rgb-cln/lightningd/channel_control.c

794 lines
24 KiB
C
Raw Normal View History

#include <bitcoin/pubkey.h>
#include <bitcoin/script.h>
#include <ccan/cast/cast.h>
#include <channeld/gen_channel_wire.h>
#include <common/features.h>
#include <common/gossip_constants.h>
#include <common/json_command.h>
#include <common/jsonrpc_errors.h>
#include <common/memleak.h>
#include <common/per_peer_state.h>
#include <common/timeout.h>
#include <common/utils.h>
#include <common/wallet_tx.h>
#include <common/wire_error.h>
#include <errno.h>
#include <hsmd/gen_hsm_wire.h>
#include <inttypes.h>
#include <lightningd/channel_control.h>
#include <lightningd/closing_control.h>
#include <lightningd/hsm_control.h>
#include <lightningd/jsonrpc.h>
#include <lightningd/lightningd.h>
#include <lightningd/log.h>
#include <lightningd/peer_control.h>
#include <lightningd/subd.h>
#include <wire/wire_sync.h>
static void update_feerates(struct lightningd *ld, struct channel *channel)
{
u8 *msg;
u32 feerate = unilateral_feerate(ld->topology);
/* Nothing to do if we don't know feerate. */
if (!feerate)
return;
msg = towire_channel_feerates(NULL, feerate,
feerate_min(ld, NULL),
feerate_max(ld, NULL));
subd_send_msg(channel->owner, take(msg));
}
static void try_update_feerates(struct lightningd *ld, struct channel *channel)
{
/* No point until funding locked in */
if (!channel_fees_can_change(channel))
return;
/* Can't if no daemon listening. */
if (!channel->owner)
return;
update_feerates(ld, channel);
}
void notify_feerate_change(struct lightningd *ld)
{
struct peer *peer;
/* FIXME: We should notify onchaind about NORMAL fee change in case
* it's going to generate more txs. */
list_for_each(&ld->peers, peer, list) {
struct channel *channel = peer_active_channel(peer);
if (!channel)
continue;
/* FIXME: We choose not to drop to chain if we can't contact
* peer. We *could* do so, however. */
try_update_feerates(ld, channel);
}
}
static void lockin_complete(struct channel *channel)
{
/* We set this once we're locked in. */
assert(channel->scid);
/* We set this once they're locked in. */
assert(channel->remote_funding_locked);
/* We might have already started shutting down */
if (channel->state != CHANNELD_AWAITING_LOCKIN) {
log_debug(channel->log, "Lockin complete, but state %s",
channel_state_name(channel));
return;
}
channel_set_state(channel, CHANNELD_AWAITING_LOCKIN, CHANNELD_NORMAL);
/* Fees might have changed (and we use IMMEDIATE once we're funded),
* so update now. */
try_update_feerates(channel->peer->ld, channel);
}
/* We were informed by channeld that it announced the channel and sent
* an update, so we can now start sending a node_announcement. The
* first step is to build the provisional announcement and ask the HSM
* to sign it. */
static void peer_got_funding_locked(struct channel *channel, const u8 *msg)
{
struct pubkey next_per_commitment_point;
if (!fromwire_channel_got_funding_locked(msg,
&next_per_commitment_point)) {
channel_internal_error(channel,
"bad channel_got_funding_locked %s",
tal_hex(channel, msg));
return;
}
if (channel->remote_funding_locked) {
channel_internal_error(channel,
"channel_got_funding_locked twice");
return;
}
update_per_commit_point(channel, &next_per_commitment_point);
log_debug(channel->log, "Got funding_locked");
channel->remote_funding_locked = true;
if (channel->scid)
lockin_complete(channel);
}
static void peer_got_announcement(struct channel *channel, const u8 *msg)
{
secp256k1_ecdsa_signature remote_ann_node_sig;
secp256k1_ecdsa_signature remote_ann_bitcoin_sig;
if (!fromwire_channel_got_announcement(msg,
&remote_ann_node_sig,
&remote_ann_bitcoin_sig)) {
channel_internal_error(channel,
"bad channel_got_announcement %s",
tal_hex(tmpctx, msg));
return;
}
wallet_announcement_save(channel->peer->ld->wallet, channel->dbid,
&remote_ann_node_sig,
&remote_ann_bitcoin_sig);
}
static void peer_got_shutdown(struct channel *channel, const u8 *msg)
{
u8 *scriptpubkey;
struct lightningd *ld = channel->peer->ld;
if (!fromwire_channel_got_shutdown(channel, msg, &scriptpubkey)) {
channel_internal_error(channel, "bad channel_got_shutdown %s",
tal_hex(msg, msg));
return;
}
/* FIXME: Add to spec that we must allow repeated shutdown! */
tal_free(channel->shutdown_scriptpubkey[REMOTE]);
channel->shutdown_scriptpubkey[REMOTE] = scriptpubkey;
/* BOLT #2:
*
* 1. `OP_DUP` `OP_HASH160` `20` 20-bytes `OP_EQUALVERIFY` `OP_CHECKSIG`
* (pay to pubkey hash), OR
* 2. `OP_HASH160` `20` 20-bytes `OP_EQUAL` (pay to script hash), OR
* 3. `OP_0` `20` 20-bytes (version 0 pay to witness pubkey), OR
* 4. `OP_0` `32` 32-bytes (version 0 pay to witness script hash)
*
* A receiving node:
*...
