File warn-nfs-udp.patch of Package nfs-utils
Index: nfs-utils-1.1.3/utils/mount/nfs.man
===================================================================
--- nfs-utils-1.1.3.orig/utils/mount/nfs.man
+++ nfs-utils-1.1.3/utils/mount/nfs.man
@@ -438,6 +438,8 @@ Specifying
.B proto=udp
forces all traffic types to use UDP.
.IP
+.B Before using NFS over UDP, please refer to the section WARNINGS below.
+.IP
If the
.B proto
mount option is not specified, the
@@ -452,6 +454,8 @@ The
option is an alternative to specifying
.BR proto=udp.
It is included for compatibility with other operating systems.
+.IP
+.B Before using NFS over UDP, please refer to the section WARNINGS below.
.TP 1.5i
.B tcp
The
@@ -799,7 +803,7 @@ dropped requests, but this can result in
and server load.
.P
However, UDP can be quite effective in specialized settings where
-the network’s MTU is large relative to NFS’s data transfer size (such
+the network's MTU is large relative to NFS's data transfer size (such
as network environments that enable jumbo Ethernet frames). In such
environments, trimming the
.B rsize
@@ -811,6 +815,8 @@ in a single frame) is advised. This r
the loss of a single MTU-sized network frame results in the loss of
an entire large read or write request.
.P
+Please see also the WARNINGS section below.
+.P
TCP is the default transport protocol used for all modern NFS
implementations. It performs well in almost every conceivable
network environment and provides excellent guarantees against data
@@ -1179,6 +1185,83 @@ of Access Control Lists that are semanti
NFS version 4 ACLs are not fully compatible with POSIX ACLs; as such,
some translation between the two is required
in an environment that mixes POSIX ACLs and NFS version 4.
+.SH WARNINGS
+Using NFS over UDP on high-speed links such as Gigabit
+.BR "can cause silent data corruption" .
+.P
+The problem can be triggered at high loads, and is caused by problems in
+IP fragment reassembly. NFS read and writes typically transmit UDP packets
+of 4 Kilobytes or more, which have to be broken up into several fragments
+in order to be sent over the Ethernet link, which limits packets to 1500
+bytes by default. This process happens at the IP network layer and is
+called fragmentation.
+.P
+In order to identify fragments that belong together, IP assigns a 16bit
+.I IP ID
+value to each packet; fragments generated from the same UDP packet
+will have the same IP ID. The receiving system will collect these
+fragments and combine them to form the original UDP packet. This process
+is called reassembly. The default timeout for packet reassembly is
+30 seconds; if the network stack does not receive all fragments of
+a given packet within this interval, it assumes the missing fragment(s)
+got lost and discards those it already received.
+.P
+The problem this creates over high-speed links is that it is possible
+to send more than 65536 packets within 30 seconds. In fact, with
+heavy NFS traffic one can observe that the IP IDs repeat after about
+5 seconds.
+.P
+This has serious effects on reassembly: if one fragment gets lost,
+another fragment
+.I from a different packet
+but with the
+.I same IP ID
+will arrive within the 30 second timeout, and the network stack will
+combine these fragments to form a new packet. Most of the time, network
+layers above IP will detect this mismatched reassembly - in the case
+of UDP, the UDP checksum, which is a 16 bit checksum over the entire
+packet payload, will usually not match, and UDP will discard the
+bad packet.
+.P
+However, the UDP checksum is 16 bit only, so there is a chance of 1 in
+65536 that it will match even if the packet payload is completely
+random (which very often isn't the case). If that is the case,
+silent data corruption will occur.
+.P
+This potential should be taken seriously, at least on Gigabit
+Ethernet.
+Network speeds of 100Mbit/s should be considered less
+problematic, because with most traffic patterns IP ID wrap around
+will take much longer than 30 seconds.
+.P
+It is therefore strongly recommended to use
+.BR "NFS over TCP where possible" ,
+since TCP does not perform fragmentation.
