6bdcd5b676
examples: add nfs-io example
...
this simple example tool allows to
- creat and unlink files
- mkdir and rmdir directories
- stat
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-23 14:50:09 +01:00
9126c9c036
add uid and gid URL param to specify alternate uid and/or gid
...
This allows to connect with an alternate uid or gid than that
of the current user.
Example:
examples/nfs-ls nfs://10.0.0.1/export?uid=1000&gid=33
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-23 14:03:26 +01:00
1c8b4547ce
add tcp-syncnt URL param to adjust TCP_SYNCNT sockopt
...
This allows indirect support for a configurable connect timeout.
Linux uses a exponential backoff for SYN retries starting
with 1 second.
This means for a value n for TCP_SYNCNT, the connect will
effectively timeout after 2^(n+1)-1 seconds.
Example:
examples/nfs-ls nfs://10.0.0.1/export?tcp-syncnt=1
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-23 14:03:18 +01:00
d2ec73c75a
add URL parsing functions
...
this adds the following 4 functions. see libnfs.h for details.
struct nfs_url *nfs_parse_url_full(struct nfs_context *nfs, char *url);
struct nfs_url *nfs_parse_url_dir(struct nfs_context *nfs, char *url);
struct nfs_url *nfs_parse_url_incomplete(struct nfs_context *nfs, char *url);
void nfs_destroy_url(struct nfs_url *url);
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-23 14:03:00 +01:00
1f3f0d5853
libnfs_zdr_opaque: make valgrind happy
...
valgrind complains about unitialized memory
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-23 12:46:31 +01:00
fadbf8cf56
nfs_creat1_cb: fix memleak
...
data is not freed in case of a failure.
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-17 11:50:00 +01:00
b554c8e8d2
libnfs_authunix_create: make valgrind happy
...
valgrind complains about uninitialized memory.
Signed-off-by: Peter Lieven <pl@kamp.de>
2013-12-17 11:47:59 +01:00
ed09b5676f
Initial support for NSM
2013-11-05 17:37:16 -08:00
558aefc736
New version 1.8
...
Fix nasty memory leak
2013-10-30 18:17:49 -07:00
751770fd43
Dont leak every single buffer we read from the socket.
2013-10-30 18:15:03 -07:00
8733f38d23
Add an async helper function to connect an rpc context to a program/version
2013-10-26 13:16:09 -07:00
3b943d2f68
Add a mechanism to control what the next XID value on a session will be.
...
This makes it possible for multiple processes/contexts to use the same
target and (with some synchronization) avoid XID collissions across processes/contexts.
2013-10-26 07:44:29 -07:00
6fda9871a0
Mix in the pid in the top 16 bits for the initial xid.
...
This helps for users which rapidly fork a lot of processes that then
immediately create a new context (I am looking at you dbench)
to awoid having lots of processes starting and using overlapping xid values.
2013-10-25 16:16:43 -07:00
e7f3a78172
Don't clamp write3 max to 32k
...
We use our own XDR and RPC layer nowadays and do not have
any external dependencies to XDR or RPC.
As such we no longer need to clamp the write size to max 32kb
since we never link to the system rpc/xdr libraries.
(and thus dont have to clamp in case the system library is broken for
pdu's > 32k)
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
2013-10-23 19:18:41 -07:00
cd67ec7d6b
Fix typos
2013-10-23 19:13:00 -07:00
b93082da2b
Make xid be a uint32_t and not either unsigned long or unsigned int
2013-10-21 19:15:11 -07:00
31820d2727
New version 1.7
...
- Allow nested eventloops so that a sync function can be called from a callback.
- Fix a bug in unmarshalling a uint64.
- Add PATHCONF support.
- WIN32/64 updates
- AROS updates
2013-10-20 16:31:32 -07:00
dd97d43aad
Reset the receive buffer processing the received PDU.
...
This allows us to have recursive eventloops and do synchronous
rpc calls from a callback.
2013-10-20 13:16:45 -07:00
c81e7d8921
Make sure the the value is unsigned before we OR it into the uint64_t
...
or else the sign might be extended.
2013-10-20 12:41:56 -07:00
7e66b28947
Merge branch 'master' of github.com:sahlberg/libnfs
2013-05-30 17:59:24 -07:00
4e019b25b5
On Windows, use 65534/65534 instead of 65535/65535 for the default uid/gid
2013-05-30 17:58:29 -07:00
292e2d9097
[win32] - add projects for examples to the solution and make them compile (all compile but nfsclient-bcast)
2013-05-29 18:41:31 +02:00
c5e346e15c
More AROS fixes. AROS compiles and works again.