* - if the `scriptpubkey` is not in one of the above forms:
* - SHOULD fail the connection.
*/
if (!is_p2pkh(scriptpubkey, NULL) && !is_p2sh(scriptpubkey, NULL)
&& !is_p2wpkh(scriptpubkey, NULL) && !is_p2wsh(scriptpubkey, NULL)) {
channel_fail_permanent(channel, "Bad shutdown scriptpubkey %s",
tal_hex(channel, scriptpubkey));
return;
}
/* If we weren't already shutting down, we are now */
if (channel->state != CHANNELD_SHUTTING_DOWN)
channel_set_state(channel,
channel->state, CHANNELD_SHUTTING_DOWN);
/* TODO(cdecker) Selectively save updated fields to DB */
wallet_channel_save(ld->wallet, channel);
}
static void channel_fail_fallen_behind(struct channel *channel, const u8 *msg)
{
if (!fromwire_channel_fail_fallen_behind(channel, msg,
cast_const2(struct pubkey **,
&channel->future_per_commitment_point))) {
channel_internal_error(channel,
"bad channel_fail_fallen_behind %s",
tal_hex(tmpctx, msg));
return;
}
/* per_commitment_point is NULL if option_static_remotekey, but we
* use its presence as a flag so set it any valid key in that case. */
if (!channel->future_per_commitment_point) {
struct pubkey *any = tal(channel, struct pubkey);
if (!channel->option_static_remotekey) {
channel_internal_error(channel,
"bad channel_fail_fallen_behind %s",
tal_hex(tmpctx, msg));
return;
}
if (!pubkey_from_node_id(any, &channel->peer->ld->id))
fatal("Our own id invalid?");
channel->future_per_commitment_point = any;
}
/* Peer sees this, so send a generic msg about unilateral close. */
channel_fail_permanent(channel, "Awaiting unilateral close");
}
static void peer_start_closingd_after_shutdown(struct channel *channel,
const u8 *msg,
const int *fds)
{
struct per_peer_state *pps;
if (!fromwire_channel_shutdown_complete(tmpctx, msg, &pps)) {
channel_internal_error(channel, "bad shutdown_complete: %s",
tal_hex(msg, msg));
return;
}
per_peer_state_set_fds_arr(pps, fds);
/* This sets channel->owner, closes down channeld. */
peer_start_closingd(channel, pps, false, NULL);
channel_set_state(channel, CHANNELD_SHUTTING_DOWN, CLOSINGD_SIGEXCHANGE);
}
static void handle_error_channel(struct channel *channel,
const u8 *msg)
{
struct command **forgets = tal_steal(tmpctx, channel->forgets);
channel->forgets = tal_arr(channel, struct command *, 0);
if (!fromwire_channel_send_error_reply(msg)) {
channel_internal_error(channel, "bad send_error_reply: %s",
tal_hex(tmpctx, msg));
return;
}
/* Forget the channel. */
delete_channel(channel);
for (size_t i = 0; i < tal_count(forgets); i++) {
assert(!forgets[i]->json_stream);
struct json_stream *response;
response = json_stream_success(forgets[i]);
json_add_string(response, "cancelled", "Channel open canceled by RPC(after fundchannel_complete)");
was_pending(command_success(forgets[i], response));
}
tal_free(forgets);
}
static unsigned channel_msg(struct subd *sd, const u8 *msg, const int *fds)
{
enum channel_wire_type t = fromwire_peektype(msg);
switch (t) {
case WIRE_CHANNEL_SENDING_COMMITSIG:
peer_sending_commitsig(sd->channel, msg);
break;
case WIRE_CHANNEL_GOT_COMMITSIG:
peer_got_commitsig(sd->channel, msg);
break;
case WIRE_CHANNEL_GOT_REVOKE:
peer_got_revoke(sd->channel, msg);
break;
case WIRE_CHANNEL_GOT_FUNDING_LOCKED:
peer_got_funding_locked(sd->channel, msg);
break;
case WIRE_CHANNEL_GOT_ANNOUNCEMENT:
peer_got_announcement(sd->channel, msg);
break;
