LDP
/
old-www
1
1
Fork 0
Code Issues Pull Requests Releases Wiki Activity
old-www/HOWTO/NFS-HOWTO/interop.html

980 lines
18 KiB
HTML
Raw Permalink Blame History

<HTML
><HEAD
><TITLE
>Using Linux NFS with Other OSes</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.76b+
"><LINK
REL="HOME"
TITLE="Linux NFS-HOWTO"
HREF="index.html"><LINK
REL="PREVIOUS"
TITLE="Troubleshooting"
HREF="troubleshooting.html"></HEAD
><BODY
CLASS="SECT1"
BGCOLOR="#FFFFFF"
TEXT="#000000"
LINK="#0000FF"
VLINK="#840084"
ALINK="#0000FF"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="3"
ALIGN="center"
>Linux NFS-HOWTO</TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="bottom"
><A
HREF="troubleshooting.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="80%"
ALIGN="center"
VALIGN="bottom"
></TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="bottom"
>&nbsp;</TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="INTEROP">8. Using Linux NFS with Other OSes</H1
><P
> Every operating system, Linux included, has quirks and deviations
in the behavior of its NFS implementation -- sometimes because
the protocols are vague, sometimes because they leave gaping
security holes. Linux will work properly with all major vendors'
NFS implementations, as far as we know. However, there may be
extra steps involved to make sure the two OSes are communicating
clearly with one another. This section details those steps.
</P
><P
> In general, it is highly ill-advised to attempt to use a Linux
machine with a kernel before 2.2.18 as an NFS server for non-Linux
clients. Implementations with older kernels may work fine as
clients; however if you are using one of these kernels and get
stuck, the first piece of advice we would give is to upgrade
your kernel and see if the problems go away. The user-space NFS
implementations also do not work well with non-Linux clients.
</P
><P
> Following is a list of known issues for using Linux together with
major operating systems.
</P
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AIX">8.1. AIX</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AIXSERVER">8.1.1. Linux Clients and AIX Servers</H3
><P
> The format for the <TT
CLASS="FILENAME"
>/etc/exports</TT
> file for our example in <A
HREF="server.html"
>Section 3</A
> is:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
> /usr slave1.foo.com:slave2.foo.com,access=slave1.foo.com:slave2.foo.com
/home slave1.foo.com:slave2.foo.com,rw=slave1.foo.com:slave2.foo.com
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AIXCLIENTS">8.1.2. AIX clients and Linux Servers</H3
><P
> AIX uses the file <TT
CLASS="FILENAME"
>/etc/filesystems</TT
> instead of <TT
CLASS="FILENAME"
>/etc/fstab</TT
>.
A sample entry, based on the example in <A
HREF="client.html"
>Section 4</A
>, looks like this:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>/mnt/home:
dev = "/home"
vfs = nfs
nodename = master.foo.com
mount = true
options = bg,hard,intr,rsize=1024,wsize=1024,vers=2,proto=udp
account = false
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
><P
> <P
></P
><OL
TYPE="i"
><LI
><P
> Version 4.3.2 of AIX, and possibly earlier versions as well,
requires that file systems be exported with
the <TT
CLASS="USERINPUT"
><B
>insecure</B
></TT
> option, which
causes NFS to listen to requests from
insecure ports (i.e., ports above 1024, to which non-root users can
bind). Older versions of AIX do not seem to require this.
</P
></LI
><LI
><P
> AIX clients will default to mounting version 3 NFS over TCP.
If your Linux server does not support this, then you may need
to specify <TT
CLASS="USERINPUT"
><B
>vers=2</B
></TT
> and/or
<TT
CLASS="USERINPUT"
><B
>proto=udp</B
></TT
> in your mount options.
</P
></LI
><LI
><P
> Using netmasks in <TT
CLASS="FILENAME"
>/etc/exports</TT
>
seems to sometimes cause clients
to lose mounts when another client is reset. This can be fixed
by listing out hosts explicitly.
</P
></LI
><LI
><P
> Apparently automount in AIX 4.3.2 is rather broken.
</P
></LI
></OL
>
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="BSD">8.2. BSD</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="BSDSERVER">8.2.1. BSD servers and Linux clients</H3
><P
> BSD kernels tend to work better with larger block sizes.