2013-05-28 22:49:19 -07:00
0556ee05e1
Aros fixes
2013-05-28 22:41:54 -07:00
265276038a
AROS fixes
2013-05-28 22:13:08 -07:00
c022471e52
Add configure test for arpa/inet.h
2013-05-28 21:25:08 -07:00
833d5b9da1
Include ifdef cleanup
2013-05-28 21:20:00 -07:00
bff8fe460d
Some more configure checks for headers
2013-05-28 17:54:12 -07:00
7057e733c1
Add configure checks for sys/socket.h
2013-05-28 17:43:47 -07:00
728970051c
Add configure checks for whether netinet/in.h is available or not
2013-05-28 17:38:37 -07:00
cb5b8be2eb
Try making the initial port used a little more random
2013-05-28 17:22:28 -07:00
67a9f57e67
WIN32 fixes
2013-05-28 16:15:16 -07:00
a60caa0945
New version 1.6
...
Add AROS/Amiga support
Chose initial XID value better to reduce probability for collissions
Fix bug in the initial default credentials and use getgid() instead of -1
2013-05-27 21:35:26 -07:00
43e0e7a7e6
On unix systems, when creating the default authentication token
...
use getgid() as the group instead of -1.
Recent linux knfsd do not allow grp==-1
On windows there are no uid/gids in the traditional sense so there I still specify a default credential of uid==gid==-1 :
rpc->auth = authunix_create("LibNFS", 65535, 65535, 0, NULL);
This is I think the sanest/safest thing to do since most servers will have
special handing of -1 meaning 'nobody' or similar.
This should work on many/most servers and give the user the minimum available
access allowed for 'nobody'.
I think on windows (or AROS for that matter) applications will probably have
to invoke and set the credentials themself explicitely.
Those apps probably, unfortunately, also need to have a configuration
setting to select which uid/gid to use when talking to the server.
(or they could hardcode it)
rpc_set_auth(rpc, libnfs_authunix_create("hostname", uid, gid, 0, NULL))
should do the trick if they call immediately after creating the rpc/nfs context.
But dont set it to 0,0 root/root for uid/gid.
First of all, most servers have root-squash so they will re-map this uid/gid
to 'nobody' internally.
But, if the user uses a server that does not do root-squash, then setting this to 0,0 would mean that your app now access the nfs share as root which is probably not what you want.
2013-05-25 06:24:15 -07:00
0c1a9d7a6b
Fix a bug when returning the filehandle in creat() and return the filehandle
...
of the object we created and not the filehandle of the parent directory
2013-04-15 21:37:03 -07:00
73f4ae7cca
Allow nfs_lookuppath_async to have path[0]=='\0' this happens when we
...
want to create an object straight under the root directory of what we
mounted.
As always, the actual object to create is then a string starting after the \0 byte
2013-04-15 20:42:44 -07:00
4945f30750
we can always include netinet/in.h all platforms have it
2013-04-14 10:34:33 -07:00
108c622a95
more include cleanups it starts to look almost decent now
2013-04-14 10:32:01 -07:00
00748f36c5
more header include cleanups
2013-04-14 10:11:48 -07:00
2142af5d47
WIN32: More ifdef cleanups
2013-04-14 09:02:23 -07:00
622489d36e
WIN32: get rid of some ifdefs
2013-04-14 08:50:38 -07:00
099be710b6
SOCKET: Use recv/send for all socket io on all platforms
...
instead of only on windows and aros and read/write on the others
2013-04-14 08:43:59 -07:00
9a9126c3d9
AROS: add inet_pton emulation and make sure we use recv/send and not read/write
2013-04-13 07:40:46 -07:00
2cdf4fcb07
AROS: it is called IoctlSocket/CloseSocket and WaitSelect on AROS.
...
Additionally add a aros_init_socket() function to open the bsd socket library and set up networking
2013-04-12 06:12:58 -07:00
e77d093c37
AROS: getting closer to link. only a handful of missing symbols now
2013-04-11 18:37:05 -07:00
d7c6e9aaa9
Initial AROS support.
...
