Novell Modular Authentication Services support;
fix some return values and decodes;
show the NDS reply buffer only in service request packets.
svn path=/trunk/; revision=10263
1. Add several return values
2. Fix the reply structure for NCP 22/48, 23/122
3. Fix the request structure for NCP 23/121
4. ServerID should be displayed as a hex value.
5. ServerStationLong should be LE.
6. JobNumberList should be JobNumberLong.
7. Fix 87/64, 87/65, 87/66, 87/68, 87/69 - These are not Advanced Audit
Service, they really 64 bit file support NCP's.
svn path=/trunk/; revision=8855
1. Added Secret Store Services group
2. Added a number of NCP return values
3. Added support for NDS defragmentation
4. Fixed type from file to bindery on ncp's 23/20, 23/21, 23/22, 23/24,
23/27, 23/28, 23/29, 23/30, and 23/31
5. Fixed type from comm to nmas on ncp 91
6. Fixed ncp 92 to include 92/01, 92/02, 92/03, 92/04, 92/05, 92/06,
92/07, 92/08, 92/09, and 92/10
7. Added return value for 104/02
8. Fixed reply structures for NCP 123/60 and NCP 123/79
svn path=/trunk/; revision=8256
Fix a spelling error.
Don't use lengths with ASCIIZ strings, as "proto_tree_add_item()" will,
with a specified length, attempt to fetch the specified number of bytes
for an FT_STRINGZ string, throwing an exception of there aren't that
many bytes present.
svn path=/trunk/; revision=7786
NCP 123,62 did not display value for the set command. Set default
value to guint32.
Add some stuff for the Novell Modular Authentication Service.
svn path=/trunk/; revision=7559
1. Added a number of NCP Error return values.
2. Added ability to pass byte strings to summary window.
3. Modified file NCP's to reflect file handle in summary window.
4. Changed all NCP's using QueueID from big endian to little endian.
This is different from Novell's NCP documentation but matches same
information obtained from NDS dissector.
5. Added over 50 more NDS error return values.
6. Partially fixed NDS Request verb 6 (Search Entries)
svn path=/trunk/; revision=7172
to be correct; remove the comment about what was there not matching.
Note that the PropertyValue item in a Write Property Value request
should perhaps be omitted if MoreFlag isn't set (it appears to be
garbage if it's not set).
svn path=/trunk/; revision=7123
Properly display the reply to a Get Name Space Information request -
there is a sequence of name space names, and a sequence of pairs of
{associated name space, data stream name}, and there are also sequences
of name space index numbers, but if we try to display anything after the
list of loaded name spaces, nothing gets displayed at all.
svn path=/trunk/; revision=7101
and in file attributes means "this is a subdirectory"; don't mix the
two.
The 1-byte and 2-byte search attributes appear to have the same bit
definitions (except, obviously, for those that are in the topmost byte).
svn path=/trunk/; revision=7083
the ones not specified as such when used), so declare them as such
rather than specifying them as such when used.
The SearchSequenceWord also appears to be big-endian.
Note that we're not cracking the bits of a DirectoryAttributes field.
The "Subdirectory" bit in search attributes is really "Subdirectories
Only", as in "just show me subdirectories".
Note some confusion about whether the bit numbers on the Novell Web site
for search attributes are bit numbers or bit flags.
Note that we appear to have gotten back attributes for a file rather
than a directory in at least one search that had "Subdirectories Only"
set - unless the problem is that the numbers in the Novell spec are bit
numbers rather than bit flags.
Update some items that claimed to have a 2-byte search attributes field
to have a 1-byte search attributes field instead, to match the spec on
the Novell site (of course, the spec could be wrong...).
svn path=/trunk/; revision=7081
structures; make it so (which makes it big-endian on other structures as
well).
The same appears to be true of the UpdateID.
There doesn't appear to be a reserved word between the
SearchSequenceWord and DirectoryID fields of a DirectoryInstance or a
FileInstance.
The dates and times in a DirectoryInstance or a FileInstance appear to
be big-endian; note that we need to add the ability to make those
instances big-endian but leave others little-endian.
Note that there's a sequence of NameSpaceNames and a sequence of
DataStreamNames in the reply to a Get Name Space Operation request.
A Login Object request has a 2-byte object type followed by a
ClientName, just a UserName. The same applies to a Get Object
Connection List request.
