Quentin Young [Fri, 28 Apr 2017 22:45:59 +0000 (22:45 +0000)]
lib: allow nonblocking thread_fetch()
This change adds three fields to thread_master and associated code to
use them. The fields are:
* long selectpoll_timeout
This is a millisecond value that, if nonzero, will override the
internally calculated timeout for select()/poll(). -1 indicates
nonblocking while a positive value indicates the desired timeout in
milliseconds.
* bool spin
This indicates whether a call to thread_fetch() should result in a loop
until work is available. By default this is set to true, in order to
keep the default behavior. In this case a return value of NULL indicates
that a fatal signal was received in select() or poll(). If it is set to
false, thread_fetch() will return immediately. NULL is then an
acceptable return value if there is no work to be done.
* bool handle_signals
This indicates whether or not the pthread that owns the thread master
is responsible for handling signals (since this is an MT-unsafe
operation, it is best to have just the root thread do it). It is set to
true by default. Non-root pthreads should set this to false.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Sun, 16 Apr 2017 03:14:36 +0000 (03:14 +0000)]
lib: add pthread manager
Adds infrastructure for keeping track of pthreads.
The general idea is to maintain a daemon-wide table of all pthreads,
running or not. A pthread is associated with its own thread master that
can be used with existing thread.c code, which provides user-space
timers, an event loop, non-blocking I/O callbacks and other facilities.
Each frr_pthread has a unique identifier that can be used to fetch it
from the table. This is to allow naming threads using a macro, for
example:
The idea here is to be relatively flexible with regard to how daemons
manage their collection of pthreads; the implementation could get away
with just some #define'd constants, or keep a dynamically allocated data
structure that provides organization, searching, prioritizing, etc.
Overall this interface should provide a way to maintain the familiar
thread.c userspace threading model while progressively introducing
pthreads.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Mon, 17 Apr 2017 18:33:58 +0000 (18:33 +0000)]
lib: additional thread.c MT-safety work
Fixes a few insufficient critical sections. Adds back locking for
thread_cancel(), since while thread_cancel() is only safe to call from
the pthread which owns the thread master due to races involving
thread_fetch() modifying thread master's ready queue, we still need
mutual exclusion here for all of the other public thread.c functions to
maintain their MT-safety.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Daniel Walton [Wed, 26 Apr 2017 21:45:32 +0000 (21:45 +0000)]
Update to draft-ietf-idr-bgp-prefix-sid-05
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
The initial implementation was against draft-keyupate-idr-bgp-prefix-sid-02
This updates our label-index implementation up to draft-ietf-idr-bgp-prefix-sid-05
- changed BGP_ATTR_LABEL_INDEX to BGP_ATTR_PREFIX_SID
- since there are multiple TLVs in BGP_ATTR_PREFIX_SID you can no longer
rely on that flag to know if there is a label-index for the path. I
changed bgp_attr_extra_new() to init the label_index to
BGP_INVALID_LABEL_INDEX
- put some placeholder code in for the other two TLVs (IPv6 and
Originator SRGB)
Daniel Walton [Wed, 26 Apr 2017 15:53:35 +0000 (15:53 +0000)]
BGP_ATTR_LABEL_INDEX fixes
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
- cleaned up the "show bgp ipv4 labeled-unicast x.x.x.x" output
- fixed some json keys to use camelCase
- bgp_attr_label_index() was clearing BGP_ATTR_LABEL_INDEX because it
was comparing mp_update->afi against SAFI_LABELED_UNICAST instead of
mp_update->safi
Don Slice [Fri, 21 Apr 2017 18:18:42 +0000 (11:18 -0700)]
zebra: fix attempt to install a second rib from imported table entries
Problem reported by a customer with prefix imported by rdnbrd not being
successfully installed in the quagga rib. Determined that this was due
to VRR resolving the same arp entry, causing two entries to be installed
in table 10. When these were imported into the quagga rib, they came in
as two different rib entries from the same table/instance, which is not
permitted and caused them to be deleted.
Added logic to zebra_add_import_table_entry to do the same actions as
rib_add and delete the older rib entry if a new rib entry is received
which matches.
Manual testing successful and automated tests for redistribute neighbor
have the same passes and failures as the base.
Ticket: CM-15926 Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
Reviewed-by:
When the transport address is changed, all interfaces and targeted
neighbors are temporary disabled in the ldpe process until new sockets
bound to the new transport address are received from the parent.
This patch fixes a problem in which adjacencies weren't being removed
after the associated targeted neighbors were disabled. This was causing
ldpd not to set some MD5 sockoptions for new neighbors are thus preventing
MD5-protected sessions to come up after a change in the transport-address.
It's doesn't make sense to enforce that a targeted-hello is received
on an LDP-enabled interface. It should be possible, for example, to use
LDP only to signal pseudowires and other another protocol (e.g. RSVP-TE)
to create end-to-end LSPs.
ldpd: schedule the sending of label messages when necessary
Once we send a Label Withdraw, we can't send a Label Mapping for the
same FEC until we receive a Label Release from the peer. This is due to
some limitations in the LDP algorithms described in Appendix A. ("LDP
Label Distribution Procedures") of RFC 5036.