The test app doesnt link yet since we are missing a bunch of symbols
but it is a start
2013-04-10 20:28:40 -07:00
a27561934c
randomize the initial rpc xid to avoid reply cache collision
2013-02-25 16:40:29 -08:00
8afcdc959f
New version 1.5
...
- Switch to our own builtin RPC/XDR replacement
- Android support
2012-12-03 06:38:18 -08:00
e803ae5760
Rearrange the header files.
...
Move the headers we install to ./include/nfsc
and start including nfsc/libnfs-zdr.h from all the other header files.
2012-11-28 20:21:53 -08:00
252aa90dc8
[droid] Support building for android
...
These are the changes necessary to build for android. Working status is yet
untested.
2012-11-28 12:04:10 -05:00
f816bf4c76
Merge branch 'zdr'
...
Manually fixyp some merge conflicts in :
lib/init.c
lib/pdu.c
lib/socket.c
nfs/Makefile.am
nfs/nfs.x
nlm/Makefile.am
2012-11-27 18:22:27 -08:00
aab6538bbd
Implement 'rejected_reply' and switch to RFC5531 definitions
...
Switch to using fields and names from RFC5531. that is the standard
for RPC.
2012-11-26 19:42:10 -08:00
dedc4675ba
New version 1.4
...
- Add trackig of freed context and assert on using a context after it has been
freed.
- Windows x64 support and fixes.
- Switch to using our own version of xdr_int64() since the one in libtirpc
crashes on some platforms.
- Fix memory leak in an error path for addrinfo.
- Fix bug dereferencing a null pointer in the mount callback on error.
2012-11-25 07:57:40 -08:00
5f08d03e16
ZDR_ARRAYS: The sub-unmarshaller returns TRUE FALSE if there was an error, not TRUE
2012-11-24 09:01:00 -08:00
f3a7507806
Add an assert to track if we try to use an rpc_context after it has been destroyed
2012-11-24 08:18:54 -08:00
2811861b96
Dont leak addrinfo in error path
2012-11-24 08:14:12 -08:00
9b6d106414
MOUNT: when processing the mount response, check status before dereferencing data pointer
...
If status is non-zero in the mount callback, then the datapointer might be NULL
so we should not dereference it into an exports pointer before we have checked the status.
This would otherwise cause crashes.
2012-11-24 08:13:46 -08:00
456c0b147a
Fixed errno processing for connect() on Windows.
2012-11-24 08:12:45 -08:00
2521374f70
Modified mechanism to assign privileged ports and enable it for Windows as well
2012-11-24 08:12:31 -08:00
4a2b08768a
Add an assert to track if we try to use an rpc_context after it has been destroyed
2012-11-21 17:10:05 -08:00
be7f59339e
Dont leak addrinfo in error path
2012-11-17 10:14:19 -08:00
5ac848e908
MOUNT: when processing the mount response, check status before dereferencing data pointer
...
If status is non-zero in the mount callback, then the datapointer might be NULL
so we should not dereference it into an exports pointer before we have checked the status.
This would otherwise cause crashes.
2012-11-04 15:55:05 -08:00
f96b24fada
Fixed errno processing for connect() on Windows.
2012-09-18 08:33:41 +02:00
ac559609a3
Modified mechanism to assign privileged ports and enable it for Windows as well
2012-09-18 08:32:40 +02:00
b14062764e
ZDR: we need to include both arpa/inet.h and time.h
2012-07-05 21:17:45 +10:00
442b829a05
Dont just cast *enum_t to *uint32_t
...
go via a temporary variable in case these scalars are different size
2012-07-05 21:13:04 +10:00
67ba2239cb
Get rid of AUTH completely and replace all uses with 'struct AUTH'
2012-07-05 10:23:19 +10:00
45670bd0d6
fix compiler warnings
2012-07-05 07:37:37 +10:00
020f172895
fix compiler warning
2012-07-05 07:35:04 +10:00
d14e28387f
squelch compiler warnings in socket.c about integer sizes
2012-07-05 07:21:48 +10:00
574095b960
Change a size variable from uint64 to uint32 to squelch a warning
2012-07-05 07:12:11 +10:00
10a0ceaaa3
librpc_rpc_call_body() should return TRUE is successful
2012-07-05 07:09:00 +10:00
242b81cfbf
setpos returns a boolean, set it to always return TRUE
2012-07-05 07:06:33 +10:00
763cd6e3e2
ZDR: New builtin replacement for RPC/XDR called ZDR
...
