managing celerra file systems

1 of 66

Managing CelerraFile Systems

P/N 300-002-120Rev A01

Version 5.4April 2005

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3Access to Celerra File Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6Restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6Cautions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

Celerra File System Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7File System Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7File System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8File System Planning Considerations. . . . . . . . . . . . . . . . . . . . . . . . . .8About Mounting a Celerra File System on a Data Mover . . . . . . . . .10Monitoring the Amount of Space in Use in a File System . . . . . . . .11File Systems and Automatic Volume Management . . . . . . . . . . . . . .12Celerra WORM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12Checking File System Consistency . . . . . . . . . . . . . . . . . . . . . . . . . .13Nested Mount File Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16EMC NAS Interoperability Matrix. . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

User Interface Choices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17Managing Celerra File Systems Roadmap . . . . . . . . . . . . . . . . . . . . . . . .18Create a Celerra File System on a Celerra Network Server . . . . . . . . . . .20Create a Mount Point for a Celerra File System on a Data Mover . . . . . .22Mount a Celerra File System on a Mount Point on a Data Mover . . . . . .23Enabling Access to Celerra File Systems . . . . . . . . . . . . . . . . . . . . . . . . .24

Enable NFS Access to a Celerra File System. . . . . . . . . . . . . . . . . . .24Enable CIFS Access to a Celerra File System . . . . . . . . . . . . . . . . . .24Enable FTP Access to a Celerra File System. . . . . . . . . . . . . . . . . . .24Enable TFTP Access to a Celerra File System. . . . . . . . . . . . . . . . . .25Enable MPFS Access to a Celerra File System . . . . . . . . . . . . . . . . .25

Managing Celerra File Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26Listing All Celerra File Systems on a Celerra Network Server . . . . .26Listing Configuration Information for a Celerra File System . . . . . .27Checking Celerra File System Capacity . . . . . . . . . . . . . . . . . . . . . . .27Finding Out If a File System Has an Associated Storage Pool . . . .30Extending a Celerra File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31Exporting a Celerra File System . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

Managing Celerra File SystemsVersion 5.4 2 of 66

Renaming a Celerra File System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36Deleting a Celerra File System and Freeing Its Disk Space . . . . . . . 36Listing Mount Points on a Data Mover . . . . . . . . . . . . . . . . . . . . . . . . 39Listing All Celerra File Systems Mounted on a Data Mover. . . . . . . 39Mounting a File System for Read/Write Access . . . . . . . . . . . . . . . . 40Mounting a File System for Read-Only Access. . . . . . . . . . . . . . . . . 40Unmounting a Celerra File System from a Data Mover . . . . . . . . . . 41Unmounting All Celerra File Systems from a Data Mover . . . . . . . . 43Changing the Size Threshold for All File Systems . . . . . . . . . . . . . . 45Changing the Size Threshold for a Data Mover. . . . . . . . . . . . . . . . . 45Enhancing File Read and Write Performance . . . . . . . . . . . . . . . . . . 46

Finding and Fixing File System Storage Errors . . . . . . . . . . . . . . . . . . . . 49Starting and Monitoring a File System Check . . . . . . . . . . . . . . . . . . 49Starting ACL Check on a File System . . . . . . . . . . . . . . . . . . . . . . . . 49Listing Current File System Checks. . . . . . . . . . . . . . . . . . . . . . . . . . 50Displaying Information on a Single File System Check . . . . . . . . . . 50Displaying Information on all Current File System Checks . . . . . . . 51

Managing Celerra Nested Mount File Systems . . . . . . . . . . . . . . . . . . . . 52Creating a Nested Mount File System . . . . . . . . . . . . . . . . . . . . . . . . 52Mounting a Nested Mount File System . . . . . . . . . . . . . . . . . . . . . . . 53Deleting the Nested Mount File System. . . . . . . . . . . . . . . . . . . . . . . 53Adding an Existing File System to the NMFS . . . . . . . . . . . . . . . . . . 55Exporting a Component File System . . . . . . . . . . . . . . . . . . . . . . . . . 56Removing a Component File System. . . . . . . . . . . . . . . . . . . . . . . . . 56Moving a Nested Mount File System . . . . . . . . . . . . . . . . . . . . . . . . . 56

Troubleshooting File Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57Error Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57Known Problems and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Related Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60Want to Know More? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Appendix: GID Support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61Restrictions for GID Support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

3 of 66Version 5.4Managing Celerra File Systems

IntroductionThe purpose of a file server is to make file systems available to users. The Celerra® Network Server is a file server that offers the administrator flexibility in creating file systems to respond to customer needs.

This technical module is part of the Celerra Network Server information set. This module is intended for the system administrator responsible for configuring and managing Celerra volumes and file systems.

Access to Celerra File SystemsThe Celerra Network Server supports file system access by NFS and CIFS users, as well as by other file system access protocols. In order to configure the file systems and access protocols for different users, you must understand the types of users you have and how they need to access files.

The Celerra Network Server lets you configure a file system for the following access protocols:

◆ NFS user access: For how to set up and manage NFS client and user access to a Celerra file system, refer to the Managing NFS Access to the Celerra Network Server technical module on the user information CD.

◆ CIFS user access: For how to set up and manage CIFS client and user access to a Celerra file system, refer to the guides Configuring CIFS on Celerra and Managing Celerra for the Windows Environment on the user information CD.

◆ Simultaneous CIFS and NFS user access: For how to set up and manage simultaneous CIFS and NFS client and user access to a Celerra file system, refer to the Configuring CIFS on Celerra for a Multiprotocol Environment technical module on the user information CD. Note that the Celerra Network Server lets you map Windows users and groups to UNIX UIDs (user IDs) and GIDs (group IDs) to provide users with seamless access to shared file system data; for details, refer to the Configuring External Usermapper for Celerra technical module on the user information CD.

◆ File Transfer Protocol (FTP) access: FTP is a client/server protocol that operates over TCP/IP and allows file uploading and downloading across heterogeneous systems. FTP includes functions to log on to the remote system, list directories, and copy files. For how to set up and manage FTP access to a Celerra file system, refer to the Using FTP on Celerra Network Server technical module on the user information CD.

◆ Trivial File Transfer Protocol (TFTP) access: A simple, UDP-based protocol to read and write files. TFTP can be used to boot remote clients from a network server. TFTP does not authenticate users or provide access control.

../../CelerraDocCD.htm#DocListing.htm





Managing Celerra File Systems4 of 66 Version 5.4

◆ Multiplex File System (MPFS) access: MPFS adds a thin, lightweight protocol known as the File Mapping Protocol (FMP) that works with NAS protocols such as NFS and CIFS to control metadata operations. FMP is responsible for exchanging file layout information and management of sharing conflicts between clients and Celerra Network Servers. The clients use the file layout information to read and write file data directly from and to the storage system. For how to set up and manage MPFS access to a Celerra file system, refer to the Using HighRoad on Celerra technical module on the user information CD.

Note: All referenced documents are available on the Celerra Network Server User Information CD.

TerminologyThis section defines terms that you need to understand in order to manage volumes and file systems on the Celerra Network Server. Refer to the Celerra Network Server User Information Glossary on the user information CD for a complete list of Celerra terminology.

Automatic Volume Management (AVM): A feature of the Celerra Network Server that creates and manages volumes automatically, without manual volume management by an administrator. AVM organizes volumes into pools of storage that can be allocated to file systems.

Business continuance volume (BCV): A Symmetrix® volume used as a mirror that attaches to and fully synchronizes with a production (source) volume on the Celerra Network Server. The synchronized BCV is then separated from the source volume and is addressed directly from the host to serve in backup and restore operations, decision support, and application testing.

Celerra WORM: Celerra Write-Once Read-Many feature.

Component file system: Refers to the file system/s mounted on the nested mount root file system and is part of the nested mount file system.

Disk volume: On Celerra systems, a physical storage unit as exported from the storage array. All other volume types are created from disk volumes. See also Metavolume, Slice volume, Stripe volume, Volume.

File system: A method of cataloging and managing the files and directories on a storage system.

LUN (Logical unit number): The identifying number of a SCSI object that processes SCSI commands. The LUN is the last part of the SCSI address for a SCSI object. The LUN is an ID for the logical unit, but the term is sometimes used to refer to the logical unit itself.

Metavolume: On a Celerra system, a concatenation of volumes, which can consist of disk, slice, or stripe volumes. Also called a hypervolume or hyper. Every file system must be created on top of a unique metavolume. See also Disk volume, Slice volume, Stripe volume, Volume.

Nested export: An export of a component file system. The export has its own specified access controls.

Nested mount file system (NMFS): File system that contains the nested mount root file system and component file systems.




Nested mount file system root: File system on which the component file systems are mounted; the NMFS root is mounted as read/only.

Slice volume: On a Celerra system, a logical piece or a specified area of a volume used to create smaller, more manageable units of storage. See also Disk volume, Metavolume, Stripe volume, Volume.

Storage pool: Automated Volume Management (AVM), a Celerra feature, organizes available disk volumes into groupings called storage pools. Storage pools are used to allocate available storage to Celerra file systems. Storage pools can be created automatically by AVM or manually by the user.

Storage system: An array of physical disk drives and their supporting processors, power supplies, and cables.

Stripe volume: An arrangement of volumes that appear as a single volume. Allows for stripe units, which cut across the volume and are addressed in an interlaced manner. Stripe volumes make load balancing possible. See also Disk volume, Metavolume, Stripe volume, Volume.

Volume: On a Celerra system, a virtual disk into which a file system, database management system, or other application places data. A volume can be a single disk partition or multiple partitions on one or more physical drives. See also Disk volume, Metavolume, Slice volume, Stripe volume.

Worm: A file's status in a Celerra WORM file system. When a file is in worm state, it cannot be moved, modified, extended, removed, or renamed.


References◆ If you are CWORM enabled, refer to the Using Celerra WORM technical module

on the user information CD.

◆ For additional information on file system user/group quotas management and tree quota enhancements, refer to the Using Quotas on Celerra technical module on the user information CD.

Restriction◆ When building volumes on a Celerra Network Server attached to a Symmetrix

storage system, use standard Symmetrix volumes (also called hypervolumes), not Symmetrix metavolumes.

