Age | Commit message (Collapse) | Author | |
---|---|---|---|
2004-06-10 | zero the cif struct correctly. | Ryan Thomas McBride | |
2004-06-10 | Plug a memory leak in carp_clone_destroy() | Ryan Thomas McBride | |
From Patrick Latifi | |||
2004-05-31 | initialize struct carp_if and let vhif_nvrs counter work. ok mcbride@ | Marco Pfatschbacher | |
2004-05-30 | compile without INET6; ok mcbride | Peter Valchev | |
2004-05-29 | Remove multicast addresses and promiscuous mode when destroying the | Ryan Thomas McBride | |
carp interface. | |||
2004-05-29 | carp_ioctl() happens after ifioctl(); check flags correctly so that | Ryan Thomas McBride | |
'ifconfig carp0 down' and 'ifconfig carp0 up' work as expected. | |||
2004-05-25 | let sc_counter wrap automatically, ok mcbride@ | Jolan Luff | |
2004-05-24 | Check for sc->sc_ifp != NULL in carp_setroute(). Running dhclient on the | Ryan Thomas McBride | |
carp interface won't work, but it doesn't mean we should panic the box! | |||
2004-05-16 | When we back off due to problems with the physical interface, advertise | Ryan Thomas McBride | |
the backoff immediately on all of the interfaces instead of waiting for the next scheduled advertisement. | |||
2004-05-16 | Remove useless test pointed out by Frank Denis <j@pureftpd.org>. | Ryan Thomas McBride | |
2004-05-13 | Act like our advskew is 240 when receiving packets as well as sending if | Ryan Thomas McBride | |
we're backing off. Makes the backoff actually work like it's supposed to instead of flip-flopping. Problem pointed out by ho@ and jakob@ | |||
2004-05-13 | Add cloned interface destruction to carp (ifconfig carp0 destroy). | Ryan Thomas McBride | |
ok deraadt@ | |||
2004-05-13 | KNF | Ryan Thomas McBride | |
2004-05-10 | Make sure advskew set correctly in advertisement when configured to a value | Ryan Thomas McBride | |
greater than 240. From Frank Denis <j@pureftpd.org>, PR3774. | |||
2004-05-08 | If a carp interface has more than CARP_SENDAD_MAX_ERRORS failed | Ryan Thomas McBride | |
advertisements in a row, back off all the other interfaces, until CARP_SENDAD_MIN_SUCCESS advertisements are successful. This makes carp deal better with ip output errors. | |||
2004-04-28 | Make carp(4) aware of its physical interface: | Ryan Thomas McBride | |
- If the physical interface goes down or the link goes down, the carp interface goes down as well. - We treat this like the preemption holdoff with pfsync. So if one of the carp interfaces is known to be bad (because the physical interface it's associated with is bad), all the other carp interfaces back off: they won't preempt, and their advskew goes to 240. ok cedric@ | |||
2004-03-26 | don't use mbuf after free; ok mcbride, deraadt, cloder | Markus Friedl | |
2004-03-26 | Also raise advskew to 240 while we wait for pfsync to get it's bulk update. | Ryan Thomas McBride | |
This is to deal with situations where the network is not working during boot so we become master by default - the high advskew value allows other carp boxes to preempt us when the network finally comes up. ok deraadt@ beck@ | |||
2004-03-24 | Correct size in error message. From mickey@ | Ryan Thomas McBride | |
ok deraadt@ | |||
2004-03-24 | Disable net.inet.carp.log by default. | Ryan Thomas McBride | |
ok deraadt@ | |||
2004-03-22 | Support for best effort bulk transfers of states when pfsync syncif is | Ryan Thomas McBride | |
configured. This this allows pfsync+carp clusters to come up gracefully without killing active connections. pfsync now prevents carp from preempting to become master until the state table has sync'd. ABI change, any application which use struct pf_state must be recompiled. Reminded about this by Christian Gut. Thanks to beck@ cedric@ and dhartmei@ for testing and comments. ok deraadt@ | |||