This patch switches libnfs over to use precompiled rpcgen files
and using ZDR. ZDR is a trivial reimplementation of XDR that is built in
into libnfs.
This removes the dependencies of rpc/xdr completely and allow us to build on any
system, even systems where rpcgen and librpc/libxdr are not generally available.
2012-07-04 16:53:12 +10:00
6916a66546
NLM: initial support for NLM
2012-03-10 20:46:03 +11:00
fa3c25beee
NFSACL: Add support for SETACL3
2012-03-08 21:25:21 +11:00
47e5a2c740
new version 1.3
2012-03-03 19:04:58 +11:00
8e25581668
change rpc_nfs_symlink_async to take a full SYMLINK3args as parameter
2012-01-14 08:18:19 +11:00
16104b27bd
make low level readlink_async function take READLINK3args as argument
2012-01-14 08:11:09 +11:00
7edc9026db
make mkdir_async take a full MKDIR3args argument
2012-01-12 16:02:32 +11:00
c985c015f2
change crate_async low level function to take a full RATE3args structure as argument instead of just name and mode
2012-01-12 15:31:49 +11:00
8724c83390
Add function to get root filehandle from a nfs context
2012-01-11 07:04:01 +11:00
e6cac86f77
Try to rotate the ports we assign when root better so we dont reuse
...
a port htat is in time-wait too frequently
2012-01-03 07:25:25 +11:00
59f298f5bf
Improve an error message to show the errno value
2012-01-03 07:18:56 +11:00
5c6b1176c2
PMAP: allow to ask for tcp/udp port, dont assume users only ever want the tcp port
2011-12-31 14:24:20 +11:00
345422a2e3
New version 1.2
2011-12-06 18:02:34 +11:00
d7ec001f1a
Set IFREG from file type, dont assume mode bits contain it
...
Needed for HaneWin which does not set S_IFMT in the modebits returned
2011-12-05 13:54:21 +11:00
183451cff5
Remove all [s]size_t / off_t and replace with [u]int64_t making libnfs 64-bit pure
2011-11-27 14:49:08 +11:00
1ec6b50aef
Add MKNOD command support
...
Add support for raw nfs3 mknod and sync and async posixlike api
2011-11-27 14:08:24 +11:00
86ef491004
New version 1.1
2011-11-27 11:20:57 +11:00
14a062eb59
Everytime we use rpc_connect_async() internally, make sure we clear the ->connect_cb so that we dont have a callback function and "private data" being invoked long after the "private data" structure has been free()d.
2011-10-09 19:47:07 +11:00
a3d44462cb
Sync API: Once the sync nfs_mount() functions is finished we have to clear rpc->connect_cb.
...
Otherwise we get a crash once the connect_cb callback is invoked the second time.
2011-10-09 19:10:18 +11:00
b990de23c6
Only call the "connect callback" if the callback is set to non-NULL
2011-10-09 19:00:24 +11:00
8704724f83
Change remaining "struct AUTH *" to "AUTH *"
2011-10-09 15:22:05 +11:00
fa33813ed4
[fix] - fix improper definition of struct AUTH from osx (caution - API change - not clear if typedef is there on all plattforms...)
2011-10-08 18:10:06 +02:00
acf31bb24d
Bump version to 1.0.0
2011-10-04 18:30:15 +11:00
d678b73e4d
Fragment reassembly. Add reassembly of fragmented RPC PDUs
2011-10-04 12:31:56 +11:00
1744ef9011
Autoreconnect: autoreconnect was completely broken. Reimplement it so that it reconnects properly on session failure.
2011-10-04 11:25:10 +11:00
fb69c64c02
READDIRPLUS emulation: Some embedded servers do not support READDIRPLUS,
...
so try to trap this and emulate it with READDIR + a LOOKUP loop to get the attributes.
Since we are fully async we can do all the lookups in concurrently so it should not be that very exopensive compared to the cheaper READDIRPLUS call.
2011-10-03 10:55:42 +11:00
fcc42bfe85
[win32] - completed win32 port
2011-09-12 19:53:33 +02:00
4e9dcb7665
remove debug message
2011-09-12 22:20:56 +10:00
07fd0cbc65
when creating a socket, spin over a number of low-numbered ports
...
and try to bind to a system port in case the user is root or the
binary has the CAP_NET_BIND_SERVICE capability
this removes the need to use 'insecure' on the server
2011-09-12 22:14:26 +10:00
18c27b73ff
Large-Writes: just like for large-reads, chop large writes up into the smallest
...
of 32kb and what the server responded as the max write size and send them out in parallell to the server.
