Add field 'nfs.ops.count' in the detail pane of NFSv4 calls and replies that
displays the number of operations in NFSv4 COMPOUND requests/replies.
From me: change the blurb wording a bit.
svn path=/trunk/; revision=33069
Display the fsid (filesystem ID) in decimal as well as hex in the "attributes"
section of the header in NFSv3/v4 replies.
svn path=/trunk/; revision=33068
Mesa moved its API description file again.
Also, while I'm here, I noticed that they stopped using signed tags, so add
--tags to "git describe" for a better description of the mesa version used to
build the X11 dissector.
svn path=/trunk/; revision=33067
"v10 of the packet format was added recently and could be added
to the wireshark dissector"
See: Bug #4833: https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4833
From me:
- Fix a compile error: ...version=tvb_get_ntohs...
- Fix a missing = NULL.
svn path=/trunk/; revision=33066
Note that the number of UDVM cycles MUST NOT be increased if a
request for additional compressed data fails.
so we *shouldn't* increase the cycle count further when
SIGCOMP_INSTR_INPUT_BYTES fails. That does *NOT* of course, mean that
we shouldn't increase the cycle count by 1 for a failed INPUT-BYTES
instruction - that would leave UDVM vulnerable to infinite loops (as per
bug 4826), and I *really* doubt that was their intent; presumably, it
means it should not be increased *by the number of cycles for the
additional data*:
Additionally, if the UDVM successfully requests n bits of
compressed data using one of the INPUT instructions then the
number of available UDVM cycles is increased by n *
cycles_per_bit once the instruction has been executed.
if the attempt to get that additional data fails.
svn path=/trunk/; revision=33065
From reading the rawshark(1) manpage my assumption was that rawshark
could be used like
$ /usr/bml/bin/rawshark -s -r test.pcap -d encap:EN10MB ...
However rawshark either expects the -r argument to be -
(read from stdin) or a pipe which results in the following error
message:
rawshark: ".../test.pcap" is neither an interface nor a pipe
The proposed rawshark.pod patch updates the -r description to
the implemented rawshark functionality.
The patch also applies to the current SVN version.
svn path=/trunk/; revision=33063
"Different people made changes to enhance the batman-adv dissector. It seems
that the batman dissector wasn't touched and misses those changes. Following
patchset should improve the dissector the same way Gerald Combs, Guy Harris and
Bill Meier improved batman-adv."
See Bug #4384: https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4834
From me: Add back a few 'if (check_col()) ...';
I probably went just a bit too far in changes I made in packet-batadv.c
svn path=/trunk/; revision=33062
The wireless meshing protocol B.A.T.M.A.N. Advanced changed their packet format
in such a way that now versions can be identified and so correct dissection of
the packets can be supported by wireshark.
Since it is a ever moving target it is very possible that the packet format is
changing slightly. The dissector was written in such a way that new version can
be supported relative easy.
I hope that it sufficient for the inclusion in wireshark.
I tried to fuzzing it some hours and no error was reported.
From me:
Initialize our dissector handles.
Merge packet-batadv.h into packet-batadv.c. It isn't included anywhere else.
Fuzz 500 passes using attached capture files.
svn path=/trunk/; revision=33052
Introduced some state to remember last dissected Tag/Length so that they can be recalled if an IMPLICIT tag is encountered and stripped. This allows its to be determined if the value has a constructed value - and so can be reassembled.
In this case, it is a IMPLICIT constructed OCTET STRING at the presentation layer.
Many thanks to Fred Gruman for identifying - and apologies for the delay in commiting.
svn path=/trunk/; revision=33048
This patch adds a new '-S' option to editcap that will rewrite timestamps of
packets to insure that the new capture file is in strict chronological order.
This option's primary use case is to fixup the occasional timestamps that have
a negative delta time relative to previous packet.
This feature is related to (but does not depend on) capinfos enhancement
submitted in bug #4315 which helps identify tracefiles with "out-of-order"
packets.
svn path=/trunk/; revision=33042
This patch adds a new '-o' option to capinfos (enabled by default) to report if
the packets within a particular capture file are in strict chronological time
order or not.
svn path=/trunk/; revision=33041
I've created a ASN.1 dissector for the IEC 61850 Sampled Values protocol. It
dissects ethernet frames of the IEC 61850-9-2LE specification form the UCA
International User Group.
There is also a new TAP for tshark (-R sv) which extracts the important
information of the frame and allows to create plots (with external tools) of
the sampled values.
I've developed under Linux (Ubuntu 8.10) but everything should be in place for
successful compilation under Windows.
It would be great if this dissector could be included in wireshark. I'm looking
forward for your comments.
svn path=/trunk/; revision=33039
Sébastien's initial commit [1] didn't contain support for embedding Python on
Windows.
[1] http://anonsvn.wireshark.org/viewvc?view=rev&revision=28529
From me:
Comment out PYTHON_EMBED for now.
Start a list of known Python+CRT versions.
Add get_wspython_dir to libwireshark.def.
svn path=/trunk/; revision=33036
Lori made a simple change to the bacapp.c file for segmentation parsing.
Basically all fragments have some common data in the APDU which is useful to
see parsed in the tree.
I added support for vendor numbers bigger than 255.
Corrected Info column for "message fragment" and "message reassembled"
svn path=/trunk/; revision=33033
Decode the EIA0 (null integrity algorithm) bit in EMM messages. This was introduced in 3GPP 24.301 CR-0422 (Release 9.2) and is also used by the LTSI (LTE/SAE Trial Initiative) for interoperability tests between constructors (with Release 8 equipments). I also updated the description of EEA0 as stated in 24.302 release 9.2.
svn path=/trunk/; revision=33030
back out the change to check its return value until we fix that.
Also back out the test suite changes to look for an error exit for
invalid capture filters and interfaces.
svn path=/trunk/; revision=33029
we shouldn't rely on getting it pulled in by other shared libraries, as
1) there's no guarantee that the other shared libraries we use
are linked with it or will continue to be linked with it;
2) there's no guarantee that we're even linking dynamically;
3) it fails on Fedora 13 as part of a change to catch programs
that implicitly (or explicitly) assume, in their build
procedures, that they'll get libraries linked in as a result
of linking with other libraries.
svn path=/trunk/; revision=33028
Fix for bug 4810 so PPP/USB dissector registers for IF_CLASS_CDC_DATA
- in addition to IF_CLASS_UNKNOWN and IF_CLASS_VENDOR_SPECIFIC, for those rare
occasions that the right set of circumstances arise where PPP traffic will be
(correctly) encapsulated in a USB frame of that interface class.
svn path=/trunk/; revision=33026
There are two typos in the RTCP TMMBR packets:
1. In the dissector of RTCP TMMBR is written "messured" instead of measured.
2. In the text title field is written TMMBN instead on TMMBR and vice-versa.
svn path=/trunk/; revision=33022
Fix for bug 4807 to handle compressed (NULL) address and control bytes.
The problem seems to be that the address and control bytes are missing,
presumably because they are compressed per section 3.2 of
http://tools.ietf.org/html/rfc1662. The attached patch tries to account for
that.
The patch also fixes another bug. If the hdlc framing character, 0x7e, is not
present, but the address and control fields are, then the offset might need to
be 3 bytes instead of only 2 bytes if the control field is escaped. Sometimes
it is escaped and sometimes it isn't, so we need to handle both cases.
svn path=/trunk/; revision=33021