case WIRE_CHANNEL_GOT_SHUTDOWN:
peer_got_shutdown(sd->channel, msg);
break;
case WIRE_CHANNEL_SHUTDOWN_COMPLETE:
/* We expect 3 fds. */
if (!fds)
return 3;
peer_start_closingd_after_shutdown(sd->channel, msg, fds);
break;
case WIRE_CHANNEL_FAIL_FALLEN_BEHIND:
channel_fail_fallen_behind(sd->channel, msg);
break;
case WIRE_CHANNEL_SEND_ERROR_REPLY:
handle_error_channel(sd->channel, msg);
break;
/* And we never get these from channeld. */
case WIRE_CHANNEL_INIT:
case WIRE_CHANNEL_FUNDING_DEPTH:
case WIRE_CHANNEL_OFFER_HTLC:
case WIRE_CHANNEL_FULFILL_HTLC:
case WIRE_CHANNEL_FAIL_HTLC:
case WIRE_CHANNEL_GOT_COMMITSIG_REPLY:
case WIRE_CHANNEL_GOT_REVOKE_REPLY:
case WIRE_CHANNEL_SENDING_COMMITSIG_REPLY:
case WIRE_CHANNEL_SEND_SHUTDOWN:
case WIRE_CHANNEL_DEV_REENABLE_COMMIT:
case WIRE_CHANNEL_FEERATES:
case WIRE_CHANNEL_SPECIFIC_FEERATES:
case WIRE_CHANNEL_DEV_MEMLEAK:
/* Replies go to requests. */
case WIRE_CHANNEL_OFFER_HTLC_REPLY:
case WIRE_CHANNEL_DEV_REENABLE_COMMIT_REPLY:
case WIRE_CHANNEL_DEV_MEMLEAK_REPLY:
case WIRE_CHANNEL_SEND_ERROR:
break;
}
return 0;
}
void peer_start_channeld(struct channel *channel,
struct per_peer_state *pps,
const u8 *funding_signed,
bool reconnected)
{
u8 *initmsg;
int hsmfd;
struct added_htlc *htlcs;
enum htlc_state *htlc_states;
struct fulfilled_htlc *fulfilled_htlcs;
enum side *fulfilled_sides;
const struct failed_htlc **failed_htlcs;
enum side *failed_sides;
struct short_channel_id scid;
u64 num_revocations;
struct lightningd *ld = channel->peer->ld;
const struct config *cfg = &ld->config;
bool reached_announce_depth;
struct secret last_remote_per_commit_secret;
secp256k1_ecdsa_signature *remote_ann_node_sig, *remote_ann_bitcoin_sig;
hsmfd = hsm_get_client_fd(ld, &channel->peer->id,
channel->dbid,
HSM_CAP_SIGN_GOSSIP
| HSM_CAP_ECDH
| HSM_CAP_COMMITMENT_POINT
| HSM_CAP_SIGN_REMOTE_TX);
channel_set_owner(channel,
new_channel_subd(ld,
"lightning_channeld", channel,
&channel->peer->id,
channel->log, true,
channel_wire_type_name,
channel_msg,
channel_errmsg,
channel_set_billboard,
take(&pps->peer_fd),
take(&pps->gossip_fd),
take(&pps->gossip_store_fd),
take(&hsmfd), NULL));
if (!channel->owner) {
log_unusual(channel->log, "Could not subdaemon channel: %s",
strerror(errno));
channel_fail_reconnect_later(channel,
"Failed to subdaemon channel");
return;
}
peer_htlcs(tmpctx, channel, &htlcs, &htlc_states, &fulfilled_htlcs,
&fulfilled_sides, &failed_htlcs, &failed_sides);
if (channel->scid) {
scid = *channel->scid;
reached_announce_depth
= is_scid_depth_announceable(&scid,
get_block_height(ld->topology));
log_debug(channel->log, "Already have funding locked in%s",
reached_announce_depth
? " (and ready to announce)" : "");
} else {
log_debug(channel->log, "Waiting for funding confirmations");
memset(&scid, 0, sizeof(scid));
reached_announce_depth = false;
}
num_revocations = revocations_received(&channel->their_shachain.chain);
/* BOLT #2:
* - if `next_revocation_number` equals 0:
* - MUST set `your_last_per_commitment_secret` to all zeroes
* - otherwise:
* - MUST set `your_last_per_commitment_secret` to the last
* `per_commitment_secret` it received
*/
if (num_revocations == 0)
memset(&last_remote_per_commit_secret, 0,
sizeof(last_remote_per_commit_secret));
else if (!shachain_get_secret(&channel->their_shachain.chain,
num_revocations-1,
&last_remote_per_commit_secret)) {