32kb is a common limitation in XDR libraries so even if the server says IT can handle writes larget than that, our XDR library might not.
2011-09-11 22:13:28 +10:00
83aa785d92
Add a call to find the queue-length so we can see how many I/O we have in flight from the application layer.
2011-09-11 21:32:48 +10:00
4b0e8df9a9
[fix] - fix connect for win32 - return values are inverted
2011-09-03 15:31:56 +10:00
99c14c9b81
[add] - implement set_non_blocking for socket
2011-09-03 15:30:01 +10:00
a8a1b85846
[win32] - make it compile on win32
2011-09-03 15:08:48 +10:00
4b1097cde2
remove old libnfs.def and update the buildscript to the new name libnfs-win32.def
2011-09-03 14:53:10 +10:00
7f7993883c
[add] - win32 compat stuff
2011-09-03 14:51:54 +10:00
db4ea5dd80
Revert "Rename libnfs.def to libnfs-win32.def"
...
This reverts commit a3d6cf111e
2011-09-03 14:51:48 +10:00
ea98629aef
memset() is posix, bzero() is not change bzero into memset
2011-09-03 14:11:22 +10:00
a3d6cf111e
Rename libnfs.def to libnfs-win32.def
2011-09-03 13:29:39 +10:00
565369e0b2
remove some includes that dont exist on windows
2011-09-03 11:30:10 +10:00
bb4e9ed6a8
cast to char * to avoid compiler warnings for get/setsockopt
2011-09-03 11:25:50 +10:00
4ed97718b3
Merge branch 'master' into win32-3
2011-09-03 11:11:36 +10:00
42f43e48cc
cast the pointer to char * for all calls to bzero
2011-09-03 11:08:58 +10:00
41d82d7d82
Merge branch 'win32' into win32-3
2011-09-03 10:57:42 +10:00
0804e67d7a
using READDIRPLUS3 we get some extra fields for the direentry structure for free, such as type, mode, size, timestamps
2011-09-02 20:45:30 +10:00
f390f18157
nfs_opendir() switch to using READDIRPLUS3 instead of READDIR3 when reading directories.
2011-09-02 20:27:01 +10:00
647d2ea11c
add support for opensolaris
2011-09-01 19:48:42 +10:00
427d71c0a7
Win32: add a libnfs.def file so we can build libraries/dlls
2011-08-30 15:54:29 +10:00
b85c7de2f7
use vsnprintf() on all platforms, dont special case win32 for vsnprintf and all other platforms use vasprintf
2011-08-29 19:59:36 +10:00
2606f9bb3d
Win32: add workaround for lack of vasprintf() and use vsnprintf() on windows on a fixed size buffer
2011-08-29 18:00:09 +10:00
0ad9a1f134
WIN32: win32 does not have poll()
...
add trivial replacement function
2011-08-28 19:48:01 +10:00
f4442bec27
wondows dont have vasprintf()
...
disable for now and reimplement later
2011-08-28 19:45:00 +10:00
6874f61e24
Win32: more win32 adaptations
2011-08-28 19:43:33 +10:00
fb6510bb68
VC10 dont like ';;'
2011-08-28 19:30:10 +10:00
d8143ff7d7
Win32: the stat structure on win32 does not have blocks or blocksize
2011-08-28 19:29:08 +10:00
93e9306f80
replace index/rindex with modern equivalents
2011-08-28 19:27:57 +10:00
a671a46510
replace index/rindex with modern equivalents
2011-08-28 19:27:23 +10:00
9896c7155c
nfs_set_auth() is a void and does not return anything
2011-08-28 19:26:07 +10:00
be337389bf
nfs_set_auth() is a void and does not return anything
2011-08-28 19:25:20 +10:00
eecdc4f3cd
Win32 changes, include files we need when compiling under win32
...
and some socket ops are slightly different
2011-08-28 19:24:18 +10:00
11ef53a2e6
callit_cb() : dont return a value from avoid function
2011-08-28 19:00:04 +10:00
b3740e09de
callit_cb() : dont return a value from avoid function
2011-08-28 18:59:13 +10:00
963c2f83be
Initialize the rootfh handle when creating the nfs context or else
...
we will try to free a wild pointer if the mount failed
and we proceed to call nfs_destroy_context();
2011-07-31 10:59:46 +10:00
b077fdeb65
Add automatic reconnect after TCP session failure for the sync interface.
