| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|/ /
| |
| |
| |
| | |
There is a possiblity of some hostile netlink user getting
bogus responses in the conversation between zebra and kernel.
|
| |
| |
| |
| |
| | |
When interface is renamed notify daemons of addresses that are
now available.
|
| |
| |
| |
| | |
More diagnostic messages in case of updating existing route.
|
| |
| |
| |
| | |
More messages in case of netlink errors, and debugging
|
| |
| |
| |
| | |
If route type or other index changes flag it as changed.
|
| |
| |
| |
| |
| |
| | |
In the case of rename, delete_retain would be called twice. Once
when the interface is removed from the system, and again when
interface is being replaced during rename.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
If a routing daemon crashes, the quagga code sets up a signal handler
and catches the exception and produces a backtrace. If the code is
not stripped then the backtrace can include symbolic information
that can be useful for debugging.
Save space by not including full debug symbols.
|
| | |
|
| |
| |
| |
| | |
these files are created by build
|
| |
| |
| |
| |
| |
| | |
When interface is created, associate it with any pre-existing
configuration. When interface is renamed, notify protocols that
is deleted/added but keep existing address information.
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This reverts commit 3b1779981cd67679dd4cddd70160859ca3d862d7.
Conflicts:
zebra/rt_netlink.c
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This resolves issue with static routes of multilink.
In the multilink case, route is created before device exists.
Later device is created as ppp0 and gets renamed to ml0.
Zebra needed to orphan the old interface structure and create
a new one to correctly match up the interface with the inactive
route.
|
| |\ \ |
|
| | |\ \ |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Need to delete then add since zebra has no clean way to pass around
rename, and the configuration is based on name not ifindex.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Need to handle the case where interface comes back after being
deleted. After deletion state is retained but ifindex is set to
IFINDEX_INTERNAL. When new interface arrives with same name, then
update index
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The name has to continue to exist after delete in order to prevent
crashes of daemon (BZ 3990) and restart of multilink interfaces (BZ 3987).
|
| | |\| | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Speed up shutdown by using the existing pid file, and having
start-stop-daemon do the retry.
|
| | | | |
| | | | |
| | | | |
| | | | | |
This optimization is by default off with -Os
|
| | |/ /
| |/| |
| | | |
| | | |
| | | | |
Kernel can sometimes send delete address messages after
device has been deleted.
|
| |\ \ \
| | |/ /
| |/| | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This addresses potential issues with link management when
multiple routing protocols are in use.
Also fix non-linux build issues
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Make the quagga upstream and vyatta where only difference was
white space.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This reverts commit ebc04ce20a871c99dbb116a4fbada967dd750c43.
Don't really need this and it would be too hard to convince upstream
to take it.
|
| | | |
| | | |
| | | |
| | | | |
Change lost in merge.
|
| | | |
| | | |
| | | |
| | | | |
Change lost in merge.
|
| | | |
| | | |
| | | |
| | | | |
A couple more trivial optimization.
|
| |\ \ \
| | | |/
| | |/|
| | | |
| | | |
| | | | |
Conflicts:
lib/checksum.c
|
| | | |
| | | |
| | | |
| | | |
| | | | |
* lib/sockunion.c: (sockunion_accept) sockunion needs to be normalised here,
as it used to before this was moved to a dedicated function..
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* bgp_vty.c: (show_bgp_views_cmd) new command to list all defined
views.
(with small edits by Paul Jakma)
Signed-off-by: Paul Jakma <paul@quagga.net>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* lib/checksum.c: (fletcher_checksum) Switch the second phase of the checksum
back to the old ospfd logic.
The isisd-derived version:
a) is very hard to follow
b) had some kind of subtle bug that caused it be wrong when c0=0 and c1=254
(potentially fixable by doing the mods before adjusting x and y)
Additionally:
- explicitely cast expressions using non-internal variables to int, to ensure
the result is signed.
- defensively change the length argument to 'size_t', to ensure the code
works with that argument being unsigned..
Thanks to Joakim Tjernlund for the investigative work into this bug.
* tests/test-checksum.c: new file to exercise the checksum code.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Capabilities are always built into the vyatta kernel (don't think
they can even be a module anymore). So avoid overhead of unneeeded
modprobe on boot.
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The new impl. of the Fletcher checkum is buggy. I have
changed back the critical part to the orginal method
and fixed the buggy one, hid it behind UNSIGNED_FLETCHER.
Also did some simple optimizations to reduce the number
of instructions in the hot path for both fletcher_checksum()
and in_cksum().
|
| | | | |
|
| | | | |
|
| |\ \ \
| | | | |
| | | | |
| | | | | |
Merge in current upstream version of quagga.
|
| | |\| |
| | | | |
| | | | |
| | | | | |
Resolve all conflicts by using upstream version.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
* Makefile.am: vtysh should also pull in commands from zebra_routemap.c
Signed-off-by: Paul Jakma <paul@quagga.net>
|
| | | |\ \ |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
* bgpd/bgp_packet.c: (bgp_open_receive) Try fix the little race in the FSM,
where a accept-peer that progress faster than realpeer gets closed down if
realpeer is still just in Connect, by allowing the realpeer to be
bgp_stop'ed and doing the regular swapping-of-FSM state.
|
| | | | | | |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
* bgp_route.c: (show_ip_bgp_view_rsclient_route_cmd) Add the missing "ip"
to the command string.
Signed-off-by: Paul Jakma <paul@quagga.net>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
* ospfd/ospfd.c: (opsf_if_update) Check if ospf_lookup() returns NULL.
Fixes bug introduced with recent ospf_network_run changes in
a49eb30a, where SEGV will happen if not the "router ospf" command has
been executed before the first interfaces are reported by zebra.
Signed-off-by: Paul Jakma <paul@quagga.net>
|