Fix bug 544: isisd produces an unexpected routing table for wide-metric.
* isis_spf.c: Accept VTYPE_PSEUDO_TE_IS and VTYPE_NONPSEUDO_TE_IS
vertex types for SPF calculation
* isis_pdu.c: Change order of TLVs to match Cisco to make bitwise
comparison easier for Wireshark
* isis_tlv.c: EXTREME_TLV_DEBUG for TLV debugging instead of
EXTREME_DEBUG
The forwarding table was filled with wrong next-hops, and which is even
worse, it was done in a totally non-deterministic way.
The next-hop set for an IP prefix by isisd was the neighbor IS from
which the flooded LSP about the IP prefix was arrived. So, if an IS
received all the LSPs through its, say, eth0 interface, all entries
in the forwarding table contained the next IS reachable via eth0 as
the next-hop.
The solution is to propagate the correct next-hop further from node to
node as the SPF algorithm traverses the graph and selects the next
node to be added to the set of already covered nodes.
Also, the construction of the tentative node list (the nodes where the
shortest path is not known yet) was buggy: if a node was already a
member of this list with a certain path cost, and an alternative path
was found to it with a lower cost while processing a pseudo-node LSP,
it was not added to the list. This way, the path selected by isisd for
a certain prefix was the first one it encountered during the LSDB
processing.
Signed-off-by: Fritz Reichmann <fritz@reichmann.nl>
After an IS has been elected as the Designated IS for a LAN, it did
not refresh the content of the pseudo-node after a new node has been
connected to the same LAN. Instead, the periodically reoriginated
pseudo-node LSP still contained only those IS neighbors that were
already present when the DIS election process was commenced.
The fix for the problem schedules an LSP regeneration rather than
just reoriginating the same LSP with the old content.
Signed-off-by: Fritz Reichmann <fritz@reichmann.nl>
isisd has a so-called circuit state machine that takes care about
the interface state changes, such as initializing, down, up. When
an interface was brought down by a link failure, the interface
information was deleted and set to NULL. When the link was restored
later, the interface was looked up by the old pointer, but since it
was cleared, it was never found again, resulting in an interface
never entering the up state again.
Also, the program regularly crashed because of a deleted pointer in
the same context which was later accessed without any further
checking.
Signed-off-by: Fritz Reichmann <fritz@reichmann.nl>
The crash is due to threads accessing data that gets destroyed
during the removal of the configuration.
* isis_circuit.c: Destroy adjacencies to stop adjacency expiry thread.
Stop PSNP threads.
* isisd.c: Change state of circuit back to INIT and reassign the
circuit structure to isis->init_circ_list rather than destroying
the circuit data structure. Stop SPF threads. Stop LSP generation
threads.
* isisd.h: Add pointers to LSP threads into area structure in order to
stop them in isisd.c
* isis_lsp.c: Store pointer to LSP thread in area structure.
* isis_pdu.c: Stop PDU generation for a circuit with a removed area.
* isis_pfpacket.c: Stop processing received PDUs for a circuit with a
removed area.
* isisd/isis_pdu.c
* maskbit: this pre-initialized array is not modified
* lib/prefix.c
* maskbit: idem
* lib/command.c
* facility_map: idem
* itoa64: idem
* default_motd: make local var static
* facility_name(): update local var accordingly
* facility_match(): idem
2008-08-13 Jingjing Duan <Jingjing.Duan@sun.com>
* ospfd/: Remove the old checksum implementation and
use the consolidated version.
* isisd/: ditto, thus fixing isisd checksuming on big-endian.
Signed-off-by: Paul Jakma <paul@quagga.net>
2005-11-20 Paul Jakma <paul.jakma@sun.com>
* (general) remove includes of very common system headers,
these are already picked up by zebra.h. Further, including
them before zebra.h messes up all our lovely (sarcasm)
autoconf'd detecting of platform specific things. Recent
addition of stdint.h to configure.ac and zebra.h appears
particularly to throw up this problem.