...
Once a NFS TCP connection has been idle for a very long time, say
10-15 minutes, it is common that NFS servers will tear down the TCP connection.
So add code to re-connect to the NFS server and re-issue any i/o that might have been in flight (unlikely unless the server has hung)
While it is sufficient to just reconnect to tcp port 2049, it was much simplet to just utilize the nfs_mount() function.
This means that instead of just reconnecting TCP port 2049, we do a full blown
nfs mount via portmapper and mountd protocols.
2011-07-31 10:49:47 +10:00
8c27363e21
since UDP is unreliable, send the CALLIT broadcasts 3 times, one second apart when probing for local NFS servers
2011-07-16 17:30:44 +10:00
e5964ef905
break out the loop for sending a NFSd probe bcast on every interface into a separate function
2011-07-16 17:27:12 +10:00
41a6209bef
[fix] - fix allocation of if list for osx
...
- removed one unneeded header in examples/nfsclient-bcast.c which was breaking build under osx
2011-07-11 19:16:13 +02:00
9a96dd46d2
We have to include config.h when testing for precense of sockaddr sa_len
2011-07-10 08:50:36 +10:00
1be803ce7f
On OSX and other systems where we have sockaddr_in, SIOCGIFCONF returns a packed list of ifreq structures, so we can not just index them as (struct ifreq *)[i]
...
Instead we have to compute the offset into the buffer for where the next record starts (this offset might or might not be sizeof(struct ifreq)
2011-07-10 08:39:54 +10:00
56320c681f
Fix out-of-tree (VPATH) builds
2011-07-06 16:38:38 -04:00
1ad6f93185
when ioctl() for getting the broadcast address fails, continue the loop to the next entry instead of exiting out
2011-07-04 08:46:19 +10:00
b989ca9c36
- fixed missing include for osx
2011-07-03 15:25:48 +02:00
f7f931c7e3
add back a check for sa_len in sockaddr structure.
2011-07-03 07:41:15 +10:00
1b9917b886
With broadcast PDUs we can receive multiple replies so must make sure we
...
delete the previuous unmarshalling buffer when we start over and
process the second pdu to the initial call.
Or else we will leak memory very slowly when processing broadcast RPC replies
2011-07-02 12:16:37 +10:00
2e4db0d145
Merge remote branch 't-nelson/automake'
2011-07-02 11:10:36 +10:00
552c766578
add example for sync interface to probe for all local servers
2011-07-02 10:30:16 +10:00
fc01d2a961
Fix compile error on solaris.
...
Include sys/filio.h explicitly as solaris doesn't define BSD_COMP and
thus doesn't pull this header via sys/ioctl.h. Needed for FIONREAD.
2011-06-29 14:15:23 -04:00
2842ff15c8
Handle versioning the libtool way.
2011-06-29 14:15:22 -04:00
6c437065d3
Rewrite build system using automake
...
Initial commit, only linux tested so far.
2011-06-29 14:15:16 -04:00
7ff2f3a01d
when destroying a udp rpc context, free any udp_dest data we may have
2011-06-26 18:58:54 +10:00
c481da67f2
add function to extract the sockaddr from which we received the current rpc pdu
...
(useful when parsing replies from broadcast rpc calls)
2011-06-26 18:45:38 +10:00
5bf60dc601
add function to set the target for UDP rpc calls.
...
including doing broadcast calls
2011-06-26 18:42:53 +10:00
485bc9b9a7
add function to bind a udp rpc context to a local port/address
2011-06-26 18:40:00 +10:00
070287e516
Dont allow a tcp _connect to a UDP context
2011-06-26 18:32:06 +10:00
0268794fe3
add capability to read UDP datagrams from the socket and pass them to the parser
2011-06-26 18:30:26 +10:00
5911f3e8ab
for UDP sockets, we ONLY wait for pollin events
...
since upd sockets are always writeable.
they just silently drop packets when congested
2011-06-26 18:26:34 +10:00
a669857d0d
add support for sending and receiving udp pdu
2011-06-26 18:24:24 +10:00
a19924120d
in allocate pdu, if context is udp, dont allocate space for a record marker
2011-06-26 17:30:00 +10:00
9e00b8c663
add function to request a UDP context
...
