< NFS

NFS/Troubleshooting

Dedicated article for common problems and solutions.

Server-side issues

exportfs: /etc/exports:2: syntax error: bad option list

Make sure to delete all space from the option list in /etc/exports.

exportfs: requires fsid= for NFS export

As not all filesystems are stored on devices and not all filesystems have UUIDs (e.g. FUSE), it is sometimes necessary to explicitly tell NFS how to identify a filesystem. This is done with the fsid option:

/etc/exports
/srv/nfs client(rw,sync,crossmnt,fsid=0)
/srv/nfs/music client(rw,sync,fsid=10)

Group/GID permissions issues

If NFS shares mount fine, and are fully accessible to the owner, but not to group members; check the number of groups that user belongs to. NFS has a limit of 16 on the number of groups a user can belong to. If you have users with more than this, you need to enable the manage-gids start-up flag on the NFS server:

/etc/nfs.conf
[mountd]
manage-gids=y

"Permission denied" when trying to write files as root

  • If you need to mount shares as root, and have full r/w access from the client, add the no_root_squash option to the export in /etc/exports:
/var/cache/pacman/pkg 192.168.1.0/24(rw,no_subtree_check,no_root_squash)
  • You must also add no_root_squash to the first line in /etc/exports:
/ 192.168.1.0/24(rw,fsid=root,no_root_squash,no_subtree_check)

"RPC: Program not registered" when showmount -e command issued

Make sure that nfs-server.service and rpcbind.service are running on the server site, see systemd. If they are not, start and enable them.

Also make sure NFSv3 is enabled. showmount does not work with NFSv4-only servers.

UDP mounts not working

nfs-utils disabled serving NFS over UDP in version 2.2.1. Arch core updated to 2.3.1 on 21 Dec 2017 (skipping over 2.2.1.) If UDP stopped working then, add under in . Then restart nfs-server.service.

Timeout with big directories

Since nfs-utils version 1.0.x, every subdirectory is checked for permissions. This can lead to timeout on directories with a "large" number of subdirectories, even a few hundreds.

To disable this behaviour, add the option to /etc/exports to the share directory.

Client-side issues

mount.nfs4: No such device

Make sure the nfsd kernel module has been loaded.

mount.nfs4: Invalid argument

Enable and start nfs-client.target and make sure the appropriate daemons (nfs-idmapd, rpc-gssd, etc) are running on the server.

mount.nfs4: Network is unreachable

Users making use of systemd-networkd or NetworkManager might notice NFS mounts are not mounted when booting.

Force the network to be completely configured by enabling or . This may slow down the boot-process because fewer services run in parallel.

mount.nfs4: an incorrect mount option was specified

This can happen if using the option without nfs-client.target and/or running. Starting and enabling those services should resolve the issue.

Unable to connect from OS X clients

When trying to connect from an OS X client, you will see that everything is ok in the server logs, but OS X will refuse to mount your NFS share. You can do one of two things to fix this:

  • On the NFS server, add the option to the share in /etc/exports and re-run .

... OR ...

  • On the OS X client, add the option to the mount command line. You can also set as a default client mount option in :
/etc/nfs.conf
nfs.client.mount.options = resvport

Using the default client mount option should also affect mounting the share from Finder via "Connect to Server...".

Unreliable connection from OS X clients

OS X's NFS client is optimized for OS X Servers and might present some issues with Linux servers. If you are experiencing slow performance, frequent disconnects and problems with international characters edit the default mount options by adding the line to on your Mac client. More information about the mount options can be found in the OS X mount_nfs man page.

Intermittent client freezes when copying large files

If you copy large files from your client machine to the NFS server, the transfer speed is very fast, but after some seconds the speed drops and your client machine intermittently locks up completely for some time until the transfer is finished.

Try adding as a mount option on the client (e.g. in ) to fix this problem.

NFSv4

If you use Kerberos (), make sure the client and server clocks are correct. Using ntpd or systemd-timesyncd is recommended. Also, check that the canonical name for the server as resolved on the client (see Domain name resolution) matches the name in the server's NFS principal.

NFSv3 and earlier

nfs-utils versions 1.2.1-2 or higher use NFSv4 by default, resulting in NFSv3 shares failing on upgrade. The problem can be solved by using either mount option or on the command line:

# mount.nfs remote target directory -o ...,vers=3,...
# mount.nfs remote target directory -o ...,nfsvers=3,...

or in :

remote target directory nfs ...,vers=3,... 0 0
remote target directory nfs ...,nfsvers=3,... 0 0

mount.nfs: Protocol not supported

This error occurs when you include the export root in the path of the NFS source. For example:

# mount SERVER:/srv/nfs4/media /mnt
mount.nfs4: Protocol not supported

Use the relative path instead:

# mount SERVER:/media /mnt

Permissions issues

If you find that you cannot set the permissions on files properly, make sure the user/user group are both on the client and server.

If all your files are owned by , and you are using NFSv4, on both the client and server, you should ensure that the nfs-idmapd.service has been started.

On some systems detecting the domain from FQDN minus hostname does not seem to work reliably. If files are still showing as after the above changes, edit , ensure that Domain is set to . For example:

Problems with Vagrant and synced_folders

If you get an error about unuspported protocol, you need to enable NFS over UDP on your host (or make Vagrant use NFS over TCP.) See #UDP mounts not working.

If Vagrant scripts are unable to mount folders over NFS, installing the net-tools package may solve the issue.