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="BSDCLIENT">8.2.2. Linux servers and BSD clients</H3
><P
> Some versions of BSD may make requests to the server from insecure ports,
in which case you will need to export your volumes with the
<TT
CLASS="USERINPUT"
><B
>insecure</B
></TT
>
option. See the man page for <EM
>exports(5)</EM
>
for more details.
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="TRU64">8.3. Tru64 Unix</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="TRU64SERVER">8.3.1. Tru64 Unix Servers and Linux Clients</H3
><P
> In general, Tru64 Unix servers work quite smoothly with Linux clients.
The format for the <TT
CLASS="FILENAME"
>/etc/exports</TT
> file for our example in <A
HREF="server.html"
>Section 3</A
> is:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>
/usr slave1.foo.com:slave2.foo.com \
-access=slave1.foo.com:slave2.foo.com \
/home slave1.foo.com:slave2.foo.com \
-rw=slave1.foo.com:slave2.foo.com \
-root=slave1.foo.com:slave2.foo.com
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
><P
> (The <TT
CLASS="USERINPUT"
><B
>root</B
></TT
> option is listed in the last
entry for informational purposes only; its use is not recommended
unless necessary.)
</P
><P
> Tru64 checks the <TT
CLASS="FILENAME"
>/etc/exports</TT
> file every time there is a mount request
so you do not need to run the <B
CLASS="COMMAND"
>exportfs</B
> command; in fact on many
versions of Tru64 Unix the command does not exist.
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="TRU64CLIENT">8.3.2. Linux Servers and Tru64 Unix Clients</H3
><P
> There are two issues to watch out for here. First, Tru64 Unix mounts
using Version 3 NFS by default. You will see mount errors if your
Linux server does not support Version 3 NFS. Second, in Tru64 Unix
4.x, NFS locking requests are made by
<TT
CLASS="COMPUTEROUTPUT"
>daemon</TT
>. You will therefore
need to specify the <TT
CLASS="USERINPUT"
><B
>insecure_locks</B
></TT
> option on all volumes you export
to a Tru64 Unix 4.x client; see the <B
CLASS="COMMAND"
>exports</B
> man pages for details.
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="HPUX">8.4. HP-UX</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="HPUXSERVER">8.4.1. HP-UX Servers and Linux Clients</H3
><P
> A sample <TT
CLASS="FILENAME"
>/etc/exports</TT
> entry on HP-UX looks like this:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>/usr -ro,access=slave1.foo.com:slave2.foo.com
/home -rw=slave1.foo.com:slave2.fo.com:root=slave1.foo.com:slave2.foo.com
</PRE
></FONT
></TD
></TR
></TABLE
>
(The <TT
CLASS="USERINPUT"
><B
>root</B
></TT
> option is listed in the last entry for informational
purposes only; its use is not recommended unless necessary.)
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="HPUXCLIENT">8.4.2. Linux Servers and HP-UX Clients</H3
><P
> HP-UX diskless clients will require at least a kernel version 2.2.19
(or patched 2.2.18) for device files to export correctly. Also, any
exports to an HP-UX client will need to be exported with the
<TT
CLASS="USERINPUT"
><B
>insecure_locks</B
></TT
> option.
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="IRIX">8.5. IRIX</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="IRIXSERVER">8.5.1. IRIX Servers and Linux Clients</H3
><P
> A sample <TT
CLASS="FILENAME"
>/etc/exports</TT
> entry on IRIX looks like this:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>/usr -ro,access=slave1.foo.com:slave2.foo.com
/home -rw=slave1.foo.com:slave2.fo.com:root=slave1.foo.com:slave2.foo.com
</PRE
></FONT
></TD
></TR
></TABLE
>
(The <TT
CLASS="USERINPUT"
><B
>root</B
></TT
> option is listed in the last entry for informational
purposes only; its use is not recommended unless necessary.)
</P
><P
> There are reportedly problems when using the nohide option on
exports to linux 2.2-based systems. This problem is fixed in the
2.4 kernel. As a workaround, you can export and mount lower-down
file systems separately.
