Veritas Cluster Server and InfoScale help
Notes from years of running Veritas clusters, brought up to date. Veritas Cluster Server (VCS), Volume Manager (VxVM) and File System (VxFS) are still sold and supported, now as InfoScale. Most everyday commands on these pages have changed little in decades, but licensing and installation have changed a lot: check the documentation for your exact version.
Who owns Veritas now?
The products have changed hands several times: Veritas Software was bought by Symantec in 2005, split off again as Veritas Technologies in 2016, and in December 2024 Veritas' data protection business (NetBackup) merged with Cohesity. The remaining products, InfoScale, Backup Exec and the data compliance tools, became a new company, Arctera. Cloud Software Group (the owner of Citrix) bought Arctera in 2025 and runs InfoScale as a separate business. Documentation is on SORT (Services and Operations Readiness Tools, sort.veritas.com); support articles, licences and downloads are on the InfoScale support portal.
Pages in this section
- Product overviewWhat each Veritas storage and availability product does, how the old names map to the InfoScale editions, which platforms are supported, and how licensing works.
- Cluster overviewHow Veritas Cluster Server (InfoScale Availability) works: nodes, service groups, resources and dependencies, LLT and GAB heartbeats, I/O fencing, what happens at startup and failover, the main.cf file, logs and how to change the configuration safely.
- Cluster installHow to plan and install a Veritas Cluster Server (InfoScale Availability or Enterprise) cluster: hardware and network preparation, the pre-installation checklist, running the installer, configuring LLT, GAB and fencing, and building the first service group.
- Cluster debuggingStep-by-step troubleshooting for Veritas Cluster Server: reading hastatus, starting a cluster that will not come up, seeding GAB on a single node, checking licences and heartbeats, clearing faults, stuck resources, frozen groups and reading the engine log.
- Testing failoverA complete failover test plan for Veritas Cluster Server: checks before testing, manual switch-over, application failure, node crash, heartbeat and network failures, storage path failure, and how to record the results.
- Volume Manager (VxVM)How Veritas Volume Manager organises storage into disks, disk groups, volumes, plexes and subdisks, and the everyday VxVM commands: initialise disks, create and resize volumes, mirror, replace a failed disk, move disk groups between servers, and DMP multipathing.
- File System (VxFS)Veritas File System (VxFS) explained: features, creating and mounting VxFS, online resizing and defragmentation with fsadm, snapshots and checkpoints, tuning with vxtunefs, checking with fsck, and cluster file systems.
Quick reference
| Task | Command |
|---|---|
| Cluster status | hastatus -sum |
| Start or stop the cluster on this node | hastart, hastop -local |
| Switch a service group to another node | hagrp -switch GROUP -to NODE |
| Clear a faulted resource | hares -clear RESOURCE -sys NODE |
| Make the configuration writable, then save it | haconf -makerw … haconf -dump -makero |
| Heartbeat links and membership | lltstat -nvv, gabconfig -a |
| Main log | /var/VRTSvcs/log/engine_A.log |
| Disks and volumes | vxdisk -o alldgs list, vxprint -ht |
| Licences | vxlicrep |
The commands live in /opt/VRTSvcs/bin, /opt/VRTS/bin and /sbin. Add them to root's PATH:
export PATH=$PATH:/opt/VRTSvcs/bin:/opt/VRTS/bin:/sbinOfficial resources
- SORT (Services and Operations Readiness Tools): documentation, patches, compatibility lists and installation checks.
- InfoScale: the InfoScale business, now part of Cloud Software Group.
- InfoScale support portal: knowledge base, licences and downloads.
Alternatives on Linux include Pacemaker with Corosync (the high-availability add-ons of RHEL and SUSE) and, on AIX, IBM PowerHA SystemMirror.