Replace XMALLOC + memset with XCALLOC. Fix some indentation issues.
The only really significant change is simplified isis_delete_adj
function in isis_adjacency.c.
its data and put it back after. Database entry MUST contain at least
correct pointers to the sysid to get correct compare results.
* isis_lsp.[ch], isis_pdu.c: Pass level to the lsp_update() function.
anything. These floats to longs and vice versa casts are starnge
indeed.
* isis_pdu.c: As we don't use %z for size_t for now because we support
older compilers, cast them to unsigned long.
Also fix previous changelog entry. Isisd compiles cleanly now again.
* (general) More cleaning up of stream abuse, isisd should be
back to previous functionality. Replace various XMALLOC+memset's
with XCALLOC
* isis_tlv.c: (tlv_add_padding) use stream_put to clear the stream
rather than forward endp, as isisd reuses streams.
* isis_pdu.c: (process_lsp) cleanup direct reference to stream endp
(send_lsp) manual copy of a stream cleaned up to use stream_copy.
* isis_network.c: (isis_recv_pdu_bcast) replace direct memcpy with
stream_write
(isis_recv_pdu_p2p) replace recvfrom directly into stream with
stream_recvfrom. Remove dangerous and now unneeded manual update
of endp.
(isis_recv_pdu_bcast / non-GNU_LINUX) Replace direct memcpy with
stream_write.
(isis_recv_pdu_p2p) replace read direct into stream with
stream_read_try, and hence remove the manual update of endp.
* isis_lsp.c: (lsp_update_data) manual stream dup replaced with
stream_dup.
(lsppdu_realloc) mempcy into stream data replaced with stream_put.
(lsp_build_nonpseudo) remove mysterious stream_forward_endp's -
which were originally stream_set_putp - shouldn't be needed
now that all the manual fiddling of private stream data has been
corrected.
(build_topology_lsp_data) remove unneeded twiddling of endp,
appears to be due to lsppdu_realloc(), but it appears to sort of
do the right thing wrt streams.
* (global): Fix up list loops to match changes in lib/linklist,
and some basic auditing of usage.
* configure.ac: define QUAGGA_NO_DEPRECATED_INTERFACES
* HACKING: Add notes about deprecating interfaces and commands.
* lib/linklist.h: Add usage comments.
Rename getdata macro to listgetdata.
Rename nextnode to listnextnode and fix its odd behaviour to be
less dangerous.
Make listgetdata macro assert node is not null, NULL list entries
should be bug condition.
ALL_LIST_ELEMENTS, new macro, forward-referencing macro for use
with for loop, Suggested by Jim Carlson of Sun.
Add ALL_LIST_ELEMENTS_RO for cases which obviously do not need the
"safety" of previous macro.
LISTNODE_ADD and DELETE macros renamed to ATTACH, DETACH, to
distinguish from the similarly named functions, and reflect their
effect better.
Add a QUAGGA_NO_DEPRECATED_INTERFACES define guarded section
with the old defines which were modified above,
for backwards compatibility - guarded to prevent Quagga using it..
* lib/linklist.c: fix up for linklist.h changes.
* ospf6d/ospf6_abr.c: (ospf6_abr_examin_brouter) change to a single
scan of the area list, rather than scanning all areas first for
INTER_ROUTER and then again for INTER_NETWORK. According to
16.2, the scan should be area specific anyway, and further
ospf6d does not seem to implement 16.3 anyway.
* (global) Update code to match stream.h changes.
stream_get_putp effectively replaced with stream_get_endp.
stream_forward renamed to stream_forward_getp.
stream_forward_endp introduced to replace some previous
setting/manual twiddling of putp by daemons.
* lib/stream.h: Remove putp. Update reference to putp with endp.
Add stream_forward_endp, which daemons were doing manually.
Rename stream_forward to stream_forward_getp.
lib/stream.c: Remove/update references to putp.
introduce stream_forward_endp.