</P
><P
> As of Kernel 2.4.17, there continue to be several minor interoperability
issues that may require a kernel upgrade. In particular:
<P
></P
><UL
><LI
><P
> Make sure that Trond Myklebust's <SPAN
CLASS="APPLICATION"
>seekdir</SPAN
>
(or <SPAN
CLASS="APPLICATION"
>dir</SPAN
>) kernel patch is applied.
The latest version (for 2.4.17) is located at:
</P
><P
> <A
HREF="http://www.fys.uio.no/~trondmy/src/2.4.17/linux-2.4.17-seekdir.dif"
TARGET="_top"
> http://www.fys.uio.no/~trondmy/src/2.4.17/linux-2.4.17-seekdir.dif</A
>
</P
></LI
><LI
><P
> IRIX servers do not always use the same
<TT
CLASS="COMPUTEROUTPUT"
>fsid</TT
> attribute field across
reboots, which results in <TT
CLASS="COMPUTEROUTPUT"
>inode number mismatch</TT
>
errors on a Linux
client if the mounted IRIX server reboots. A patch is available from:
</P
><P
><A
HREF="http://www.geocrawler.com/lists/3/SourceForge/789/0/7777454/"
TARGET="_top"
> http://www.geocrawler.com/lists/3/SourceForge/789/0/7777454/</A
>
</P
></LI
><LI
><P
> Linux kernels v2.4.9 and above have problems reading large directories
(hundreds of files) from exported IRIX XFS file systems that were made
with <TT
CLASS="USERINPUT"
><B
>naming version=1</B
></TT
>.
The reason for the problem can be found at:
</P
><P
> <A
HREF="http://www.geocrawler.com/archives/3/789/2001/9/100/6531172/"
TARGET="_top"
> http://www.geocrawler.com/archives/3/789/2001/9/100/6531172/</A
>
</P
><P
> The naming version can be found by using (on the IRIX server):
</P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
> xfs_growfs -n mount_point
</PRE
></FONT
></TD
></TR
></TABLE
><P
>
The workaround is to export these file systems using the
<TT
CLASS="USERINPUT"
><B
>-32bitclients</B
></TT
>
option in the <TT
CLASS="FILENAME"
>/etc/exports</TT
> file.
The fix is to convert the file system to 'naming version=2'.
Unfortunately the only way to do this is by a
<TT
CLASS="USERINPUT"
><B
>backup</B
></TT
>/<TT
CLASS="USERINPUT"
><B
>mkfs</B
></TT
>/<TT
CLASS="USERINPUT"
><B
>restore</B
></TT
>.
</P
><P
> <B
CLASS="COMMAND"
>mkfs_xfs</B
> on IRIX 6.5.14 (and above)
creates <TT
CLASS="USERINPUT"
><B
>naming version=2</B
></TT
> XFS file
systems by default. On IRIX 6.5.5 to 6.5.13, use:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
> mkfs_xfs -n version=2 device
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
><P
> Versions of IRIX prior to 6.5.5 do not support
<TT
CLASS="USERINPUT"
><B
>naming version=2</B
></TT
> XFS file systems.
</P
></LI
></UL
>
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="IRIXCLIENT">8.5.2. IRIX clients and Linux servers</H3
><P
> Irix versions up to 6.5.12 have problems mounting file systems exported
from Linux boxes - the mount point "gets lost," e.g.,
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
> # mount linux:/disk1 /mnt
# cd /mnt/xyz/abc
# pwd
/xyz/abc
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
><P
> This is known IRIX bug (SGI bug 815265 - IRIX not liking file handles of
less than 32 bytes), which is fixed in <SPAN
CLASS="APPLICATION"
>IRIX 6.5.13</SPAN
>.
If it is not possible
to upgrade to <SPAN
CLASS="APPLICATION"
>IRIX 6.5.13</SPAN
>, then the unofficial
workaround is to force the Linux <B
CLASS="COMMAND"
>nfsd</B
>
to always use 32 byte file handles.
