Fields

Metadata including an encrypted file name can be attached to a directory entry for a file. For example, a method may include a processing device performing operations. The operations may include receiving, from a client system, a request to generate a directory entry in a distributed file system for a file. The operations may include generating the directory entry in a storage node of the plurality of storage nodes. The directory entry may include a file name and an index number. The operations may include encrypting the file name to generate an encrypted file name. The operations may include attaching the encrypted file name to the directory entry as metadata, the metadata being distinct from the file name and the index number. The metadata may be usable by the client system for decrypting the encrypted file name.

Encrypted file name metadata in a distributed file system directory entry

Systems and method for improved creation and management of distributed file systems are provided. In one embodiment, a method is provided that includes creating a plurality of containers within a cloud computing environment and executing, within the plurality of containers, a plurality of gateways for a distributed file system. Exports of the distributed file system may be assigned to a subset of the plurality of gateways. The exports may correspond to an exclusive subset of the distributed file system. The method may also include responding to requests to access exports of the distributed file system using containers executing gateways assigned to corresponding exports of the distributed file system.

Containerized gateways and exports for distributed file systems

A system includes a plurality of clustered servers, a plurality of clients communicating with the plurality of clustered servers, and a grace database. Each of the plurality of clustered servers maintains a respective client recovery list. A client of the plurality of clients is configured to send a reclaim request to a server of the plurality of clustered servers. The grace database is configured to maintain an epoch value and a reclaim value. The epoch value indicates the current epoch, and the reclaim value indicates from which epoch reclaim is allowed. The server of the plurality of clustered servers is configured to maintain a client list and receive the reclaim request from the client. The client list indicates each client currently holding state on the server.

Clustered network file system server reclaim via decentralized, coordinated grace period

The technology disclosed herein that may enable a client of a distributed storage system to recover a storage session after a failure occurs. An example method may include: identifying a storage session of a distributed storage service, the storage session comprising session data that corresponds to a storage object of the distributed storage service; providing, by a processing device of a client, an indication that the client is recovering the storage session; and obtaining, by the client, the session data of the storage session from one or more devices that accessed the storage object of the distributed storage service.

Client session reclaim for a distributed storage system

Some embodiments of a system and a method to notify applications of lost computing resources have been presented. For instance, a processing device running on a client machine can monitor a computing resource used by an application, which also runs on the client machine. If the computing resource is lost, then the processing device can notify the application of the loss and provides details about the lost computing resource to the application so that the application can take appropriate action in response to the loss.

Lost computing resource notification

A system and a method are disclosed for managing file locks, including initiating, by a processing device executing a kernel, executions of a number of active tasks that each has acquired a respective lock to a record, and in response to release of a first lock to the record by an active task, waking up a previously-designated worker task out of a number of idle tasks, in which the worker task is to attempt an acquisition of a second lock on behalf of at least one remaining task of the idle tasks.

File lock

A computing device identifies a data packet received at a computing device. The computing device allocates memory having a fixed size to store the network data packet. A latency reducer identifies a free space in the memory allocation, the free space comprising a difference between the fixed size of the memory allocation and a size of the network data packet. The latency reducer creates a socket buffer list for the network data packet in the free space, the socket buffer list comprising a plurality of entries to serve as socket queue objects for a plurality of applications.

Reducing latency in multicast traffic reception

A method and system for sending data in a file system that uses cryptographic signatures to protect data integrity. A computer system calculates a signature based on the content of a page of a memory. The memory is shared by processes that run on the computer system. The computer system write-protects the page while the page is used for calculation of the signature. When a first process attempts to modify the page, a page fault is triggered. In response to the page fault, the content of the page in memory is copied to a new page in the memory. The new page is accessible by the processes. Access to the page by the first process is redirected to the new page. Subsequent to the page fault, access to the page by the second process is also redirected to the new page.

Performing zero-copy sends in a networked file system with cryptographic signing

A system configures page tables to cause an operating system to copy original page data in a data store when any one of the application processes makes a first write request for the original page data. The system detects a page fault from a memory management unit receiving a first write request from one of the application processes and creates the copy in physical memory to allow the application process to modify the page data copy. The other application processes have read access to the original page data. The system replaces the original page data in the data store with the page data copy in response to receiving a first synchronization request from the application process and updates a page table for one of the other application processes to configure access to the replaced page data in response to receiving a second synchronization request from the one other application process.

Growing community of inventors

Raleigh, NC, United States of America

Jeffrey Thomas Layton