Age | Commit message (Collapse) | Author |
|
All daemons modified to support custom path to zserv
socket.
lib: generalize a zclient connection
zclient_socket_connect added. zclient_socket and
zclient_socket_un were hidden under static expression.
"zclient_serv_path_set" modified.
|
|
Move zserv socket creation code into zebra_zserv_socket_init() and
call it only after pidfile lock has been acquired exclusively. This
keeps subsequent zebra daemons from deleting the working socket of
an already running process (bug #403).
|
|
|
|
|
|
2007-05-01 David L Stevens <dlstevens@us.ibm.com>
* (general) These changes collectively add route-map and
prefix-list support to zebra and fix a bug in "show
route-map" (with no argument).
* doc/main.texi: added route-map, prefix-list, ip protocol
and set src documentation
* lib/command.h: added PROTOCOL_NODE type
* lib/log.c: (proto_name2num) new function, protocol name to
number translation.
* lib/routemap.c: (vty_show_route_map) fixed "show route-map"
without route-map name
* lib/routemap.h: added RMAP_ZEBRA type
* lib/zebra.h: added proto_name2num() prototype
* vtysh/extract.pl.in: added VTYSH_ZEBRA flag for route-map and
plist
* vtysh/Makefile.am: added zebra_routemap.c
* vtysh/vtysh.h: added VTYSH_ZEBRA flag to VTYSH_RMAP
* zebra/connected.c: (connected_up_ipv4) added src preference argument
to rib_add_ipv4()
* zebra/kernel_socket.c: (rtm_read) ditto
* zebra/main.c: added prefix list initialization
* zebra/Makefile.am: added zebra_routemap.c source file
* zebra/rib.h: added generic address union "g_addr" and use in
existing places that had an explicit union.
Added "src" to struct nexthop.
Added preferred src arg to nexthop_ipv4_add and rib_add_ipv4.
* zebra/rt_netlink.c: (netlink_routing_table) set preferred source on
netlink messages.
(netlink_route_change) ditto
(netlink_route_multipath) ditto.
* zebra/rtread_getmsg.c: (handle_route_entry) added (NULL) src to
rib_add_ipv4() call.
* zebra/rtread_proc.c: (proc_route_read) ditto
* zebra/zebra_rib.c: (nexthop_ipv4_add) add src argument.
(nexthop_ipv4_ifindex_add) ditto
(rib_add_ipv4) ditto
(nexthop_active_check) Add route-map processing.
* zebra/zebra_routemap.c: new file for zebra route-map commands.
* zebra/zebra_vty.c: (ip_protocol_cmd) Apply route-map to protocol
(vty_show_ip_route_detail) added "src" printing
(vty_show_ip_route) ditto
(show_ip_protocol_cmd) new command, list routemaps.
(config_write_protocol) write out routemap protocl config.
(zebra_vty_init) Install the new routemap protocol commands.
* zebra/zserv.c: (zread_ipv4_add) added (NULL) src arg
(zebra_init) init zebra route-maps.
* zebra/zserv.h: add zebra_route_map_init
|
|
2005-11-24 Paul Jakma <paul.jakma@sun.com>
* kernel_socket.h: New header for functions exported to sysctl
methods.
* kernel_socket.c: include previous.
Remove static qualifier from couple of functions which are
used by sysctl methods, incorrectly added in previous commit.
Add a workaround for a bogus gcc warning to the RTA_ macros.
* Makefile.am: Add kernel_socket.h to noinst_HEADERS
* if_sysctl.c: include rt.h and kernel_socket.h and remove
redundant prototypes.
* rtread_sysctl.c: ditto.
(route_read) fix mismatch of return values.
* {rt,zserv,rib}.h: Include lib headers depended on.
|
|
* interface.c: (if_delete_update) should always be available, not
just on RTM_IFANNOUNCE/NETLINK systems.
* kernel_socket.c: (ifan_read) only call if_delete_update when
interface departs, dont if_delete, because we wish to retain
interface configuration state even when interfaces are removed.
(ifm_read) If we dont have RTM_IFANNOUNCE, then transitioning
to down state is only chance we have to clean up interface in case
it is deleted (eg Solaris down -> unplumb -> plumb up).
* redistribute.c: (zebra_interface_delete_update) should always be
available, we /will/ call it now on all systems, via
if_delete_update.
* zserv.c: (zsend_interface_delete) ditto
(zsend_interface_address) Update the call-flow diagramme, to
reflect that if_delete_update /is/ now called on all systems,
potentially.
* zserv.h: (zsend_interface_delete) unconditionally exported, as
above.
|
|
* (global) Extern and static'ification, with related fixups
of declarations, ensuring files include their own headers, etc.
if_ioctl.c: (interface_info_ioctl) fix obvious arg mis-order in
list loop
|
|
* zserv.h: Remove ZEBRA_PORT definition, its in lib/zebra.h now
|
|
* rib.h: (struct rib) Add lock field for refcounting.
* zserv.h: (struct zebra_t) Add a ribq workqueue to the zebra
'master' struct.
* zserv.c: (zread_ipv4_add) XMALLOC then memset should be XCALLOC.
* zebra_rib.c: Clean up refcounting of route_node, make struct rib
refcounted and convert rib_process to work-queue. In general,
rib's should be rib_addnode'd and delnode'd to route_nodes, and
these symmetrical functions will manage the locking of referenced
route_node and freeing of struct rib - rather than having users
manage each seperately - with much scope for bugs..
(newrib_free) removed and replaced with rib_lock
(rib_lock) new function, check state of lock and increment.
(rib_unlock) new function, check lock state and decrement. Free
struct rib if refcount hits 0, freeing struct nexthop's, as
newrib_free did.