Performance issues

This NFS Howto page has some useful information regarding performance. Here are some further tips:

Diagnose the problem

  • Htop should be your first port of call. The most obvious symptom will be a maxed-out CPU.
  • Press F2, and under "Display options", enable "Detailed CPU time". Press F1 for an explanation of the colours used in the CPU bars. In particular, is the CPU spending most of its time responding to IRQs, or in Wait-IO (wio)?

Close-to-open/flush-on-close

Symptoms: Your clients are writing many small files. The server CPU is not maxed out, but there is very high wait-IO, and the server disk seems to be churning more than you might expect.

In order to ensure data consistency across clients, the NFS protocol requires that the client's cache is flushed (all data is pushed to the server) whenever a file is closed after writing. Because the server is not allowed to buffer disk writes (if it crashes, the client will not realise the data was not written properly), the data is written to disk immediately before the client's request is completed. When you are writing lots of small files from the client, this means that the server spends most of its time waiting for small files to be written to its disk, which can cause a significant reduction in throughput.

See this excellent article or the nfs manpage for more details on the close-to-open policy. There are several approaches to solving this problem:

The nocto mount option

If all of the following conditions are satisfied:

  • The export you have mounted on the client is only going to be used by the one client.
  • It does not matter too much if a file written on one client does not immediately appear on other clients.
  • It does not matter if after a client has written a file, and the client thinks the file has been saved, and then the client crashes, the file may be lost.

Use the nocto mount option, which will disable the close-to-open behavior.

The async export option

Does your situation match these conditions?

  • It is important that when a file is closed after writing on one client, it is:
    • Immediately visible on all the other clients.
    • Safely stored on the server, even if the client crashes immediately after closing the file.
  • It is not important to you that if the server crashes:
    • You may lose the files that were most recently written by clients.
    • When the server is restarted, the clients will believe their recent files exist, even though they were actually lost.

In this situation, you can use instead of in the server's /etc/exports file for those specific exports. See the exports manual page for details. In this case, it does not make sense to use the mount option on the client.

Buffer cache size and MTU

Symptoms: High kernel or IRQ CPU usage, a very high packet count through the network card.

This is a trickier optimisation. Make sure this is definitely the problem before spending too much time on this. The default values are usually fine for most situations.

See this article for information about I/O buffering in NFS. Essentially, data is accumulated into buffers before being sent. The size of the buffer will affect the way data is transmitted over the network. The Maximum Transmission Unit (MTU) of the network equipment will also affect throughput, as the buffers need to be split into MTU-sized chunks before they are sent over the network. If your buffer size is too big, the kernel or hardware may spend too much time splitting it into MTU-sized chunks. If the buffer size is too small, there will be overhead involved in sending a very large number of small packets. You can use the rsize and wsize mount options on the client to alter the buffer cache size. To achieve the best throughput, you need to experiment and discover the best values for your setup.

It is possible to change the MTU of many network cards. If your clients are on a separate subnet (e.g. for a Beowulf cluster), it may be safe to configure all of the network cards to use a high MTU. This should be done in very-high-bandwidth environments.

See NFS#Performance tuning for more information.

Debugging

Using rpcdebug

Using is the easiest way to manipulate the kernel interfaces in place of echoing bitmasks to /proc.

OptionDescription
-cClear the given debug flags
-sSet the given debug flags
-m moduleSpecify which module's flags to set or clear.
-vIncrease the verbosity of rpcdebug's output
-hPrint a help message and exit. When combined with the -v option, also prints the available debug flags.

For the -m option, the available modules are:

ModuleDescription
nfsdThe NFS server
nfsThe NFS client
nlmThe Network Lock Manager, in either an NFS client or server
rpcThe Remote Procedure Call module, in either an NFS client or server

Examples:

Once the flags are set you can tail the journal for the debug output, usually by running as root or similar.

Using mountstats

The nfs-utils package contains the tool, which can retrieve a lot of statistics about NFS mounts, including average timings and packet size.

Kernel Interfaces

A bitmask of the debug flags can be echoed into the interface to enable output to syslog; 0 is the default:

/proc/sys/sunrpc/nfsd_debug
/proc/sys/sunrpc/nfs_debug
/proc/sys/sunrpc/nlm_debug
/proc/sys/sunrpc/rpc_debug

Sysctl controls are registered for these interfaces, so they can be used instead of echo:

At runtime the server holds information that can be examined:

A rundown of (the userspace tool pretty-prints this info):

NFS debug flags

/usr/include/linux/nfs_fs.h
/*
 * NFS debug flags
 */
#define NFSDBG_VFS              0x0001
#define NFSDBG_DIRCACHE         0x0002
#define NFSDBG_LOOKUPCACHE      0x0004
#define NFSDBG_PAGECACHE        0x0008
#define NFSDBG_PROC             0x0010
#define NFSDBG_XDR              0x0020
#define NFSDBG_FILE             0x0040
#define NFSDBG_ROOT             0x0080
#define NFSDBG_CALLBACK         0x0100
#define NFSDBG_CLIENT           0x0200
#define NFSDBG_MOUNT            0x0400
#define NFSDBG_FSCACHE          0x0800
#define NFSDBG_PNFS             0x1000
#define NFSDBG_PNFS_LD          0x2000
#define NFSDBG_STATE            0x4000
#define NFSDBG_ALL              0xFFFF

See also

gollark: You could, in theory, probably implement it with a macro(n).
gollark: And lambdas are `|| println!("bee")`, `match` just uses normal blocks.
gollark: It's pattern matching; `Ordering::Less` is a simple pattern.
gollark: `match` is an expression, yes, but it does NOT have a map.
gollark: Well, I can* explæin it.
This article is issued from Archlinux. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.