| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
It is accessable e.g. through:
git pull git://repo.or.cz/netsniff-ng-old.git with-mausezahn
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Maybe in short term future, before the next release, we will think
about putting the README file for users into DOCDIRE, since the
rest is not really interesting for users, but probably only for
developers.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
We should only put the things here that are interesting for users,
thus, add steps to verify a release from the Workflow document.
[1] http://pub.netsniff-ng.org/docs/Workflow
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Lets move CodingStyle, SubmittingPatches, and Sponsors into the
root directory of netsniff-ng and remove the Documentation folder.
Some of those files are quite bloated, and most of these things
should be in the man-pages anyway. They should be the only big
sources of documentation, nothing else.
The rest is currently put here: http://pub.netsniff-ng.org/docs/
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Add the known bug from Documentation/KnownIssues into the man page
so that at some point in time, we can remove this document and
maintain that stuff within the man page.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
When we're still root, we tell the kernel to lock/protect all current
and future pages in memory so that they will not be swapped out in
case the system uses up too much. Now when we do xzmalloc_aligned(),
it calls internally posix_memalign() that can call mmap(2), thus we
will get an EAGAIN as errno, since we're not root anymore and since
we wanted to touch sth. that belongs to root. Nasty. Fix this up by
only protecting these pages when we do not use -u/-g.
Reported-by: Doug Burks <doug.burks@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Jon Schipp reported that a arbitrary packet config
such as ...
{
"MyPacket",
"MyPacket",
"MyPacket",
"MyPacket",
"MyPacket",
"MyPacket",
"MyPacket",
"MyPacket",
}
... doesn't work anymore, although it used to work. This
regression was introduced by commit f9591eebe40c ("trafgen:
lexer/parser: fix cpu() selection and whitespacing").
Reported-by: Jon Schipp <jonschipp@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Fix wording and phrasing in some of the usage examples.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
|
|
Replace 'priviledges' by 'privileges'.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
|
|
Replace 'networkoing' by 'networking'.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
|
|
When using netsniff-ng with dropping priviledges, we have to introduce
another pcap helper function that is called once before we drop the
priviledges. In this function we have to invoke the disc I/O scheduler
policy, because it needs priviledges. Otherwise netsniff-ng will fail
with "Failed to set io prio for pid" on startup, since we're not root
anymore.
Reported-by: Doug Burks <doug.burks@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
E.g. when executing trafgen on tap devices, trafgen will complain
that it is not up and running although it was brought up. Remove
this usability restriction and let the kernel decide if it may
send the packets or not.
Reported-by: Aimal Khan <aimal.khan98@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Add a section of known bugs/todos for bpfc.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Fill in the rest of the gaps for this man page.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
This patch continues on the netsniff-ng man page. It adds a program
description and documents some options, not yet all. More to come.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
The usage examples are good so far. Most of the things one needs
to know. Also, add a short howto on how to use bpfc filtering or
pcap-like high-level filters.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Not yet finished, but it already contains some useful example
descriptions and notes for tuning. More to come very soon.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
This patch adds BPF architecture and instruction description with
supported Linux extensions. Also, essential bpfc examples are given.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
These are the writings so far. For sure they are not yet finished, so
I will push updates for both of them as soon as possible, but for now
this is a first version people can start to read.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
There is actually no reason why netsniff-ng should fail if the NIC is
down at startup. We still can setup everything and already capture at
the time it goes up. This might be useful when replugging cables on
servers, for instance.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Finally I found some time and started writing man pages. This is a
first one in a series of man pages for all tools.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Still untested, since we did not yet tag a release, but for now,
also credit all involved people in a particular release.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
When we really clean up everything, also remove crap files that do
not belong to the Git tree.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
If we unmap TX ring buffers and still have timer shots that trigger
the kernel to traverse the TX_RING, it can send out random crap in
some situations. Prevent this by destroying the timer and flush the
TX_RING first in wait mode.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Do not enforce whitespace at the beginning or end of a packet