Cautions ◆ All parts of a Celerra file system must use the same type of disk storage and must

be stored on a single storage system.◆ If you plan to set quotas on a file system to control the amount of space that

users and groups can consume, you should turn on quotas immediately after creating the file system. Turning on quotas later, when the file system is in use, can cause temporary file system disruption, including slow file system access. For information about quotas and how to turn on quotas, refer to the technical module Using Quotas on Celerra on the user information CD.

◆ If your user environment requires international character support (i.e., support of non-English character sets or Unicode characters), EMC strongly recommends configuring your Celerra Network Server to support this feature before creating file systems. For detailed information about international character support and how to configure it on a Celerra Network Server, refer to the technical module Using International Character Sets with Celerra on the user information CD.

◆ If you plan to create TimeFinder®/FS (local, NearCopy, or FarCopy) snapshots of a production file system (PFS), do not use slice volumes (nas_slice) when creating the PFS. Instead, use the full disk presented to the Celerra Network Server, since TimeFinder copies and restores volumes in their entirety and does not recognize sliced partitions created by the host (in this example, the Celerra Network Server).

◆ Do not manually edit the nas_db database without consulting EMC Customer Support. Any changes to this file could cause problems with your Celerra installation.






Celerra File System ConceptsA file system lets you arrange and manage files and directories on a storage system. File systems on a Celerra Network Server are stored in volumes configured to meet your business requirements.

A metavolume is required to create a file system. Metavolumes provide the storage capacity that can be used to expand file systems. A metavolume is also a way to form a logical volume that is larger than a single disk. For information on volumes and how to configure metavolumes, refer to the Managing Celerra Volumes technical module on the user information CD.

After you configure metavolumes, you are ready to create your file systems.

In this section, we discuss file system:

◆ Types

◆ Requirements

◆ Planning considerations

File System Types The Celerra Network Server creates different file systems depending on how they are used. You might see the types listed in Table 1.

Table 1 File System Types

File System Type ID Description

uxfs 1 Default file system

sfs -- Share file system

rawfs 5 Raw file system

mirrorfs 6 Mirrored file system

ckpt 7 Checkpoint file system

mgfs 8 Migration file system

group 100 Stores members of a file system group

vpfs -- Volume pool file system (multiple services sharing a single metavolume)

rvfs -- Local configuration volume used to store replication-specific internal information

nmfs 102 Nested mount file system



File System Requirements The following lists the requirements for creating a file system:

◆ You can create a file system on only nonroot metavolumes that are not in use.

◆ A metavolume must be at least 2 megabytes to accommodate a file system.

◆ A file system name must be unique on a particular Celerra Network Server. It can include up to 254 characters, including hyphens ( - ), underscores ( _ ), and periods ( . ), but cannot begin with a hyphen or period.

!CAUTION!All parts of a Celerra file system should use the same type of disk storage and should be stored on a single storage system.

File System Planning ConsiderationsConsider the following when you are planning file systems for your Celerra Network Server:

Table 2 File System Planning Considerations

Issues Considerations

Size of the production file systems

Plan your backup and restore strategy. Larger file systems might take more time to back up and restore.If a file system becomes inconsistent (rare occurrence), larger file systems can take more time for a consistency check to run.

Data growth rates Be sure to consider your space needs as your data increases.

Use of TimeFinder/FS or remote TimeFinder/FS

If you plan to create multiple copies of your production file system, you need to plan for that number of BCVs. For example, from one production file system you may create 10 copies. Therefore, you need to plan for 10 BCVs, not one.TimeFinder/FS uses the physical disk, not the logical volume, when it creates BCV copies. The copy is done track by track, so unused capacity is carried over to the BCVs. Volumes used for BCVs need to be the same size as the standard volume. For information about TimeFinder/FS, refer to the Using TimeFinder/FS with Celerra and Using TimeFinder/FS Near Copy and Far Copy with Celerra technical modules on the user information CD.




File System Size Guidelines The size of your file systems depends on a variety of factors within your own business organization. Contact your EMC representative or refer to the Celerra Network Server Release Notes, which are available at Powerlink at http://powerlink.emc.com.

On a per-Data-Mover basis, the total size of all file systems, and the size of all SavVols used by SnapSure, and the size of all SavVols used by the Celerra Replicator feature, must be less than the total supported capacity of the Data Mover. For a list of Data Mover capacities, refer to the Celerra Network Server Release Notes, which are available at Powerlink at http://powerlink.emc.com.

Extending Capacity of a File SystemThe Celerra Network Server provides two ways to extend capacity of a file system:

◆ by volume

◆ by size if the file system has an associated storage pool

When you extend a file system by volume, you should extend it with the same type as the original file system. For example, if the original file system was created using a stripe volume, extend it using a stripe volume of the same size and depth.

You can extend by size a file system that has an associated storage pool. In this case, the AVM feature extends the file system with storage from the same storage pool.

Use of SRDF® disaster recovery All file systems on the Data Mover must be built on SRDF volumes. For information about SRDF, refer to the Using SRDF/S with Celerra for Disaster Recovery technical module or the Using SRDF/A with Celerra technical module on the user information CD. If you use the AVM feature to create the file systems, you must specify the symm_std_rdf_src storage pool. This storage pool directs AVM to allocate space from volumes configured at installation time for remote mirroring using SRDF.

Use of HighRoad® You cannot enable HighRoad access to file systems with a stripe depth of less than 32 KB. For more information, refer to the Using HighRoad on Celerra technical module on the user information CD.

Use of CWORM When you create a new file system, you have the option of applying the CWORM feature (Celerra Write-Once Read-Many) to the file system. Selecting the CWORM option designates the file system as CWORM; the file system is persistently marked as such until it is deleted.CWORM is a licensed feature of Celerra. You must have a CWORM software license to create a CWORM file systemThe CWORM feature is disabled by default.

Table 2 File System Planning Considerations (continued)

Issues Considerations




https://powerlink.emc.com





Deleting File Systems and Freeing Disk SpaceIf you want to delete a Celerra file system and free its disk space, you must delete or disconnect all entities associated with the file system: all checkpoints, business continuance volumes, slice volumes, stripe volumes, and metavolumes. After you delete or disconnect all of the file system entities, the disk volumes that provided storage space to the file system become part of the available free space on the file server.

About Mounting a Celerra File System on a Data Mover Before you can make a Celerra file system available for client access, you must create a mount point on a Data Mover, and then mount the file system on that mount point. After mounting the file server on a Data Mover, you can set up client access to the file system.

A file system can be mounted read/write on only one Data Mover, or read-only on many Data Movers. The same file system cannot be mounted both read/write and read-only on multiple Data Movers.

A mount point name must begin with a / (slash), followed by alphanumeric characters (for example, /new). The name can include multiple components that make the mounted file system appear to be a logical subdirectory (for example,/new1/new2 or /new1/new2/new3/new4).

With the exception of nested mount file systems, you should mount a file system only on the end of a multicomponent mount point name. No file systems should be mounted on intermediate components of a mount point name. For example, if you have a file system mounted on the /new1 mount point, you should not mount a file system on a mount point named /new1/new2.

A mount point name can have a maximum of seven components (for example,/new1/new2/new3/new4/new5/new6/new7). International characters are not supported as mount point names.

Options for Mounts on Data MoversWhen defining a mount point for a file system on a Data Mover, you can use different options to define how the file system mount is controlled.

Read/Write Option for Data Mover Mounts

When a file system is mounted read/write on a Data Mover, only that Data Mover is allowed access to the file system. No other Data Mover is allowed read/write or read-only access to that file system. This is the default for all file systems but checkpoint and TimeFinder/FS file systems.

Read-Only Option for Data Mover Mounts

When a file system is mounted read-only on a Data Mover, clients cannot write to the file system regardless of the permissions defined when the client access is set up. A file system can be mounted read-only on several Data Movers concurrently, and no Data Mover can mount the file system as read/write. This is the default for checkpoint and TimeFinder/FS file systems.


Monitoring the Amount of Space in Use in a File SystemThe Celerra Network Server monitors the amount of space in use in its file systems. It is useful to monitor this file system size threshold because the performance of a file system can degrade as its used space approaches 100 percent full.

By default, the Celerra Network Server is set up to trigger an event when the used space in a file system exceeds 90 percent full. To be notified of this event, you must set up event logging (an SNMP trap or e-mail notification) for it as described in the Configuring Celerra Events and Notifications technical module

When the file system size threshold is exceeded, you can take one of two corrective actions: Remove files from the file system or extend the file system size. Both actions have the side effect of reducing the amount of used space to a percentage below the threshold.

You may want to change the file system size threshold, for example, to a smaller percentage if the amount of space in use in your file systems grows at a rapid rate. You change the threshold by changing the percentage number in the global parameter file (if you want to change the threshold for all file systems on a Celerra Network Server) or in the parameter file for a single Data Mover (if you want to change the threshold only for file systems on that Data Mover).

Using Quotas to Prevent File Systems from Filling UpTo prevent a file system from getting full, you can impose quota limits on users and groups that create files or on directory trees.

You can set a hard quota limit on user, group, or directory tree quotas to prevent allocation of all of the space in the file system. Once the hard limit is reached, the system denies user requests to save additional files and notifies the administrator that the hard limit was reached. At this point, existing files can be read, but action must be taken either by the user or administrator to delete files from the file system, or increase the hard limit to allow the saving of additional files.

To avoid degradation of file system performance, you may choose to set the hard quota limit to between 80 percent and 85 percent of the total file system space. In addition to setting the hard quota limit, you can set a lower soft quota limit so that the administrator is notified when the hard limit is being approached.

For example, to prevent a file system containing 100 gigabytes of storage from filling up, you can set a soft quota limit of 80 gigabytes and a hard quota limit of 85 gigabytes using user, group, or directory tree quotas. When used space in the file system totals 80 gigabytes, the administrator is notified that the soft limit was hit. When used space totals 85 gigabytes, the system denies user requests to save additional files, and the administrator is notified that the hard limit was hit.

For detailed information about quotas and how to set up user, group, or directory tree quotas, refer to the Using Quotas on Celerra technical module on the user information CD.




