Abstract
This paper presents the experience in providing CERN users with direct online access to their EOS/CERNBox-powered user storage from Windows. In production for about 15 months, a High-Available Samba cluster is regularly used by a significant fraction of the CERN user base, following the migration of their central home folders from Microsoft DFS in the context of CERN’s strategy to move to open source solutions. We describe the configuration of the cluster, which is based on standard components: the EOS-backed CERNBox storage is mounted via FUSE, and an additional mount provided by CephFS is used to share the cluster’s state. Further, we describe some typical shortcomings of such a setup and how they were tackled. Finally, we show how such an additional access method fits in the bigger picture, where the storage is seamlessly accessed by user jobs, sync clients, FUSE/Samba mounts as well as the web UI, whilst aiming at a consistent view and user experience.
Highlights
At CERN, the offer of personal storage for the user community has historically been based on two independent services: on one hand the Microsoft Distributed File System (DFS) [11], for Windows-based users, and on the other hand the Andrew File System (AFS), for Linux-based users
One of the most striking evidences when analyzing the strace profiles was the large amount of system calls related to extended attributes: as a matter of fact, when a Windows client requests the properties of a file, a Samba server is requested to retrieve the list of all its xattrs and their values
The EOS storage system extensively uses xattrs for internal metadata information and bookkeeping, making the getxattr system call by far the most recurrent call in any of our profiling reports. To mitigate this performance bottleneck, we have enabled the possibility to hide the system-level extended attributes in the EOS FUSE mounts, such that only user-specific ones like the Access Control Lists (ACLs) are exposed to Samba
Summary
At CERN, the offer of personal storage for the user community has historically been based on two independent services: on one hand the Microsoft Distributed File System (DFS) [11], for Windows-based users, and on the other hand the Andrew File System (AFS), for Linux-based users. Providing a universal CERN home directory with easy access to user files from any device in any location enables and fosters collaboration and mobility. CERNBox [1] is built upon modern sync & share functionalities and it is the adopted solution to fulfill the requirements of a universal home directory service, supporting any OS on the market. As CERNBox is built on top of Linux, the Samba service presented in this paper is instrumental to meeting such requirements.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