</P
><P
> A number of patches exist - see:
<P
></P
><UL
><LI
><P
> <A
HREF="http://www.geocrawler.com/archives/3/789/2001/8/50/6371896/"
TARGET="_top"
> http://www.geocrawler.com/archives/3/789/2001/8/50/6371896/</A
>
</P
></LI
><LI
><P
> <A
HREF="http://oss.sgi.com/projects/xfs/mail_archive/0110/msg00006.html"
TARGET="_top"
> http://oss.sgi.com/projects/xfs/mail_archive/0110/msg00006.html</A
>
</P
></LI
></UL
>
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SOLARIS">8.6. Solaris</H2
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SOLARISSERVER">8.6.1. Solaris Servers</H3
><P
> Solaris has a slightly different format on the server end from
other operating systems. Instead of
<TT
CLASS="FILENAME"
>/etc/exports</TT
>, the configuration
file is <TT
CLASS="FILENAME"
>/etc/dfs/dfstab</TT
>. Entries are of
the form of a <B
CLASS="COMMAND"
>share</B
> command, where the syntax
for the example in <A
HREF="server.html"
>Section 3</A
> would look like
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>share -o rw=slave1,slave2 -d "Master Usr" /usr
</PRE
></FONT
></TD
></TR
></TABLE
>
and instead of running <B
CLASS="COMMAND"
>exportfs</B
> after editing, you run <B
CLASS="COMMAND"
>shareall</B
>.
</P
><P
> Solaris servers are especially sensitive to packet size. If you
are using a Linux client with a Solaris server, be sure to set
<TT
CLASS="USERINPUT"
><B
>rsize</B
></TT
> and <TT
CLASS="USERINPUT"
><B
>wsize</B
></TT
>
to 32768 at mount time.
</P
><P
> Finally, there is an issue with root squashing on Solaris: root gets
mapped to the user <TT
CLASS="COMPUTEROUTPUT"
>noone</TT
>, which
is not the same as the user <TT
CLASS="COMPUTEROUTPUT"
>nobody</TT
>.
If you are having trouble with file permissions as root on the client
machine, be sure to check that the mapping works as you expect.
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SOLARISCLIENT">8.6.2. Solaris Clients</H3
><P
> Solaris clients will regularly produce the following message:
</P
><TABLE
BORDER="1"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="SCREEN"
>svc: unknown program 100227 (me 100003)
</PRE
></FONT
></TD
></TR
></TABLE
><P
> This happens because Solaris clients, when they mount, try to obtain
ACL information - which Linux obviously does not have. The messages
can safely be ignored.
</P
><P
> There are two known issues with diskless Solaris clients: First, a kernel
version of at least 2.2.19 is needed to get <TT
CLASS="FILENAME"
>/dev/null</TT
> to export
correctly. Second, the packet size may need to be set extremely
small (i.e., 1024) on diskless sparc clients because the clients
do not know how to assemble packets in reverse order. This can be
done from <TT
CLASS="FILENAME"
>/etc/bootparams</TT
> on the clients.
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SUNOS">8.7. SunOS</H2
><P
> SunOS only has NFS Version 2 over UDP.
</P
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SUNOSSERVER">8.7.1. SunOS Servers</H3
><P
> On the server end, SunOS uses the most traditional format for its
<TT
CLASS="FILENAME"
>/etc/exports</TT
> file. The example in <A
HREF="server.html"
>Section 3</A
> would look like:
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><FONT
COLOR="#000000"
><PRE
CLASS="PROGRAMLISTING"
>/usr -access=slave1.foo.com,slave2.foo.com
/home -rw=slave1.foo.com,slave2.foo.com, root=slave1.foo.com,slave2.foo.com
</PRE
></FONT
></TD
></TR
></TABLE
>
</P
><P
> Again, the <TT
CLASS="USERINPUT"
><B
>root</B
></TT
> option is listed for informational
purposes and is not recommended unless necessary.
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SUNOSCLIENT">8.7.2. SunOS Clients</H3
><P
> Be advised that SunOS makes all NFS locking requests
as <TT
CLASS="COMPUTEROUTPUT"
>daemon</TT
>, and
therefore you will need to add the <TT
CLASS="USERINPUT"
><B
>insecure_locks</B
></TT
> option to any
volumes you export to a SunOS machine. See the <B
CLASS="COMMAND"
>exports</B
> man page
for details.
</P
></DIV
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="troubleshooting.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>&nbsp;</TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Troubleshooting</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
>&nbsp;</TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>&nbsp;</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
Reference in New Issue View Git Blame Copy Permalink