Note that the DirectoryInstance or FileInstance in a File Search
Continue reply appears to be garbage if the reply doesn't succeed with
"OK".
svn path=/trunk/; revision=7077
qualifiers as necessary to ensure that we don't have to.
"strcmp()", "strcasecmp()", and "memcmp()" don't return booleans; don't
test their results as if they did.
Use "guint8", not "guchar", for a pointer to (one or more) 8-bit bytes.
Update Michael Tuexen's e-mail address.
svn path=/trunk/; revision=6726
Various NDS dissector enhancements including NDS verb 0x07, and
proper ping flags for NDS ping version 10 request.
Updated error codes for several NCP reply packets.
NDPS enhancements.
svn path=/trunk/; revision=6463
string, it's a tree name (as the field's display name suggests).
Get rid of the unused "ncp.nds_string_value" field.
Fix up the names of some fields, and use "proto_tree_add_uint()" rather
than "proto_tree_add_uint_format()" to add them to the protocol tree.
svn path=/trunk/; revision=6408
This update adds NCP 87/66.
Also add additional error codes, added hf_definition for super class
(NDS). Cleaned up some unused variables (Warnings when building
Ethereal). Fixed NCP type 0x9999 (server busy packets to echo
information to summary information), Added subtree format to NDS
dissector to cleanup display.
svn path=/trunk/; revision=6331
without having to know the numerical values for the verbs.
Use that table to convert the verb value to a verb name.
Fix indentation.
svn path=/trunk/; revision=6075
1. Secret Store Services (NCP 94) (ncp2222.py)
2. NMAS (NCP 92) (ncp2222.py)
3. NDS information in summary screen (packet-ncp.c & packet-ncp2222.inc)
4. Sever broadcast packets (NCP type 0xbbbb) to notify workstation to clear op-lock (packet-ncp.c)
5. Large Internet Packets (LIP) (packet-ncp.c)
6. Unicode Support. (unicode_to_string function in packet-ncp2222.inc & ncp2222.py)
svn path=/trunk/; revision=6069
1. Fixes for NCP decodes (87/6), (30), (87/32)
2. Endianness fixes for a number of decodes
3. Added error codes for return values not defined
4. Added most common SAP types under ObjectType table
svn path=/trunk/; revision=5758
the source to an "ipxdump" utility from the Linux ncpfs package.
The NetworkAddress field in a 23/26 GetInternetAddress reply appears
to be big-endian; we assume it's big-endian in all messages.
The NetworkSocket field in that reply also appears to be big-endian;
it was already set up to be big-endian in other messages.
Put in comments noting other things seen in captures.
svn path=/trunk/; revision=5658
they're the right width.
Fix the definitions of AccessRightsMask and JobControlFlagsWord,
and add AccessRightsMaskWord.
svn path=/trunk/; revision=5546
them in "packet-ncp2222.inc".
The page at
http://www.odyssea.com/whats_new/tcpipnet/tcpipnet.html
indicates that a positive ACK (0x9999) NCP packet has the same
completion code and connection status fields as a reply (0x3333) packet
(but nothing after them); hand "dissect_ncp_reply()" the packet type as
one of its arguments, and have it handle positive ACK packets as well as
reply packets.
It also indicates that bit 4 of the connection status indicates that the
server is unavailable, and the page at
http://www.unm.edu/~network/presentations/course/appendix/appendix_f/tsld088.htm
speaks of that and of the significance of other bits; put a comment in
"ncp2222.py", before the "hf_ncp_connection_status" field, about that.
From looking at a capture, it appears that a "destroy service
connection" (0x5555) packet should be treated like a "create service
connection" (0x1111) packet and be handed to "dissect_ncp_request()".
Note that perhaps watchdog packets should be handled by
"dissect_ncp_reply()" as well.
svn path=/trunk/; revision=5489
A little work still needs to be done on the new NCP dissector -- make
some of the COL_INFO texts more useful, handle a Unicode issue, and
modify some of the cases that use "request conditions".
But the NCP dissector as it stands is very usable now.
Note: I didn't merge in the PROTO_LENGTH_UNTIL_END macro... I wanted
to think about the various possible macros and review an email conversation
I had with Guy on the subject.
svn path=/trunk/; revision=5432
Guy Harris