To workaround this issue, make it possible to schedule the sending of
a Label Mapping as soon as a Label Release is received for the same FEC.
The easiest way to test this patch is by typing the "label local advertise
explicit-null" command. ldpd will withdraw all null labels using a
Wildcard FEC and then send new Label Mappings as soon the corresponding
Label Releases are received.
On unstable networks, routes can be lost and relearned very often. If
we deallocate the input label every time a route is lost and allocate
a new one when the route is relearned, a lot of changes are made in vain.
This patch introduces a logic in which labels are preserved for at least
five minutes before being deallocated by the LIB garbage collector. This
is consistent with what other implementations do.
ldpd: accept notifications during the session establishment process
If we don't do this, we'll never trigger the backoff exponential timer
since it's impossible to distinguish between Initialization NAK's and
general errors.
Also:
* Implement some missing bits from RFC 5036;
* remove superfluous log message in session_shutdown()
(send_notification() logs that we're sending a fatal notification).
ldpd: use synchronous channels for sending log messages
This is necessary to guarantee that all log messages sent from the child
processes are received in the parent process right away.
Without this patch, when a child process calls fatal() or fatalx(),
the log messages don't make it to the parent because the child doesn't
have a chance to flush its buffers before exiting.
When ldpd fails to start for some reason, like failing to create a pid
file, the child processes call their shutdown functions without being
completely initialized. This patch adds some protections to prevent a
segmentation fault on such circumstances.
ldpd: simplify initialization of the child processes
In order to have separate ASLR/cookies per process, ldpd calls exec()
in the child processes after fork() (this is also known as the fork+exec
model).
This is an important security feature but it makes the initialization
of the child processes a bit more complicated as they're not a copy of
the parent anymore, so all parameters given via command line are lost.
To solve this problem, we were creating an argv array by hand with all
necessary parameters and providing it to the exec() syscall. This works
but it's a very ugly solution. This patch introduces a different approach
to solve the problem: send an IMSG_INIT message to the child processes
with all parameters they need in order to initialize properly. This
makes adding additional initialization parameters much more convenient
and less error prone.
'ip route show' displays all routes as belonging to protocol zebra.
The user has to run an additional command (in vtysh) to get the actual
source of a route (bgp/ospf/static etc.). This patch addresses that by
pushing the appropriate protocol string into the protocol field of the
netlink route update message. Now you can see routes with the correct
origin as well as filter on them (ip route show proto ospf).
'ospf' is used for both IPv4 and IPv6 routes, even though the OSPF
version is different in both cases.
Sample output (old):
9.9.12.13 via 69.254.2.38 dev swp3.2 proto zebra metric 20
9.9.13.3 proto zebra metric 20
nexthop via 69.254.2.30 dev swp1.2 weight 1
nexthop via 69.254.2.34 dev swp2.2 weight 1
nexthop via 69.254.2.38 dev swp3.2 weight 1
Sample output (new):
9.9.12.13 via 69.254.2.38 dev swp3.2 proto bgp metric 20
9.9.13.3 proto bgp metric 20
nexthop via 69.254.2.30 dev swp1.2 weight 1
nexthop via 69.254.2.34 dev swp2.2 weight 1
nexthop via 69.254.2.38 dev swp3.2 weight 1
Martin Winter [Wed, 12 Apr 2017 23:00:43 +0000 (16:00 -0700)]
snapcraft: Improve README.usage.md based on feedback received
- Fix snap connect (it's now called core, not ubuntu-core)
- Add section on MPLS configuration
- Add FAQ topic on ospfd/ospf6d crashing before privs are assigned to snap
- Add pointer to official webpage
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
Don Slice [Tue, 18 Apr 2017 12:11:32 +0000 (08:11 -0400)]
zebra: stop crash on process termination due to stale ifp->node
Problem reported that crash occurred when stopping quagga in certain
circumstances. Determined that this was due to a stale pointer on the
ifp for a deleted interface. The ifp->node had been freed but the ifp
still kept a pointer to it, and when later the process was stopped, it
attempted to delete it again.
Ticket: CM-15550 Signed-off-by: Don Slice <dslice@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Sat, 15 Apr 2017 17:25:03 +0000 (13:25 -0400)]
zebra: Add some more checks to fec [un]registration
Be a bit more rigoruous about what we can receive
from another protocol and attempt to make the code
less likely to crash and to just safely bail
out when an error is received.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Sat, 15 Apr 2017 13:31:59 +0000 (09:31 -0400)]
bgpd: Fix up several issues in bgp_route.c
1) Some commands were installed in the wrong node
2) Fix output to show labeled information to correctly display
3) Whitespace issue in route-map command
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Sat, 15 Apr 2017 12:42:18 +0000 (08:42 -0400)]
zebra: Fix crash with dereference of NULL pointer
We only create the v4 and v6 mpls fec tables currently.
Follow the code pattern for the rest of the code and
ensure that the table exists before we attempt to access it.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Sat, 15 Apr 2017 12:33:43 +0000 (08:33 -0400)]
lib, bgpd: Remove UNDEFINED_NODE
Remove the UNDEFINED_NODE as that it's implementation breaks
our ability in BGP to figure out where we are by allowing
default: in the switch statement.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
projects / mirror / frr.git / log