(rib_addnode) Add RIB to route_node, locking both.
(rib_delnode) Delete RIB from route_node, unlocking each.
(rib_process) Converted to a work-queue work function.
Functional changes are minimal, just arguments, comments and
whitespace.
(rib_queue_add_qnode) Helper function to setup a ribq item.
(rib_queue_add) Helper function, same arguments as old
rib_process, to replace in callers of rib_process.
(rib_queue_qnode_del) ribq deconstructor.
(rib_queue_init) Create the ribq.
(rib_init) call rib_queue_init.
(remainder) Sanitise refcounting of route_node's. Convert to
rib_queue_add, rib_addnode and rib_delnode. Change XMALLOC/memset
to XCALLOC. Remove calls to nexthop_delete and nexthop_free.
|
|
* zserv.c: Must include network.h and buffer.h for non-blocking I/O.
Remove global message_queue and t_write (need separate buffering for
each client).
(zebra_server_dequeue,zebra_server_enqueue) Remove functions
related to old buggy buffering code.
(zserv_delayed_close) New thread callback function to delete a client.
(zserv_flush_data) New thread callback function to flush buffered
data to client.
(zebra_server_send_message) Rewritten to use buffer_write (so
buffering of writes and non-blocking I/O work properly).
(zsend_interface_add,zsend_interface_delete,zsend_interface_address,
zsend_interface_update) Return 0 instead of -1 if !client->ifinfo
(this is not really an error). Return value from
zebra_server_send_message.
(zsend_route_multipath,zsend_ipv4_nexthop_lookup,
zsend_ipv4_import_lookup) Return value from zebra_server_send_message.
(zsend_ipv6_nexthop_lookup) Fix scope to static, and return value
from zebra_server_send_message.
(zsend_router_id_update) Must use zebra_server_send_message instead
of deprecated writen function. Return 0 instead of -1 if this client
is not subscribed to router-id updates (since this is not really
an error).
(zread_interface_add) Change type to static int. If
zsend_interface_add fails or zsend_interface_address fails, return -1
immediately (since the client has had an I/O error).
(zread_interface_delete,zread_ipv4_add,zread_ipv4_delete,
zread_ipv6_add,zread_ipv6_delete,zread_router_id_delete) Return 0
to indicate success.
(zread_ipv4_nexthop_lookup) Return value from
zsend_ipv4_nexthop_lookup.
(zread_ipv4_import_lookup) Return value from zsend_ipv4_import_lookup.
(zebra_read_ipv6) Remove unused function.
(zread_ipv6_nexthop_lookup) Return value from
zsend_ipv6_nexthop_lookup.
(zread_router_id_add) Return value from zsend_router_id_update.
(zebra_client_close) Call buffer_free(client->wb) and
thread_cancel(client->t_suicide).
(zebra_client_create) Allocate client->wb using buffer_new.
(zebra_client_read) Support non-blocking I/O by using stream_read_try.
Use ZEBRA_HEADER_SIZE instead of 3.
(zebra_accept) Fix bug: reset accept thread at top. Make client
socket non-blocking using the set_nonblocking function.
(config_write_forwarding) Fix scope to static.
(zebra_init) Remove initialization code for old buggy write buffering.
* zserv.h: Add 2 new fields to struct zserv: struct buffer *wb
(to enable buffered writes with non-blocking I/), and
struct thread *t_suicide to support delayed close on I/O
errors.
* router-id.h: Remove prototypes for zread_router_id_add and
zread_router_id_delete (their scope should be static to zserv.c).
|
|
|
|
|
|
* if_ioctl_solaris.c: HAVE_IPV6 ifdef fixups
* zserv.c: ditto
* ioctl_solaris.c: ditto.
* interface.c: cast for LLADDR
* interface.h: Add guards, include redistribute.h and remove
extraneous definitions of zebra_interface_{up,down}_update
* ioctl.h: Add AF_IOCTL define for non SOLARIS_IPV6
* redistribute.h: include dependent header, zserv.h
* zserv.h: include dependent header, rib.h
|
|
* zserv.c: (zsend_route_multipath) Set the nexthop_num
field correctly. Add NEXTHOP_TYPE_IPV6_IFNAME for v6.
Conditionally set ZAPI_MESSAGE_METRIC flag - only for adds.
(zsend_ipv4_add) cruft, deleted.
(zsend_ipv4_delete) ditto.
(zsend_ipv6_add) ditto.
(zsend_ipv6_delete) ditto.
2004-05-09 Sowmini Varadhan <sowmini.varadhan@sun.com>
* zserv.c: (zsend_ipv{4,6}_{add,delete}_multipath) collapsed
into single zsend_route_multipath function.
(zsend_interface_{up,down}) collapsed into zsend_interface_update.
(zsend_interface_address_{add,delete}) collapsed into
zsend_interface_address.
(zsend_interface_add) send mtu6.
(zsend_interface_delete) ditto.
(zebra_write) remove unused function.
(various) Apply static qualifier. Add comments.
* zserv.h: Definitions changed as per above.
* redistribute.c: Changes as per zserv.c.
|
|
* lib/vty.{c,h}: Remove vty layer depending on a 'master' global,
pass the thread master in explicitly to vty_init. Sort out some
header dependency problems with lib/command.h
* zebra/: Move globals to struct zebrad. Update vty_init().
* (.*)/\1_main.c: update call to vty_init().
|
|
|
|
moved definition of the various socket paths from the
per daemon header files into configure.ac. it will set the paths to
be in the directory specified by --localstatedir=<prefix> or
otherwise will try to guess as best it can ( a la pid file path
detection - which probably should try reference ${prefix} too).
the present hardcoded socket path, /tmp, isnt really correct. should
be in /var somewhere really.
|
|
|