File Systems and Automatic Volume ManagementThe Celerra management interfaces implement support of AVM to automate volume creation and management. When you use the AVM to create a file system in a Celerra interface, the file server automatically creates and manages volumes as part of file system creation and expansion operations.

AVM and the Celerra Command Line InterfaceThe Celerra command line interface (CLI) uses AVM to create a file system when you specify a size and storage pool from which to allocate the space, using the nas_fs -create size=<size> pool=<pool> options.

When you extend a file system created with AVM, you specify the extension size instead of a volume to add to the file system. AVM handles the extension by creating and allocating new storage from the storage pool associated with the file system. This ensures that the file system and its extensions consist of the same type of storage, with the same cost, performance, and availability characteristics.

AVM and the Celerra ManagerThe Celerra Manager uses AVM to create a file system when you create the file system from a storage pool.

When you use the Celerra Manager to extend a file system created from a storage pool, it is recommended that you extend the file system using space from the storage pool already associated with the file system. If the file system you want to extend has no associated storage pool, the Celerra Manager will determine the best storage pool to use for the extension and will present that pool as the default for the extension.

Celerra WORMCelerra WORM (Write-Once Read-Many) allows you to archive data to WORM storage on standard rewritable magnetic disks. You can write data to a CIFS or NFS file system a single time to create a permanent, noneditable set of records that cannot be altered, corrupted, or deleted.

CWORM can be applied to any standard Celerra file system. In a CWORM environment, administrators can use WORM protection on a per-file basis. Files can be stored with specified retention periods, which until expiration will prohibit the files from being deleted. Only an administrator has the ability to delete the CWORM file system.

A Celerra WORM-enabled file system:

◆ Safeguards data while ensuring its integrity and accessibility.

◆ Simplifies the task of archiving data for administrators and applications.

◆ Improves storage management flexibility and application performance.

If your system is CWORM enabled, refer to the Using Celerra WORM technical module on the user information CD for additional information on this feature.



Checking File System ConsistencyThe file system check (fsck) feature checks file system consistency on a specified file system by detecting and correcting file system storage errors.

When file system corruption is detected, the server either begins fsck automatically (auto fsck) or panics itself to initiate fsck on reboot. Also, you can start manual fsck on a particular file system using the nas_fsck command.

The fsck process cannot run on a read/only file system and you do not need to run fsck for normal reboot/shutdown operations. File system consistency is maintained through a logging mechanism and reboot/shutdown operation causes no corruption.

!! CAUTION

◆ During fsck, the file system is not available to the users. NFS clients receive a “NFS server not responding” message and CIFS clients lose the server connection and must remap shares.

◆ Depending on the size of the file system, the fsck utility may use a significant chunk of the system’s resources (memory and CPU) and may affect overall system performance.

◆ Two is the maximum number of fsck processes that can be run on a single Data Mover at the same time.

◆ A fsck of permanently unmounted file system can be executed on a standby Data Mover. The fsck process continues to run during and after a Data Mover failover.

◆ If a Data Mover reboots or experiences failover or failback while running fsck on an unmounted file system, fsck will not start automatically after the Data Mover reboots. You must start fsck manually on the unmounted file system.

Using auto fsck (default)The Celerra Network Server begins auto fsck when file system corruption is detected. The first step in the fsck process is to ensure the corruption can be safely corrected without bringing down the server. The fsck process also corrects any inconsistencies in the ACL (Access Control List) database.

The corrupted file system is not available to users during the fsck process. After fsck finds and corrects the corruption, users again have access to the file system. While fsck is running, other file systems mounted on the same server are not affected and are available to users.

You can check the status of fsck through the Control Station using the nas_fsck command. Refer to the section on displaying information on a single file system check.


Using fsck on RebootIf auto fsck cannot safely correct the corruption or if there are not enough system resources available to run fsck, the server panics itself to initiate fsck on reboot. During reboot, the server detects the corrupted file system and begins fsck.

By default, the reboot process runs fsck on any corrupted file system. If you set the ufs.skipFsck parameter, the reboot process does not run fsck and the corrupted file systems are not mounted. Refer to the Celerra Network Server Parameters Guide on the user information CD for additional information on parameters.

Using Manual fsckThe nas_fsck command allows you to manually start fsck on a particular file system. The nas_fsck command also lists and displays the status of fsck and aclchk. The aclchk utility finds and corrects any errors in the ACL database.

The fsck utility should not be run on a server under heavy load to prevent the server from running out of resources. In most cases, the user is notified when sufficient memory is not available to run a fsck. In these cases, users can choose one of the following options:

◆ Start the fsck during off-peak hours.

◆ Reboot the server and start fsck immediately.

◆ Run fsck on a different server if the file system is unmounted.

For more information on manually starting and using the nas_fsck command, refer to the section on finding and fixing file system storage errors and the Celerra Network Server Command Reference Manual on the user information CD.




Nested Mount File SystemsThe Nested Mount File System (NMFS) allows you to manage a collection of file systems as a single file system. The NMFS contains the component file systems that are mounted on the NMFS root.

The client sees the component file systems as a single share or single export. The following features are supported on component file systems only:

◆ Snap

◆ Replicator

◆ Quotas

◆ Security policies

◆ Backup

◆ SRDF

◆ TimeFinder FS

◆ MPFS access to MPFS enabled FS

◆ CDMS

◆ FileMover

When using NMFS, keep in mind the following considerations and procedures:

◆ NMFS contain component file systems only.

◆ Capacity is managed independently for each component file system.

◆ You can increase the total capacity of the NMFS either by extending an existing component file system or by adding new component file systems.

◆ There are no additional limitations on the number of nested file systems or component file systems beyond the existing limitations for the number of file systems on a Data Mover.

◆ NMFS root is a read/only file system.

◆ Hard links (NFS), renames, and simple moves are not possible from one component file system to another.

◆ NMFS can contain different classes of storage.


System RequirementsTable 3 lists the requirements for creating and managing file systems on the Celerra Network Server using this document.

EMC NAS Interoperability MatrixRefer to the EMC NAS Interoperability Matrix for definitive information on supported software and hardware, such as backup software, Fibre Channel switches, and application support for Celerra network-attached storage (NAS) products.

To view the EMC NAS Interoperability Matrix:

1. Go to http://powerlink.emc.com.

2. Search for NAS Interoperability Matrix.

The EMC NAS Interoperability Matrix appears in the list with a high score. The default order for search results is from highest to lowest score.

Table 3 System Requirements for Managing File Systems

Software Celerra Network Server Version 5.2 or higher

Hardware No specific hardware requirements

Network No specific network requirements

StorageNo specific storage requirements

Note: Any Celerra-qualified storage system.



User Interface ChoicesThe Celerra Network Server offers flexibility in managing networked storage based on your support environment and interface preferences.

This technical module describes how to configure Celerra volumes and file systems using the command line interface (CLI). You can also perform most of these tasks using the Celerra Manager management interface.

For more information about Celerra Manager, refer to Getting Started with Celerra Management and the Celerra Manager online help in the application or on the user information CD.




Managing Celerra File Systems RoadmapThis roadmap shows the process for configuring and managing file systems. This process contains components that represent the sequential phases of the roadmap. In addition, any nonsequential phases are represented in blocks at the base of the roadmap. Each phase contains the tasks required to complete the process.

Note: When viewing online, click the text in the roadmap to access that phase. To return to this roadmap from other pages, click the roadmap symbol at the center bottom of the page.

Table 4 lists the tasks to manage Celerra volumes as described in this technical module.

For a detailed synopsis of the commands presented in this section or to view syntax conventions, refer to the online Celerra man pages or the Celerra Network Server Command Reference Manual on the user information CD.

Table 4 File Systems Roadmap Tasks

Task Procedure

Explain procedures for creating a Celerra file system with existing volumes or using the AVM feature.

Create a Celerra File System on a Celerra Network Server on page 20.

Create a mount point on a Data Mover for a file system

Create a Mount Point for a Celerra File System on a Data Mover on page 22.

Mount a Celerra file system on a Data Mover Mount a Celerra File System on a Mount Point on a Data Mover on page 23



Create a Mount Point for a Celerra File System on a Data Mover

Create a Celerra File System on a Celerra Network Server

Mount a Celerra File System on a Mount Point on a Data Mover

Enable NFS Access to a Celerra File System

Enable CIFS Access to a Celerra File System

Enable FTP Access to a Celerra File System

Enable MPFS Access to a Celerra File System

Enable TFTP Access to a Celerra File System


Create a Celerra File System on a Celerra Network ServerThe following sections explain procedures for creating a Celerra file system with existing volumes or using the AVM feature. For an explanation of AVM, refer to Celerra File System Concepts on page 7.

Creating a Celerra File System by VolumeUse this procedure to create a Celerra file system that includes existing volumes.

Prerequisite: Perform the procedure for creating a metavolume in the Managing Celerra Volumes technical module on the user information CD.

Action

To create a file system with existing volumes, use this command syntax: $ nas_fs -name <name> -create <volume_name>

Where:<name> = the name assigned to a file system<volume_name> = the name of the existing volumeExample:To create a file system with existing volumes called ufs1, type:$ nas_fs -name ufs1 -create mtv1

Output

id = 18name = ufs1acl = 0in_use = Falsetype = uxfsvolume = mtv1rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs = 002806000209-006, 002806000209-007, 002806000209-008, 002806000209-009disks = d3,d4,d5,d6

Note

A file system name must be unique on a particular Celerra Network Server. It can include up to 254 characters, including hyphens ( - ), underscores ( _ ), and periods ( . ), but cannot begin with a hyphen or period.Refer to the Celerra Network Server Command Reference Manual on the user information CD for information about the options available for creating a file system with the nas_fs command.




Creating a Celerra File System by Size Using AVMWhen you create a Celerra file system using the AVM feature, you do not have to create volumes before setting up the file system. AVM allocates space to the file system from the storage pool you specify, and automatically creates any required volumes when it creates the file system.

For information about AVM and storage pools, refer to Celerra File System Concepts on page 7 and storage pools in the Managing Celerra Volumes technical module on the user information CD. For information about TimeFinder/FS as well as SRDF/S, refer to the technical modules Using TimeFinder/FS with Celerra and Using SRDF/S with Celerra for Disaster Recovery on the user information CD.