configuration, there's no reason to do so.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Also return if no needle has been found in the shell code haystack.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
After commit f9591eebe40c (trafgen: lexer/parser: fix cpu() selection
and whitespacing), we also need to fixup the help description, so that
users start using the cpu(X-Y) notation instead. Also, reintroduce the
alias csum16 for csumip.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
The previous commit reduced it, so now we allow it within cX() containers
in order to negate a result, if someone needs that.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
It looks unintuitive when once needs to write cpu(0:3), so also allow
cpu(0-3). Then it's quite stupid to enforce whitespace after the double
colon before the curley braces open. So make this rather optional here.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
When having a config like cpu(1): {...}, and one runs it with -n1, trafgen
behaves buggy, I also noted that in other situations. Fix this by letting
the loop also return on CPU state RES, and do not perform this stupid magic
in the main routine. So far it seems to work now after some basic tests I
did.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
In both, the RX_RING and TX_RING we need to unmap first and then destroy
the buffer, otherwise, we get a device or resource busy.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
In preparation for [1], add the #poff operation, so that users can
dynamically truncate packets on the payload start offset. This is
useful for several reasons: privacy, speed, ...
Example for bpfc:
ld #poff
ret a
[1] http://thread.gmane.org/gmane.linux.network/262512
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Only attempt to close the file descriptor in error cases where it is
actually open.
This was discovered by the coverity scanner.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
|
|
The return value is never used by the callers, so we don't need to
return anything.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
|
|
Add a uniform header comment to the two scripts as present in
other scripts in this repository as well. Also rename build_nacl.sh
into nacl_build.sh to stay conform in naming. For this, also update
all mentions of build_nacl.sh to nacl_build.sh.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Add a missing header file description as present in other scripts,
for example. Also credit Stefan for his help.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Both, the IP and ICMP checksums use the same algorithm from RFC1071.
Thus, it's enough if we give csumip an alias to csumicmp, so that it
can be used in trafgen packet configurations. It's calculated from
the ICMP header + data, with value 0 for this field.
Suggested-by: Jon Schipp <jonschipp@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Add different bpf opcode pretty printer to support a variety of
output formats for netsniff-ng, xt_bpf, tcpdump, C-like and make
future integration of formats easier.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Since we list all authors for their contributions in the AUTHORS,
add Peter Stuge for his improvement on the build system.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
When setting a prefix from the build system, also pass this through the
hard coded strings for config file locations in the toolkit, so that the
installed files (which use PREFIX) can be properly found.
Suggested-by: Peter Stuge <peter@stuge.se>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Hw timestaming is currently done, if we have the header file from
the kernel available. If it fails, we currently bail out, which is
bullshit. Just returning is enough, since some users might have the
header available, but not a NIC supporting hw timestaming.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
We do not assign or use this particular state, so just remove it from
the enum. We continue anyway in the while loop in that case.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Make whitespacing more uniform to coding conventions and also simplify
some fubar if + xfree combinations by using free only.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Remove various unreachable sections from the switch statements. This
was detected by the coverty scanner.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
If the new pointer is NULL anyway, this results in a constant expression
during compile time (NULL != NULL) where a barrier is left out. Thus, it
will have no different effect using rcu_assign_pointer() than assigning
it directly.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Check various setsockopt return values and exit accordingly if some
of them just fuck up.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
If something screws up, which is rather unlikely, but if it happens,
let the user know.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Before we do calculations on statistics, check if we really got them.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Let us check the return value when the timer triggers a TX flush
request to the kernel. However, ignore the case of BADFS and NOBUFS.
The socket could already have been closed before the timer triggers
in the first case, and in the second, we just let the next timer
continue processing if currently the buffer space is exhausted.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
|
Coverty detected a misuse of the getnameinfo API, thus, we
need to fix the length parameter.
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
|