+.P
+If you absolutely have to use NFS over UDP over Gigabit Ethernet,
+some steps can be taken to mitigate the problem and reduce the
+probability of corruption:
+.TP +1.5i
+.I Jumbo frames:
+Many Gigabit network cards are capable of transmitting
+frames bigger than the 1500 byte limit of traditional Ethernet, typically
+9000 bytes. Using jumbo frames of 9000 bytes will allow you to run NFS over
+UDP at a page size of 8K without fragmentation. Of course, this is
+only feasible if all involved stations support jumbo frames.
+.IP
+To enable a machine to send jumbo frames on cards that support it,
+it is sufficient to configure the interface for a MTU value of 9000.
+.TP +1.5i
+.I Lower reassembly timeout:
+By lowering this timeout below the time it takes the IP ID counter
+to wrap around, incorrect reassembly of fragments can be prevented
+as well. To do so, simply write the new timeout value (in seconds)
+to the file
+.BR /proc/sys/net/ipv4/ipfrag_time .
+.IP
+A value of 2 seconds will greatly reduce the probability of IPID clashes on
+a single Gigabit link, while still allowing for a reasonable timeout
+when receiving fragmented traffic from distant peers.
.SH FILES
.TP 1.5i
.I /etc/fstab
Index: nfs-utils-1.1.3/utils/mount/nfsmount.c
===================================================================
--- nfs-utils-1.1.3.orig/utils/mount/nfsmount.c
+++ nfs-utils-1.1.3/utils/mount/nfsmount.c
@@ -263,6 +263,9 @@ parse_options(char *old_opts, struct nfs
if (!strcmp(opteq+1, "udp")) {
nfs_pmap->pm_prot = IPPROTO_UDP;
mnt_pmap->pm_prot = IPPROTO_UDP;
+ fprintf(stderr,
+ "Using NFS over UDP can cause data corruption.\n"
+ "Please refer to the WARNINGS section of the nfs(5) manual page.\n");
#if NFS_MOUNT_VERSION >= 2
data->flags &= ~NFS_MOUNT_TCP;
} else if (!strcmp(opteq+1, "tcp") &&
Index: nfs-utils-1.1.3/utils/mount/stropts.c
===================================================================
--- nfs-utils-1.1.3.orig/utils/mount/stropts.c
+++ nfs-utils-1.1.3/utils/mount/stropts.c
@@ -389,14 +389,28 @@ static struct mount_options *nfs_rewrite
po_remove_all(options, "proto");
}
if (strcmp(option, "udp") == 0) {
+ static int once = 0;
nfs_server.pmap.pm_prot = IPPROTO_UDP;
po_remove_all(options, "proto");
+ if (!once)
+ fprintf(stderr,
+ "Using NFS over UDP can cause data corruption.\n"
+ "Please refer to the WARNINGS section of the nfs(5) manual page.\n");
+ once = 1;
+
}
}
p = po_rightmost(options, "tcp", "udp");
switch (p) {
case PO_KEY2_RIGHTMOST:
nfs_server.pmap.pm_prot = IPPROTO_UDP;
+ {static int once = 0;
+ if (!once)
+ fprintf(stderr,
+ "Using NFS over UDP can cause data corruption.\n"
+ "Please refer to the WARNINGS section of the nfs(5) manual page.\n");
+ once = 1;
+ }
break;
case PO_KEY1_RIGHTMOST:
nfs_server.pmap.pm_prot = IPPROTO_TCP;
@@ -565,11 +579,24 @@ static int nfs_try_nfs23mount(struct nfs
static int nfs_try_nfs4mount(struct nfsmount_info *mi)
{
char **extra_opts = mi->extra_opts;
+ char *proto;
+ static int once = 0;
if (po_join(mi->options, extra_opts) == PO_FAILED) {
errno = EIO;
return 0;
}
+ if (po_contains(mi->options, "udp"))
+ proto = "udp";
+ else
+ proto = po_get(mi->options, "proto");
+ if (proto && strcmp(proto, "udp") == 0 && !once) {
+ fprintf(stderr,
+ "Using NFS over UDP can cause data corruption.\n"
+ "Please refer to the WARNINGS section of the nfs(5) manual page.\n");
+ once=1;
+ }
+
if (verbose)
printf(_("%s: text-based options: '%s'\n"),