Use this procedure to create a Celerra file system using AVM.

Action

To create a file system using AVM, use this command syntax:$ nas_fs -name <name> -create size=<size> pool=<pool> [storage=<system_name>]

Where:<name> = the name assigned to a file system<size> = size of a file system in megabytes<pool> = the storage pool for the file system<system_name> = the backend storage system from which to allocate space for the file system

Note: If you do not specify a storage system, any single storage system with available space for the specified pool will be used.

Example:To create a file system using AVM called ufs1, type: $ nas_fs -name ufs1 -create size=10G pool=clar_r5_performance

Output

id = 24name = ufs1acl = 0in_use = Falsetype = uxfsvolume = v121pool = clar_r5_performancemember_of = root_avm_fs_group_3rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs = WRE00100100430-0010, WRE00100100430-0011disks = d7,d13

Note

A file system name must be unique on a particular Celerra Network Server. It can include up to 254 characters, including hyphens ( - ), underscores ( _ ), and periods ( . ), but cannot begin with a hyphen or period. The Celerra Network Server Command Reference Manual on the user information CD identifies the nas_fs command options available for creating a file system.






Create a Mount Point for a Celerra File System on a Data MoverUse this procedure to create a mount point on a Data Mover for a file system.

Action

To create a mount point on a Data Mover, use this command syntax:$ server_mountpoint <movername> -create <pathname>

Where:<movername> = name of the specified Data Mover<pathname> = path to mount pointExample:To create a mount point called server_3, type:$ server_mountpoint server_3 -create /ufs1

Output

server_3: done


Mount a Celerra File System on a Mount Point on a Data MoverPrerequisite: Create a Mount Point for a Celerra File System on a Data Mover on page 22.

To mount a Celerra file system on a Data Mover, you must know the name of the Data Mover that contains the mount point, the name of the mount point on the Data Mover, and the name of the file system you want to mount. You can use the:

◆ nas_server -list command to view Data Mover names.

◆ server_mountpoint <movername> -list command to view mount points on a specified Data Mover.

◆ nas_fs -list command to list existing file systems.

Use this procedure to mount the file system on the mount point on a Data Mover.

File systems are mounted permanently by default. If you unmount a file system temporarily and then reboot the file server, the file system is remounted automatically.

Action

To mount a file system on a mount point on a Data Mover, use this command syntax:$ server_mount <movername> -option <options><fs_name> <mount_point>

Where:<movername> = name of the specified Data Mover<options> = specifies mount options, separated by commas<fs_name> = named file system to be mounted<mount_point> = path to mount point for the specified Data MoverExample:To mount a file system on a mount point on a Data Mover with a nolock option, type:$ server_mount server_2 -option nolock,accesspolicy=NATIVEufs1 /ufs1

Output

server_2: done


Enabling Access to Celerra File SystemsThis section discusses how to enable access to file systems using the NFS, CIFS, FTP, TFTP, and MPFS protocols.

Enable NFS Access to a Celerra File SystemTo enable NFS protocol access to a Celerra file system, you export the file system explicitly for NFS access from a Data Mover and then, on the client system, mount the exported Celerra file system using the NFS tools available on that system.

For how to enable NFS access to a Celerra file system, refer to the Managing NFS Access to the Celerra Network Server technical module on the user information CD.

For how to mount the Celerra file system on the client system, refer to the documentation provided with the client’s NFS software.

Enable CIFS Access to a Celerra File SystemTo enable CIFS protocol access to a Celerra file system, you share (export) the Celerra file system explicitly for CIFS access from a Data Mover and configure CIFS service and CIFS user authentication on the Celerra Network Server. Then, on the client system, connect to the shared Celerra file system using the CIFS tools available on that system.

For how to enable CIFS access to a Celerra file system, refer to the guides Configuring CIFS on Celerra and Managing Celerra for the Windows Environment available on the user information CD.

For how to set up and manage simultaneous CIFS and NFS access to a Celerra file system, refer to the Configuring CIFS on Celerra for a Multiprotocol Environment technical module on the user information CD. Note that the Celerra Network Server lets you map Windows users and groups to UNIX UIDs (user IDs) and GIDs (group IDs) to provide users with seamless access to shared file system data; for details, refer to the Configuring Celerra User Mapping technical module on the user information CD.

For how to connect to the Celerra file system from the client, refer to the documentation provided with the client’s CIFS software.

Enable FTP Access to a Celerra File SystemTo enable FTP protocol access to a Celerra file system, you configure the FTP feature on the Celerra Network Server to allow FTP access to the file system by specified users.

For how to configure the FTP feature on the Celerra Network Server, refer to the Using FTP on Celerra Network Server technical module on the user information CD.









Enable TFTP Access to a Celerra File SystemTo enable TFTP protocol access to a Celerra file system, you configure the FTP feature on the Celerra Network Server to allow TFTP access to the file system by specified users.

For how to configure the TFTP feature on the Celerra Network Server, refer to the Using TFTP on Celerra Network Server technical module on the user information CD.

Enable MPFS Access to a Celerra File SystemYou must enable MPFS (multiplex file system) protocol access to a Celerra Network Server if you intend to use the EMC Celerra HighRoad product.

For how to enable MPFS access to a Celerra Network Server, refer to the Using HighRoad on Celerra technical module on the user information CD.





Managing Celerra File SystemsThe following sections explain procedures you use to manage Celerra file systems. Unless otherwise noted, the procedures apply to all Celerra models.

Listing All Celerra File Systems on a Celerra Network ServerUse this procedure to view a list of all file systems on a Celerra Network Server.

Action

To view a list of all Celerra file systems on a Celerra Network Server, type:$ nas_fs -list

Output

id inuse type acl volume name server1 n 1 0 166 root_fs_12 y 1 0 168 root_fs_2 13 y 1 0 170 root_fs_3 24 y 1 0 172 root_fs_4 35 y 1 0 174 root_fs_5 46 n 1 0 176 root_fs_67 n 1 0 178 root_fs_78 n 1 0 180 root_fs_89 n 1 0 182 root_fs_910 n 1 0 184 root_fs_1011 n 1 0 186 root_fs_1112 n 1 0 188 root_fs_1213 n 1 0 190 root_fs_1314 n 1 0 192 root_fs_1415 n 1 0 194 root_fs_1516 y 1 0 196 root_fs_common 4,3,2,117 n 5 0 245 root_fs_ufslog18 n 1 0 247 ufs1

Note

Column definitions:id — ID of the file system (assigned automatically).inuse — Whether the file system registered into the mount table of a Data Mover; y indicates yes, n indicates no.type — Type of file system.acl — Access control value for the file system.volume — Volume on which the file system resides.name — Name assigned to the file system.server — ID of the Data Mover accessing the file system.


Listing Configuration Information for a Celerra File SystemUse this procedure to list the configuration information for a Celerra file system.

Checking Celerra File System CapacityYou can check the space capacity of a Celerra file system or of all file systems on a specific Data Mover. Checking the space capacity tells you the total, free, and used disk space allocated to the file systems.

You can also check the inode capacity of a Celerra file system or of all file systems on a specific Data Mover. A specific number of inodes is allocated to a file system when you create it. The number of inodes defines the number of filenames and directory names (folder names) the file system can contain. Checking inode capacity tells you the total, used, and free number of inodes for the file systems.

Action

To view configuration information for a specific file system, use this command syntax:$ nas_fs -info <fs_name>

Where:<fs_name> = name of the file system for which you want to view informationExample:To view configuration information on ufs1, type:$ nas_fs -info ufs1

Output

id = 18name = ufs1acl = 0in_use = Falsetype = uxfsvolume = mtv1pool =rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs =002806000209-006,002806000209-007,002806000209-008,002806000209-009disks = d3,d4,d5,d6


Checking Space Capacity of a File SystemUse this procedure to check the space capacity of a file system.

Checking Space Capacity of All File Systems on a Data MoverUse this procedure to check the space capacity of all file systems on a Data Mover.

Action

To see the total amount of space allocated to a file system, the amount of free and used disk space available to the file system, and the percentage of space used, use this command syntax:$ nas_fs -size <fs_name>

Where:<fs_name> = name of the file systemExample:To view the total space available on ufs1, type:$ nas_fs -size ufs1

Output

total = 8654 avail = 8521 used = 133 ( 1% ) (sizes in MB)

Note

In this example, the file system is mounted

Action

To see the total amount of space allocated to every file system on a Data Mover, the free and used space available to each file system, and the percentage of used space, use this command syntax:$ server_df <movername>

Where:<movername> = name of the specified Data MoverExample:$ server_df server_2

Output

server_2:Filesystem kbytes used avail capacity Mounted onroot_fs_common 15360 1392 13968 9% /.etc_commonufs1 34814592 54240 34760352 0% /ufs1ufs2 104438672 64 104438608 0% /ufs2ufs1_snap1 34814592 64 34814528 0% /ufs1_snap1root_fs_2 15360 224 15136 1% /


Checking Inode Capacity of a File SystemUse this procedure to check the inode capacity of a file system on a Data Mover.

Action

To see the number of inodes allocated to a file system on a Data Mover, the number of used and available inodes for the file system, and the percentage of total inodes in use by the file system, use this command syntax:$ server_df <movername> -inode <fs_name>Where:<movername> = name of the specified Data Mover<fs_name> = name of the file systemExample:To view the inode allocation, use, and availability on ufs2, type:$ server_df server_2 -inode ufs2

Output

server_2:Filesystem inodes used avail capacity Mounted onufs2 12744766 8 12744758 0% /ufs2

Note

Column definitions:Filesystem — Name of the file system.inodes — Total number of inodes allocated to the file system.used — Number of inodes in use by the file system.avail — Number of free inodes available for use by the file system.capacity — Percentage of total inodes in use.Mounted on — Name of the mount point of the file system on the Data Mover.


Checking Inode Capacity of All File Systems on a Data MoverUse this procedure to check the inode capacity of all file systems on a Data Mover.

Finding Out If a File System Has an Associated Storage PoolAll file systems created using the AVM feature have an associated storage pool. For details, refer to Creating a Celerra File System by Size Using AVM on page 21.