2004-03-20 | Unbreak routing change handling for state changes; count addresses differently | Ryan Thomas McBride | |
depending on where we're calling carp_addrcount() from. | |||
2004-03-18 | Allow the state to be changed explicitly via the ioctl. | Ryan Thomas McBride | |
ok markus@ | |||
2004-03-10 | Don't permit a negative index. Pointed out by Patrick Latifi. | Ryan Thomas McBride | |
2004-03-05 | Only send route add or delete messages if it's the first identical | Ryan Thomas McBride | |
address being added or the last identical address being removed, respectively. Part of a larger diff approved by markus@ and dhartmei@, API changes held back for now. | |||
2004-01-18 | send RTM_IFINFO on carp state changes; with mpf@; ok mcbride@, mpf@ | Markus Friedl | |
2004-01-15 | missing #ifdef INET6, ok henning@ | Daniel Hartmeier | |
2004-01-13 | Add and remove IPv6 routes so when we're MASTER, we can connect to the | Ryan Thomas McBride | |
common address. | |||
2003-12-10 | de-register. deraadt ok | Jun-ichiro itojun Hagino | |
2003-12-06 | Mark u_int64_t constants with ULL to make gcc3 happy. | Alexander Yurchenko | |
ok drahn@ | |||
2003-12-03 | add support for ifconfig clone; ok henning deraadt | Markus Friedl | |
2003-11-16 | compile without INET6 | Markus Friedl | |
2003-11-14 | Don't include KAME link-local interface ID in HMAC. | Ryan Thomas McBride | |
from itojun@ | |||
2003-11-09 | No need to pass an unused struct carp_softc * into carp_input_c() from | Ryan Thomas McBride | |
carp_input() and carp6_input(). | |||
2003-11-08 | typos from Jonathon Gray; | Jason McIntyre | |
2003-11-07 | Check to make sure that the packet was received on a carp-enabled interface. | Ryan Thomas McBride | |
Pointed out by Marco Pfatschbacher, ok mickey@ | |||
2003-11-07 | Clean up carp_iamatch() behaviour so we don't reply to arp requests when | Ryan Thomas McBride | |
the interface is ifconfig'd down. Patch from Marco Pfatschbacher | |||
2003-11-07 | Add missing #ifdef INET6. | Ryan Thomas McBride | |
Courtesy of markus@ | |||
2003-11-05 | Cleanups. | Ryan Thomas McBride | |
- Remove error variable which is not needed. - Add missing /* INET6 */ comments. | |||
2003-11-04 | Add carp_setroute(), add and remove routes when the carp(4) interface | Ryan Thomas McBride | |
enters and leaves MASTER state. Allows the system to connect to the common address when it is master. ok cedric@ henning@ | |||
2003-11-04 | Use MGETHDR() instead of m_gethdr() | Ryan Thomas McBride | |
Patch from Marco Pfatschbacher (Marco_Pfatschbacher at genua dot de) | |||
2003-11-04 | - Remove bogus function prototype | Ryan Thomas McBride | |
- Put the correct address family in the BPF header Pointed out by Max Laier. | |||
2003-11-04 | kill a no longer relevant todo item | Michael Shalayeff | |
2003-11-04 | Some cleanups from Andrey Matveev <andrushock@korovino.net>: | Ryan Thomas McBride | |
- remove duplicate string; - carp_set_addr() -> carp_set_addr6() when INET6; - spacing police. Oh yeah, and we can take inet6 support off the todo list. | |||
2003-11-03 | Cleanups: | Ryan Thomas McBride | |
- missing #ifdef INET6. - fix check for interface going down in carp_send_ad(). - fix memory leak which occurs when when ifpromisc() fails. All from Max Laier, thanks. | |||
2003-11-03 | spacing | Theo de Raadt | |
2003-11-03 | Some KNF cleanup. There's probably more to do. | Ryan Thomas McBride | |
2003-11-02 | Typo; #ifdef INET not INET6. | Ryan Thomas McBride | |
From Max Laier. | |||
2003-10-31 | Add IPv6 support to CARP. | Ryan Thomas McBride | |
ok deraadt@ | |||
2003-10-27 | Skew the advertisement interval correctly when we reschedule. | Ryan Thomas McBride | |