(private use only since we dont want applications to use UDP)
2011-06-26 17:27:40 +10:00
7ed9d87a5b
redo the previous patch to instead initialize max_offset correctly before we invoke the chunks.
...
this is the correct way to fix this bug.
2011-06-25 12:23:52 +10:00
215228a23d
handle the case where we try a multi-read that is completely beyond the end of file.
...
when a multi-read was completely beyond the end of file, this caused us to invoke the callback with a 'read-count' of <0 which the callback would treat as a failure.
This would then cause the callback to treat the data pointer as an error string and try to use it for nfs_set_error().
Since the data pointer was actually a real binary databuffer and not an error string this would cause the NFS error string to look like it contained garbage data.
In this case, where the multi-read fails to read any data at all since it if fully beyond end of file, make sure to invoke the callback with a read-count of 0
2011-06-25 12:13:11 +10:00
739df145d6
fix crash in mount/export
...
need to check that status is OK before we start dereferencing the 'export' pointers or else we will segv.
this could happen if trying to pull the list of exports from a host that does not have mountd running, or if we can not connect at all to the host.
2011-06-25 11:06:53 +10:00
1e8994af0f
Add a very subtle bug in nfs_set_error()
...
when nfs_set_error is called with error_string being the result of a lower layer
problem so it is passed as "rpc_get_error()"
This meant that since free that nfs->rpc->error_string before we reference it a few lines further down in the vasprintf(..., error_string,
adn memory corruption triggers.
Valgrind found this, Valgrind rules!
2011-06-21 18:33:50 +10:00
e4a5ba42b8
When doping chunked multi-reads of a blob bigger than the max readsize,
...
a read can sometimes span beyond the end of file.
In which case individual smal read chunks beyond the end of file will return OK, 0 bytes of data and EOF==1
For this case, trap when len==0 and ignore these chunks.
Do not update the read-count if / when len is 0.
2011-06-21 06:07:29 +10:00
e8cab72a79
From Paul van der Zwan
...
Set the directory bit in st_mode properly.
2011-06-20 01:02:48 +10:00
cdb19ec164
Redo the buffer handling for input buffers and make sure we only read one PDU at a time
2011-06-20 00:56:47 +10:00
e210bd2af8
remove the need for an explicit rpc context for the highlevel function mount_getexports()
2011-06-19 23:20:35 +10:00
df5af25fd2
add a sync function to get the export list from a server and add an example of how to use this functio n to nfsclient-sync
2011-06-19 22:31:34 +10:00
7f0242ca9a
Add a high-level async function to read the export list
2011-06-19 14:54:17 +10:00
1896d37bfa
get rid of all remaining printfs from the library
...
add a helper function nfs_set_error() to use instead of the printfs
2011-06-19 09:38:20 +10:00
5f9d02f839
get rid of compiler warning
2011-06-19 09:05:20 +10:00
cbbf9d3e81
remove all printf's and replace them with rpc_set_error()
...
also change all error return codes to always return -1
2011-06-19 09:03:04 +10:00
921f877be4
Add support in read/pread to be aware of the Maximum Read size that the server
...
suggested.
When a read/pread operation that is 'oversize' is issued, the read/pread command will internally chop this up into smaller chunsk and send then concurrently to the server.
As replies come back from the server we add the data to a reassembly buffer
and as soon as all replies have been received we invoke the callback and pass the reassembly buffer back to the application.
2011-06-19 08:43:28 +10:00
e2ba5764d4
Remove the _sync postfix for the synchronous functions
2011-06-18 15:07:53 +10:00
98f5fee87d
- linux, osx, and ios need rpc/rpc.h includede before rpc/xdr.h
...
- fixed some missing includes for bzero, malloc and free
2011-06-01 18:06:15 +02:00
ea214e4566
Use HAVE_SOCK_SIN_LEN and set ->sin_len
...
for legacy unixens (AIX)
so that we will compile and run on those systems with a "weird"
sockaddr_in definition.
2011-06-01 21:31:40 +10:00
17ef62fad7
Read the max read/write sizes supported when connecting to NFS
...
and store in the nfs context.
Privide accessor functions to read the max sizes.
2011-06-01 00:08:29 +10:00
912f7ad5b0
Use getsockopt to extract better failure condition when tcp layer receives a failure
2011-02-07 22:11:12 +11:00
84004dbf9b
initial libnfs checkin
2011-02-06 15:45:09 +11:00
Peter Lieven