Use this procedure to find out if a file system has an associated storage pool.

Action

To see the number of inodes allocated to each file system on a Data Mover, the number of used and available inodes for each file system, and the percentage of total inodes in use by each file system, use this command syntax:$ server_df <movername> -inodeWhere:<movername> = name of the specified Data MoverExample:$ server_df server_2 -inode

Output

server_2:Filesystem inodes used avail capacity Mounted onroot_fs_common 7870 14 7856 0% /.etc_commonufs1 4250878 1368 4249510 0% /ufs1ufs2 12744766 8 12744758 0% /ufs2ufs1_snap1 4250878 8 4250870 0% /ufs1_snap1root_fs_2 7870 32 7838 0% /

Note

Column definitions:Filesystem - Name of the file system.inodes — Total number of inodes allocated to the file system.used — Number of inodes in use by the file system.avail — Number of free inodes available for use by the file system.capacity — Percentage of total inodes in use.Mounted on — Name of the mount point of the file system on the Data Mover.

Action

Check the file system configuration to determine if the file system has an associated storage pool:$ nas_fs -info <fs_name>

Where:<fs_name> = name of the file system whose configuration you want to checkExample:To check the file system configuration for associated storage pool on ufs1, type:$ nas_fs -info ufs1


Extending a Celerra File SystemIf a Celerra file system nears its maximum capacity, you can increase its size by extending the file system. You extend a file system either by volume (by adding volumes to it), or by size (using the AVM feature to add space if the file system has an associated storage pool).

Note: For information about AVM, refer to Celerra File System Concepts on page 7. For information about storage pools, refer to the Managing Celerra Volumes technical module on the user information CD.

The following sections explain procedures for extending a Celerra file system by volume or by size using AVM.

Extending a Celerra File System by Volume

Note: If the file system has an associated storage pool, we recommend that you extend it using the procedure Extending a Celerra File System by Size Using AVM on page 33. To figure out if a file system has an associated storage pool, use the procedure Finding Out If a File System Has an Associated Storage Pool on page 30.

You can extend a file system with any unused disk volume, slice volume, stripe volume, or metavolume. Adding volume space to a file system adds the space to the metavolume on which it is built. So when you extend a file system, the total size of its underlying metavolume is also extended.

If the metavolume underlying the file system is made up of stripe volumes, you should extend the file system with stripe volumes of the same size and type.

Output

id = 24name = ufs1acl = 0in_use = Falsetype = uxfsvolume = v121pool = clar_r5_performancemember_of = root_avm_fs_group_3rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs = WRE00100100430-0010,

WRE00100100430-0011disks = d7,d13

Note

The pool line includes the value clar_r5_performance, which is the storage pool associated with the file system.If the pool line contains no value (is blank), the file system has no associated storage pool.



Use this procedure to extend a file system by adding a volume to it.

Step Action

1. Check the size of the file system before extending it:$ nas_fs -size <fs_name>Example:$ nas_fs -size ufs1Output:total = 67998 avail = 67997 used = 0 (0%) (sizes in MB)volume: total = 69048 (sizes in MB)

2. To extend the file system, use this command syntax:$ nas_fs -xtend <fs_name> <volume_name>

Where:<fs_name> = the name of the file system. <volume_name> = the volume you want to add to the file system. The name of the volume you added appears in the volume output field.Example:$ nas_fs -xtend ufs1 emtv2b

Output:id = 18name = ufs1acl = 0in_use = Truetype = uxfsvolume = mtv1, emtv2bprofile =rw_servers= server_2ro_servers=rw_vdms =ro_vdms =symm_devs = 002804000190-0034,002804000190-0035,002804000190-0036,002804000190-0037,002804000190-0040,002804000190-0041,002804000190-0042,002804000190-0043disks = d3,d4,d5,d6,d15,d16,d17,d18 disk=d3 symm_dev=002804000190-0034 addr=c0t3l8-15-0 server=server_2 disk=d4 symm_dev=002804000190-0035 addr=c0t3l9-15-0 server=server_2 disk=d5 symm_dev=002804000190-0036 addr=c0t3l10-15-0 server=server_2 disk=d6 symm_dev=002804000190-0037 addr=c0t3l11-15-0 server=server_2 disk=d15 symm_dev=002804000190-0040 addr=c0t4l4-15-0 server=server_2 disk=d16 symm_dev=002804000190-0041 addr=c0t4l5-15-0 server=server_2 disk=d17 symm_dev=002804000190-0042 addr=c0t4l6-15-0 server=server_2 disk=d18 symm_dev=002804000190-0043 addr=c0t4l7-15-0 server=server_2


Extending a Celerra File System by Size Using AVMAll file systems created using the AVM feature have an associated storage pool. For details, refer to Creating a Celerra File System by Size Using AVM on page 21.

If a file system has an associated storage pool, you can extend it by specifying only the file system name and the extension size. AVM handles the extension by creating and allocating new storage from the storage pool already associated with the file system.

Note: For information about AVM, refer to Celerra File System Concepts on page 7. For information about storage pools, refer to Storage Pools in the Managing Celerra Volumes technical module on the user information CD.

Use this procedure to extend a file system by size using AVM.

3. Check the size of the file system after extending it to see how much the size increased:$ nas_fs -size <fs_name>Where: <fs_name> = the name of the file systemExample:$ nas_fs -size ufs1Output:total = 138096 avail = 138096 used = 0 ( 0% ) (sizes in MB)volume: total = 138096 (sizes in MB)

Step Action

1. Check the file system configuration to confirm that the file system has an associated storage pool. If you see a pool defined in the output, the file system was created with AVM and has an associated storage pool.$ nas_fs -info <fs_name>

Example:$ nas_fs -info ufs1

Output:id = 24name = ufs1acl = 0in_use = Falsetype = uxfsvolume = v121pool = clar_r5_performancemember_of = root_avm_fs_group_3rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs = WRE00100100430-0010, WRE00100100430-0011disks = d7,d13

Step Action



2. To extend the file system, use the command syntax:$ nas_fs -xtend <fs_name> size=<size> [pool=<pool>] [storage=<system_name>]

Where:<fs_name> = the name of the file system<size> = amount of space you want to add to the file system<pool> = storage pool for extending the file system <system_name> = the backend storage system from which to allocate space for extending the file system

Note: If you do not specify a storage system, the default storage system is the one on which the file system resides. If the file system spans multiple storage systems, the default is to use all the storage systems on which the file system resides.

Enter the size in gigabytes by typing <number>G (for example, 250G) or in megabytes by typing <number>M (for example, 500M).

Example:$ nas_fs -xtend ufs1 size=10G

Output:id = 24name = ufs1acl = 0in_use = Falsetype = uxfsvolume = v121pool = clar_r5_performancemember_of = root_avm_fs_group_3rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs = WRE00100100430-0010, WRE00100100430-0011disks = d7,d13,d19,d25

3. Check the size of the file system after extending it to confirm that the size increased.$ nas_fs -size <fs_name>Where: <fs_name> = the name of the file system.

Example:$ nas_fs -size ufs1

Output:total = 138096 avail = 138096 used = 0 ( 0% ) (sizes in MB)volume: total = 138096 (sizes in MB)

Step Action


Exporting a Celerra File SystemUse this procedure to export a Celerra file system.

Action

To export a file system on an NFS client, use this command syntax:$ server_export <movername> -Protocol <Protocol> -option <options>/<pathname>

Where:<movername> = name of the specified Data Mover<Protocol> = NFS or CIFS. NFS is default<options> = NFS export options<pathname> = pathname of the file system to the export

Example:To export the file system ufs2 on an NFS client, type:$ server_export server_3 -Protocol nfs -option root=10.1.1.1 /ufs2

To export a Celerra file system on an CIFS client, use this command syntax:$ server_export <movername> -Protocol <Protocol> -name <sharename>/<pathname>

Where:<movername> = name of the specified Data Mover<Protocol> = NFS or CIFS. NFS is default<sharename> = CIFS sharename<pathname> = pathname of the file system to exportExample:To export the file system ufs2 on an CIFS client, type:$ server_export server_3 -Protocol cifs -name ufs2 /ufs2

Output

For both command examples:server_3: done

Note

NFS protocol is the default.


Renaming a Celerra File SystemUse this procedure to rename a Celerra file system.

Deleting a Celerra File System and Freeing Its Disk Space

!CAUTION!Deleting a file system deletes all the data of the file system. In case of file system with checkpoints, all the checkpoints should be deleted prior to the file system deletion. The file system deletion does not delete data from business continuance volumes associated with the file system.

Use this procedure to delete a Celerra file system and free its associated disk space.

Action

To rename a file system, use this command syntax:$ nas_fs -rename <old_name> <new_name>

Where:<old_name> = current name of file system<new_name> = new name of file systemExample:To rename a file system ufs1, type:$ nas_fs -rename ufs ufs1

Output

id = 18name = ufs1acl = 0in_use = Falsetype = uxfsvolume = mtv1rw_servers=ro_servers=rw_vdms =ro_vdms =symm_devs =002806000209-006,002806000209-007,002806000209-008,002806000209-009disks = d3,d4,d5,d6

Step Action

1. Back up from the Celerra file system all data that you want to keep.


2. Check the file system configuration to determine if the file system has an associated storage pool (you will need this information for a later step):$ nas_fs -info <fs_name>

If the pool output line includes a value (is not blank), the file system has an associated storage pool.If the Celerra file system does not have an associated storage pool, continue with step 3.If the Celerra file system has an associated storage pool, continue with step 4.

3. Determine and note the name of the metavolume on which the Celerra file system is built (you will need this name for a later step):$ nas_fs -info <fs_name>

Refer to Listing Configuration Information for a Celerra File System on page 27for example output.The volume field contains the name of the metavolume. The disks field lists the disks that provide storage to the file system.

4. If the Celerra file system has associated checkpoints, permanently unmount and then delete the checkpoints and their associated volumes.

5. If the Celerra file system has associated business continuance volumes, break the connection between (unmirror) the file system and its business continuance volumes. For how to unmirror a business continuance volume, refer to the Using TimeFinder/FS with Celerra technical module on the user information CD.

6. Permanently disable client access to the Celerra file system:For NFS-exported file systems, use a command of the form:$ server_export <movername> -Protocol nfs -unexport -perm <pathname>

Where:<movername> = name of the specified Data Mover<pathname> = NFS entry

For CIFS-shared file systems, use a command of the form:$ server_export <movername> -Protocol cifs -unexport <sharename>

7. Permanently unmount the Celerra file system from its associated Data Movers:$ server_umount <movername> -perm <fs_name>

Refer to Unmounting a File System Permanently on page 42 for more information.

8. Delete the Celerra file system from the Celerra Network Server:$ nas_fs -delete <fs_name>

If the Celerra file system had an associated storage pool, you are finished. As part of the file system delete operation, AVM deletes all underlying volumes and frees the space for use by other file systems.If the Celerra file system had no associated storage pool, continue with step 9. The volumes underlying the file system were created manually and must be manually deleted.

Step Action



9. Delete the metavolume on which the Celerra file system was created:$ nas_volume -delete <meta_volume_name>

Refer to deleting a metavolume or stripe volume to in the Managing Celerra Volumes technical module on the user information CD for more information.

10. If the metavolume included stripe volumes, delete all stripe volumes associated with the metavolume until the disk space is free:$ nas_volume -delete <stripe_volume_name>

Refer to deleting a metavolume or strip volume to the Managing Celerra Volumes technical module on the user information CD for more information.

11. If the metavolume included slice volumes, delete all slice volumes associated with the metavolume until the disk space is free:$ nas_volume -delete <slice_volume_name>

Refer to deleting a slice volume in the Managing Celerra Volumes technical module on the user information CD for more information.

12. To finish freeing the disk space, check for slice volumes, stripe volumes, and metavolumes that are not in use (n in inuse column) with these commands:$ nas_volume -list

$ nas_slice -list

As needed, delete unused volumes until you free all of the disk space you want to free.

Step Action





Listing Mount Points on a Data MoverUse this procedure to list the mount points on a Data Mover.

Listing All Celerra File Systems Mounted on a Data MoverYou can view a list of all Celerra file systems currently mounted on a specific Data Mover and the options assigned to each mounted file system.

Use this procedure to view a list of file systems mounted on a Data Mover.

Action

To list the mount points on a Data Mover, use this command syntax:$ server_mountpoint <movername> -list

Where:<movername> = name of the specified Data MoverExample:To list mount points on server_3, type:$ server_mountpoint server_3 -list

Output

server_3 :/.etc_common/ufs1/ufs1_snap1

Action

To view a list of all Celerra file systems mounted on a Data Mover, use this command syntax:$ server_mount <movername>

Where:<movername> = name of the specified Data MoverExample:To list all Celerra file Systems mounted on server_3, type:$ server_mount server_3

Output

server_3:fs2 on /fs2 uxfs,perm,rwfs1 on /fs1 uxfs,perm,rwroot_fs_3 on / uxfs,perm,rw


Mounting a File System for Read/Write AccessA Celerra file system can be mounted for read/write access on only one Data Mover. If a file system is mounted for read/write access, it cannot be mounted for read-only access on any Data Mover.

Use this procedure to mount a file system for read/write access on a Data Mover.

Mounting a File System for Read-Only AccessA Celerra file system can be mounted for read-only access on several Data Movers. If a file system is mounted for read-only access, it cannot be mounted for read/write access on any Data Mover.

Use this procedure to mount a file system for read-only access on Data Movers..

Step Action

1. If the file system is mounted on any Data Mover, use the server_umount command to unmount the file system permanently from all Data Movers on which it is mounted.For how to unmount a file system from a Data Mover, refer to Unmounting a Celerra File System from a Data Mover on page 41.

2. Use the server_mount command to mount the file system on a Data Mover for read/write access.For how to mount a file system on a Data Mover, refer to Mount a Celerra File System on a Mount Point on a Data Mover on page 23.

Step Action

1. If the file system is mounted for read/write access on a Data Mover, use the server_umount command to unmount the file system permanently from that Data Mover.For how to unmount a file system from a Data Mover, refer to Unmounting a Celerra File System from a Data Mover on page 41.

2. Use the server_mount command to mount the file system for read-only access on each Data Mover from which you want to provide read-only access to the file system.For how to mount a file system on a Data Mover, refer to Mount a Celerra File System on a Mount Point on a Data Mover on page 23.


Unmounting a Celerra File System from a Data MoverYou can unmount a single Celerra file system from a Data Mover temporarily or permanently, or unmount all Celerra file systems from a Data Mover at once. You specify the file system you want to unmount either by the file system name or the name of its mount point on the Data Mover.

Unmounting a File System TemporarilyUse this procedure to temporarily unmount a file system from a Data Mover. The option for a temporary unmount, -temp, is the default and need not be specified as part of the command.

Note: It is good practice to unexport NFS exports and unshare CIFS shares of the file system before unmounting the file system from the Data Mover.

Action

To temporarily unmount a file system by specifying the file system name, use this command syntax:$ server_umount <movername> -temp <fs_name>

Where:<movername> = name of the specified Data Mover<fs_name> = name of file system to unmountExample:To temporarily unmount a file system named ufs1, type:$ server_umount server_2 -temp ufs1

Output



Unmounting a File System PermanentlyUse this procedure to permanently unmount a file system from a Data Mover.


Action

To permanently unmount a file system from a Data Mover by specifying the file system name, use this command syntax:$ server_umount <movername> -perm <fs_name>

Where: <movername> = name of the specified Data Mover<fs_name> = name of file system to unmountExample:To permanently unmount a file system named mtv1, type:$ server_umount server_2 -perm ufs1

To permanently unmount a file system from a Data Mover by specifying the name of its mount point, type:$ server_umount <movername> -perm <mount_point>

Where:<movername> = name of the specified Data Mover<mount_point> = path to mount point for the specified Data MoverExample:To permanently unmount a file system to server_2, type:$ server_umount server_2 -perm /ufs1

Output



Unmounting All Celerra File Systems from a Data MoverUse the procedures in this section to temporarily or permanently unmount all file systems from a Data Mover.

Unmounting All File Systems on a Data Mover TemporarilyUse this procedure to temporarily unmount all file systems on a Data Mover. The option for a temporary unmount, -temp, is the default and need not be specified as part of the command.


Action

To temporarily unmount all file systems on a Data Mover, use this command syntax:$ server_umount <movername> -temp all

or$ server_umount <movername> all

Where:<movername> = name of the specified Data MoverExample:To temporarily unmount all file systems on server_2, type:$ server_umount server_2 -temp all

Output

Unmount all file systems output:server_2: done


Unmounting All File Systems on a Data Mover Permanently

!CAUTION!Permanently unmounting all file systems from a Data Mover should be done with caution because the operation deletes the contents of the mount table. To re-establish client access to the file systems after this operation, you must rebuild the mount table by remounting each file system on the Data Mover.

Use this procedure to permanently unmount all file systems on a Data Mover. We recommend that you unexport NFS exports and unshare CIFS shares of the file systems before unmounting the file systems from the Data Mover.

Action

To permanently unmount all file systems on a Data Mover, use this command syntax:$ server_umount <movername> -perm all

Where:<movername> = name of the specified Data MoverExample:To permanently unmount all file systems on server_2, type:$ server_umount server_2 -perm all

Output

Unmount all file systems output:server_2: done


Changing the Size Threshold for All File SystemsUse this procedure to change the file system size threshold for all file systems on a Celerra Network Server.

Changing the Size Threshold for a Data MoverUse this procedure to change the file system size threshold for all file systems on one Data Mover.

Step Action

1. To change the size threshold for all file systems, use this command syntax:$ server_param <movername> -facility <facility_name> -modify <param_name> -value <new_value>

Where:<movername> = name of Data Mover<facility_name> = facility for the parameters<param_name> = name of parameter <new_value> = new value for parameter Example:To change the size threshold for all file systems to 85%, type:$ server_param ALL -facility file -modify fsSizeThreshold -value 85

2. To make the change take effect, reboot the Data Movers:$ server_cpu ALL -reboot now

Step Action

1. $ server_param <movername> -facility <facility_name> -modify <param_name> -value <new_value>

Where:<movername> = name of Data Mover<facility_name> = facility for the parameters<param_name> = name of parameter <new_value> = new value for parameter Example:To change the size threshold for all file systems on server_2, type:$ server_param server_2 -facility file -modify fsSizeThreshold -value 85

2. To make the change take effect, reboot the Data Mover:$ server_cpu <data_mover> -reboot now

Example:$ server_cpu server_2 -reboot now


Enhancing File Read and Write PerformanceThe Celerra Network Server includes internal mechanisms for enhancing read/write performance characteristics for particular types of files.

The read mechanisms are designed to optimize reads of large files and are turned on by default on the server. These mechanisms can improve read speed up to 100 percent on large files. This default mechanism should be turned off if the read access pattern for files in the file system involves small random accesses. For how to turn off this mechanism for a file system, refer to Turning Off the Read Prefetch Mechanism for a File System on page 46. For how to turn off this mechanism for all file systems on a Data Mover, refer to Turning Off and On the Read Prefetch Mechanism for a Data Mover on page 47.

The write mechanisms are designed to improve performance for applications, such as databases, that have many connections to a large file. These mechanisms can enhance databases access through the NFS protocol by 30 percent or more. The mechanism is off by default and can be turned on by file system. For how to turn on the write performance improvement mechanism, refer to Turning On the Uncached Write Mechanism for a File System on page 48.

Turning Off the Read Prefetch Mechanism for a File SystemUse this procedure to turn off the read prefetch performance improvement mechanism for a file system.

Action

To turn off the read prefetch mechanism for a file system, use this command syntax:$ server_mount <movername> -option <options>,noprefetch<fs_name> <mount_point>

Where:<movername> = name of the specified Data Mover<options> = specifies mount options, separated by commas<fs_name> = name of the file system<mount_point> = path to mount point for the specified Data MoverExample:To read the prefetch mechanism on ufs1, type:$ server_mount server_3 -option rw,noprefetch ufs1 /ufs1

Output

server_3: done


Turning Off and On the Read Prefetch Mechanism for a Data MoverUse this procedure to turn off the read prefetch performance improvement mechanism for all file systems on one Data Mover.

Step Action

1. To turn off the read prefetch mechanism for all file systems on one Data Mover, use this command syntax:server_param <movername> -facility <facility_name> -modify <param_name> -value <new_value>

Where:<movername> = name of the data mover<facility_name> = facility for the parameters<param_name> = name of parameter <new_value> = new value for parameterExample:To turn off the prefetch mechanism for all file systems on server_2, type$ server_param server_2 -facility file -modify prefetch -value 0

To turn on the read prefetch mechanism for all file systems on one Data Mover, use this command syntax:server_param <movername> -facility <facility_name> -modify <param_name> -value <new_value>

Example:To turn on the read prefetch mechanism for all file systems on server_2, type:$ server_param server_2 -facility file -modify prefetch -value 1

2. To make the change take effect, reboot the Data Mover:$ server_cpu <data_mover> -reboot now

Example:$ server_cpu server_2 -reboot now


Turning On the Uncached Write Mechanism for a File SystemUse this procedure to turn on the uncached write performance improvement mechanism for a file system.

Action

To turn on the write performance improvement mechanism for a file system, use this command syntax:$ server_mount <movername> -option <options>,uncached<fs_name> <mount_point>

Where:<movername> = name of the specified Data Mover<options> = specifies mount options, separated by commas<fs_name> = name of the file system<mount_point> = path to mount point for the specified Data MoverExample:$ server_mount server_3 -option rw,uncached ufs1 /ufs1

Output

server_3: done


Finding and Fixing File System Storage ErrorsThe following sections explain procedures you use check file system consistency.

Starting and Monitoring a File System CheckUse this procedure to start and monitor fsck on a specified file system.

Starting ACL Check on a File SystemTo start ACL check on a specified file system, use the following procedure.

Action

To start and monitor fsck on a specified file system, use this command syntax:$ nas_fsck -start <name> -monitor

Where:<name> = name of the specified file systemExample:To start fsck on ufs1 and monitor the progress, type: $ nas_fsck -start ufsl -monitor

Output

id = 27name = ufs1volume = mtv1fsck_server = server_2inode_check_percent = 10..20..30..40..60..70..80..100directory_check_percent = 0..0..100used_ACL_check_percent = 100free_ACL_check_status = Donecylinder_group_check_status = In Progress..Done

Action

To start ACL check a specified file system, use this command syntax:$ nas_fsck -start <name> -aclchkonly

Where:<name> = name of the specified file systemExample:To start ACL check on ufs1 and monitor the progress, type: $ nas_fsck -start ufsl -aclchkonly

Output

ACLCHK: in progress for file system ufs1


Listing Current File System ChecksTo list all current file system checks, use the following procedure.

Displaying Information on a Single File System CheckTo display information on a file system check for a single file system, use the following procedure.

Action

To list current file system checks, use this command syntax:$ nas_fsck -list

Example:To list current file system checks, type: $ nas_fsck -list

Output

id type state volume name server23 1 FSCK 134 ufs2 427 1 ACLCHK 144 ufs1 1

Action

To display information about a file system check on a single file system, use this command syntax:$ nas_fsck -info <name>

Where:<name> = name of the specified file systemExample:To display information about file system check for ufs2, type: $ nas_fsck -info ufs2

Output

name = ufs2id = 23volume = v134fsck_server = server_5inode_check_percent = 100directory_check_percent = 100used_ACL_check_percent = 100free_ACL_check_status = Donecylinder_group_check_status = In Progress


Displaying Information on all Current File System ChecksTo display information on all file system checks currently running, use the following procedure.

Action

To display information on all file system checks currently running, use this command syntax:$ nas_fsck -info -all

Example:To display information on all files system checks currently running, type: $ nas_fsck -info -all

Output

name = ufs2id = 23volume = v134fsck_server = server_5inode_check_percent = 30directory_check_percent = 0used_ACL_check_percent = 0free_ACL_check_status = Not Startedcylinder_group_check_status = Not Started

name = ufs1id = 27volume = mtv1fsck_server = server_2inode_check_percent = 100directory_check_percent = 0used_ACL_check_percent = 0free_ACL_check_status = Not Startedcylinder_group_check_status = Not Started


Managing Celerra Nested Mount File SystemsThe following sections explain procedures you use to manage Celerra nested mount file systems.

Creating a Nested Mount File System A NMFS must be initially created. Use this procedure to create a nested mount file system.

Action

To create a nested mount file system, use this command syntax:$ nas_fs -name <name> -type <nmfs> -create

Where:<name> = name of the nested mount file system<type> = type of file systemExample:To create a nested mount file system, nmfs1, type: $ nas_fs -name nmfs4 -type nmfs -create

Output

id = 719name = nmfs4acl = 0in_use = Falsetype = nmfsworm = offvolume = 0pool =rw_servers=ro_servers=rw_vdms =ro_vdms =stor_devs =disks =

Note

There is no volume allocation.


Mounting a Nested Mount File System Use the following procedure to mount the NMFS. A NMFS must be mounted as read/only.

Deleting the Nested Mount File SystemTo delete a NMFS, you must first permanently unmount all component file systems in the NMFS and then you permanently unmount the NMFS. Refer to Unmounting a File System Permanently on page 42 for more information.

After permanently unmounting all the component file systems and the NMFS, delete the file system using the nas_fs command.

Action

To mount the NMFS, use this command syntax:$ server_mount <movername> -o ro <nmfs name> /<nmfs path>

Where:<movername> = name of the specified Data Mover<nmfs name> = name of nested mount file system<pathname> = path to mount pointExample:To mount a NMFS named nmfs4, type$ server_mount server_3 -o ro nmfs4 /nmfs4

Output

server_3: done

Step Action

1. To permanently unmount the NMFS, use this command syntax:$ server_umount <movername> -perm <fs_name>

Where: <movername> = name of the specified Data Mover<fs_name> = name of file system to unmountExample:To permanently unmount a file system named nmfs4, type:$ server_umount server_3 -p nmfs4

Output:server_3 : done


2. To delete the NMFS, use this command syntax:$ nas_fs -delete <fs_name>

Where:<fs_name> = name of file system to be deleted

Example:To delete a nested mount file system, nmfs4, type:$ nas_fs -d nmfs4

Output:id = 719name = nmfs4acl = 0in_use = Falsetype = nmfsworm = offvolume = 0pool =rw_servers=ro_servers=rw_vdms =ro_vdms =stor_devs =disks =

Step Action


Adding an Existing File System to the NMFSIf you are creating a new component file system to add to the NMFS, see Creating a Celerra File System by Size Using AVM on page 21. If the file system you want to add to the NMFS already exists, use the following procedure to add the file system to a NMFS.

Step Action

1. If the existing file system is mounted, permanently unmount the file system using this command syntax:To permanently unmount a file system from a Data Mover by specifying the file system name, use this command syntax:$ server_umount <movername> -perm <fs_name>

Where: <movername> = name of the specified Data Mover<fs_name> = name of file system to unmountExample:To permanently unmount a file system named fs1, type:$ server_umount server_2 -perm fs1

To permanently unmount a file system from a Data Mover by specifying the name of its mount point, type:$ server_umount <movername> -perm <mount_point>

Where:<movername> = name of the specified Data Mover<mount_point> = path to mount point for the specified Data MoverExample:To permanently unmount a file system to server_2, type:$ server_umount server_2 -perm /fs1

Output:For both command examples:server_2: done

2. Mount the file system in the nested mount file system using this command syntax:$ server_mount <movername> -option <options> <component file system name> /<nmfs pathname>/<component file system name>

Where:<movername> = name of the specified Data Mover<options> = specifies mount options, separated by commas<component file system name> = name of the component file system<nmfs pathname> = pathname of the nested mount file system<component file system name> = name of file system to be mounted<mount_point> = path to mount point for the specified Data MoverExample:To mount a file system on a mount point on a Data Mover with a nolock option, type:$ server_mount server_3 fs5 /nmfs4/fs5

Output:server_2: done


Exporting a Component File SystemTo export a component file system, refer to Exporting a Celerra File System on page 35. When exporting a component file system, use the full path to the component file system. If you are a CIFS client, the component file system has to be shared.

Removing a Component File SystemYou can both permanently and temporarily remove component file systems from the NMFS. To permanently remove a component file system from the NMFS, refer to Unmounting a File System Permanently on page 42. To temporarily remove a component file system from the NMFS, refer to Unmounting a File System Temporarily on page 41.

Moving a Nested Mount File SystemYou can move a nested mount file system from one Data Mover to another Data Mover. To do this, perform the following steps:

1. Permanently unmount each of the component file systems. Refer to Unmounting a File System Permanently on page 42 for more information.

2. Permanently unmount the NMFS. Refer to Unmounting a File System Permanently on page 42 for more information.

3. Mount the NMFS on the new Data Mover. Refer to Mounting a Nested Mount File System on page 53 for more information.

4. Mount each component file system on the NMFS on the new Data Mover. Refer to Adding an Existing File System to the NMFS on page 55.


Troubleshooting File SystemsThis section provides guidance for troubleshooting problems you may encounter using Celerra file systems. You can query the EMC WebSupport database for related problem information, obtain release notes, or report a Celerra technical problem to EMC at Powerlink™, EMC’s secure extranet site, at http://powerlink.EMC.com. For additional information about using Powerlink and resolving problems, refer to the Problem Resolution Roadmap technical module on the user information CD.

Error MessagesThis section contains two tables to assist in troubleshooting Celerra file system problems. Table 5 presents error messages, their definition, and corrective action. Refer to the Celerra Network Server Error Messages Guide on the user information CD for additional information on error messages.

Note: There can be more than one symptom for the same problem and, in some cases more than one cause.

Table 5 File System Error and Event Messages

Server error log Definition Corrective Action

filesystem is mounted, can not delete

You may be trying to delete a mounted file system.

Verify that this is the correct file system to be deleted, permanently unmount the file system, and then retry.

filesystem is not mounted

You may be attempting to execute a command to file system that must be mounted.

Mount the file system, and then retry.

filesystem unavailable for read_write mount

The file system is already mounted by another Data Mover.

Verify the list of mounted file systems.

item is currently in use by movername

The file system is still mounted.

Unmount the file system, and then retry.

Mount Point Name [name] is not valid. Please Re-enter

The mount point was not entered correctly or does not exist.

When entering the mount point name, the slash (/) that precedes the mount point name may have been omitted. Type a slash before entering the mount point name, and then retry. If the error message reappears, check your list of mount points.








must unmount from server(s): movername

You may be trying to execute a file system check against a mounted file system.

Unmount the file system, and then retry.

no such file or directory

The value being typed either does not exist or does not exist as typed (typing error).

Check that you are entering the correct value and ensure that the uppercase and lowercase letters match.Note: All mount points begin with a forward slash (/).

Path busy: filesystem fsname is currently mounted on mountpoint

A file system is already using the mount point you are attempting to mount.

Create a new mount point, or unmount the file system, and then retry.

requires root command You may be attempting to execute a command against a root file system or volume to which you do not have access.

Select a non root volume or file system, and then retry. Alternately, retry the command as a root command by prepending it with the word root, using this command syntax: root<command>.

server_x: Device busy You may have tried to execute the same command more than once.

Check the list of mounted file systems to verify that the path is already successfully mounted.

server_x: No such file or directory

The mount point may not exist on the specified Data Mover.

Confirm that the mount point is on the specified Data Mover.

undefined netgroup The netgroup you are trying to export for is not recognized.

Enter the name of the netgroup into the system, and then retry.For information about entering the netgroup name into the system, refer to the Configuring Celerra Naming Services technical module.

Filesystem size threshold (##%) crossed (fs <fs_name>

Note: This message appears in the event log on the server.

The amount of space in use in the file system (<fs_name>) has reached its maximum threshold (##%). The threshold default is 90 percent space in use. File system performance may degrade when a file system is almost full.

Extend the file system to increase its available space. This reduces the percentage of used space.For how to extend a file system, refer to Extending a Celerra File System on page 31.

When you create a new file in the NMFS root directory, a file exists error appears.

NMFS root directory is read/only.

Do not try to create files or folders in the NMFS root directory.

Table 5 File System Error and Event Messages (continued)

Server error log Definition Corrective Action


Known Problems and LimitationsThis section explains known limitations and restrictions of the Celerra management applications. Refer to the Release Notes for additional and late-breaking information about using the management applications.

Table 6 describes known problems that might occur when using Celerra file systems and workarounds.

Table 6 File Systems Known Problems and Workarounds

Known Problem Symptom Workaround

The Celerra Manager does not provide an interface for renaming a file system.

To rename a file system, you can enter the appropriate CLI command in the CLI command entry page available in Celerra Manager, or directly in the CLI.

You are unable to mount a file system.

There are many probable causes for this scenario. Many will provide an error message, though occasionally, there will not be one. In this case, the mount table entry already exists.

Perform server_mount -all to activate all entries in the mount table. Obtain a list of mounted file systems, and then observe the entries. If the file system in question is already mounted (temporary or permanent), perform the necessary steps to unmount it, and then retry.

An unmounted file system reappears in the mount table after a system reboot.

The file system may have been temporarily unmounted prior to reboot.

Perform a permanent unmount to remove the entry from the mount table.

When you create a new file in the NMFS root directory, a file exists error appears.

NMFS root directory is read/only.

Do not try to create files or folders in the NMFS root directory.


Related InformationFor specific information related to the features and functionality described in this technical module, refer to the following publications, which are available on the user information CD.

◆ Celerra Network Server Command Reference Manual

◆ Online Celerra man pages

◆ Celerra Network Server Parameters Guide

For general information on other EMC Celerra publications, refer to the Celerra Network Server User Information CD, which is available at Powerlink at http://powerlink.emc.com.

Want to Know More?EMC Customer Education Courses are designed to help you learn how EMC storage products work together and integrate within your environment in order to maximize your entire infrastructure investment. EMC Customer Education features online and hands-on training in state-of-the-art labs conveniently located throughout the world. EMC customer training courses are developed and delivered by EMC experts. For course information and registration, refer to EMC Powerlink, our customer and partner website on http://powerlink.emc.com.







Appendix: GID SupportThe Celerra Network Server software supports 32-bit GID (group IDs) on both NFS and CIFS file systems. This support enables a maximum GID value of 2147483647 (approximately 2 billion).

Restrictions for GID SupportWith GID support, the following restrictions apply:

◆ Enabling 16-bit GID support on a Data Mover does not decrease its maximum of 64,000 GIDs per file system. Regardless of the GID support setting (32-bit or 16-bit), there is a maximum limit of 64,000 GIDs per file system.

◆ File systems with 16-bit and 32-bit GIDs can co-exist on a single Data Mover. Changing the gid32 parameter setting from 1 to 0 allows you to create file systems with 16-bit GIDs without disabling 32-bit GIDs on file systems already created with the parameter set to 1. Conversely, changing the gid32 parameter value from 0 to 1 allows you to create file systems with 32-bit GIDs without disabling 16-bit GID support on existing file systems.

◆ You cannot convert file systems created with 32-bit GID support to use 16-bit GIDs, and cannot convert file systems created with 16-bit GID support to use 32-bit GIDs. The 32-bit GID support works only for file systems created with the parameter set to 1, and the 16-bit GID support works only for file systems created with the parameter set to 0.

◆ If you back up a file system with 32-bit GIDs, you risk truncating the GID values when the data is restored if you use any of the following server_archive formats:

• emctar up to 31-bit

• ustar up to 21-bit

• cpio up to 15-bit

• bcpio up to 16-bit

• sv4cpio up to 21-bit

• sv4crc up to 21-bit

• tar up to 18-bit

If you use these server_archive formats to back up file systems with 32-bit GIDs, a message appears indicating that the UIDs are being forced to 0 and the GIDs are being forced to 1.

◆ Some backup applications have restrictions. Ensure that the application can handle 32-bit UIDs/GIDs.

◆ There is no NAS command you can use to find out whether a file system supports 16-bit or 32-bit GIDs.


Index

AAccess

CIFS and Celerra file systems 24FTP and Celerra file systems 3MPFS and Celerra file systems 4NFS and Celerra file systems 24TFTP and Celerra file systems 3

Automatic Volume ManagementCelerra Manager support 12command line interface support 12creating file system using 21definition 4extending file system using 33

AVM, see Automatic Volume Management 4

CCapacity

checking file system 27Data Mover 27

Celerra Native Manager 17Celerra WORM 12Character support, international 6CIFS access 24CIFS and Celerra file systems 24Considerations, file system 8Creating

file system 20

DData Mover

checkingfree space 27system capacity 27

creating a mount point 22free space 27, 28listing

mount points 39mounted file systems 39

unmounting file systempermanent 42temporary 41

Deletingfile system 36

Disk volumefreeing file system space 36

Displaying mounted file systems 39

EExporting

file system 35Extending file system 31

FFile system

capacity 27CIFS access 24configuration information 27creating 20definition 4deleting 36displaying mounted 39exporting 35extending 31extending capacity 9free space 28freeing disk space allocated to 36FTP access 24inode capacity 29listing all 26managing 26mounting 10mounting on Data Mover 23MPFS access 25NFS access 24permanent

mount 23planning considerations 8quotas 6, 11renaming 36requirements 8size guidelines 9TFTP access 25troubleshooting 57types 7unmount all 43

permanent 44temporary 43

unmounting from Data Mover 41permanent 42temporary 41

used space 28file system check

listing 50starting ACL check 49starting and monitoring 49

File system size threshold 11change

for all file systems 45for Data Mover 45

change for Data Mover 47fsSizeThreshold parameter 45, 47

Free spaceall file systems on Data Mover 28checking on Data Mover 27file system 28

fscklisting file system checks 50starting ACL check 49starting and monitoring 49

fsSizeThreshold parameter 45, 47FTP access to Celerra file systems 3, 24

64 of 66 Managing Celerra File SystemsVersion 5.4

GGID support

restrictions 61

HHighRoad 25

IInformation

training 60Inode capacity 29Interface choices

Celerra Native Manager 17command line interface 17

International character support 6Interoperability Matrix 16

LListing all file systems 26LUN, definition 4

MMetavolume

definition 4Monitoring file system size 11Mount point

creating 22listing 39

Mountingfile system 10file system on Data Mover 23

MPFS access to Celerra file systems 4, 25

NNested Mount File System 15Nested mount file system

definition 4managing 52

NFS access 24NFS and Celerra file systems 24

PParameters

fsSizeThreshold 45, 47Performance enhancements

file read/write 46turning off

read for Data Mover 47read for file system 46

turning on write for file system 48Powerlink 16

QQuotas for file system 6, 11

RRenaming

file system 36Restrictions

Celerra file systems 6GID support 61Symmetrix volumes 6TimeFinder/FS 6

Sserver_mount command 46, 48Size Threshold

changing for a data mover 45changing for all file systems 45

Slice volumedefinition 5

Storage poolassociated with file system 30definition 5finding out if file system has 30

System storage errorsfinding and fixing 49

TTFTP

definition 3TFTP access to Celerra file systems 3, 25Trivial File Transfer Protocol. See TFTPTroubleshooting 57

file system 57

UUnicode characters 6Unmounting file systems from Data Mover

permanent 44temporary 43

Used space, file system 28User interface choices 17

VVolume

metavolumedefinition 4

slice volumedefinition 5


Notes

Copyright © 1998–2005 EMC Corporation. All rights reserved.

EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without notice.

THE INFORMATION IN THIS PUBLICATION IS PROVIDED “AS IS.” EMC CORPORATION MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.

About This Technical ModuleAs part of its effort to continuously improve and enhance the performance and capabilities of the Celerra Network Server product line, EMC from time to time releases new revisions of Celerra hardware and software. Therefore, some functions described in this document may not be supported by all revisions of Celerra software or hardware presently in use. For the most up-to-date information on product features, see your product release notes. If your Celerra system does not offer a function described in this document, please contact your EMC representative for a hardware upgrade or software update.

Comments and Suggestions About the DocumentationYour suggestions will help us continue to improve the accuracy, organization, and overall quality of the user publications. Please send a message to [email protected] with your opinions of this document.

Version 5.4 66 of 66

managing celerra file systems

Documents