channel_fail_permanent(channel,
"Could not get revocation secret %"PRIu64,
num_revocations-1);
return;
}
/* Warn once. */
if (ld->config.ignore_fee_limits)
log_debug(channel->log, "Ignoring fee limits!");
if(!wallet_remote_ann_sigs_load(tmpctx, channel->peer->ld->wallet, channel->dbid,
&remote_ann_node_sig, &remote_ann_bitcoin_sig)) {
channel_internal_error(channel,
"Could not load remote announcement signatures");
return;
}
initmsg = towire_channel_init(tmpctx,
chainparams,
&channel->funding_txid,
channel->funding_outnum,
channel->funding,
channel->minimum_depth,
&channel->our_config,
&channel->channel_info.their_config,
channel->channel_info.feerate_per_kw,
feerate_min(ld, NULL),
feerate_max(ld, NULL),
&channel->last_sig,
pps,
&channel->channel_info.remote_fundingkey,
&channel->channel_info.theirbase,
&channel->channel_info.remote_per_commit,
&channel->channel_info.old_remote_per_commit,
channel->funder,
channel->feerate_base,
channel->feerate_ppm,
channel->our_msat,
&channel->local_basepoints,
&channel->local_funding_pubkey,
&ld->id,
&channel->peer->id,
cfg->commit_time_ms,
cfg->cltv_expiry_delta,
channel->last_was_revoke,
channel->last_sent_commit,
channel->next_index[LOCAL],
channel->next_index[REMOTE],
num_revocations,
channel->next_htlc_id,
htlcs, htlc_states,
fulfilled_htlcs, fulfilled_sides,
failed_htlcs, failed_sides,
/* This is an approximation, but failing
* on restart is a corner case */
get_block_height(ld->topology),
channel->scid != NULL,
channel->remote_funding_locked,
&scid,
reconnected,
channel->state == CHANNELD_SHUTTING_DOWN,
channel->shutdown_scriptpubkey[REMOTE] != NULL,
channel->shutdown_scriptpubkey[LOCAL],
channel->channel_flags,
funding_signed,
reached_announce_depth,
&last_remote_per_commit_secret,
channel->peer->features,
channel->remote_upfront_shutdown_script,
remote_ann_node_sig,
remote_ann_bitcoin_sig,
/* Set at channel open, even if not
* negotiated now! */
channel->option_static_remotekey,
IFDEV(ld->dev_fast_gossip, false));
/* We don't expect a response: we are triggered by funding_depth_cb. */
subd_send_msg(channel->owner, take(initmsg));
/* On restart, feerate might not be what we expect: adjust now. */
if (channel->funder == LOCAL)
try_update_feerates(ld, channel);
}
bool channel_tell_depth(struct lightningd *ld,
struct channel *channel,
const struct bitcoin_txid *txid,
u32 depth)
{
const char *txidstr;
txidstr = type_to_string(tmpctx, struct bitcoin_txid, txid);
/* If not awaiting lockin/announce, it doesn't care any more */
if (channel->state != CHANNELD_AWAITING_LOCKIN
&& channel->state != CHANNELD_NORMAL) {
log_debug(channel->log,
"Funding tx %s confirmed, but peer in state %s",
txidstr, channel_state_name(channel));
return true;
}
if (!channel->owner) {
log_debug(channel->log,
"Funding tx %s confirmed, but peer disconnected",
txidstr);
return false;
}
subd_send_msg(channel->owner,
take(towire_channel_funding_depth(NULL, channel->scid,
depth)));
if (channel->remote_funding_locked
&& channel->state == CHANNELD_AWAITING_LOCKIN
&& depth >= channel->minimum_depth)
lockin_complete(channel);
return true;
}
/* Check if we are the fundee of this channel, the channel
* funding transaction is still not yet seen onchain, and
* it has been too long since the channel was first opened.
* If so, we should forget the channel. */
static bool
is_fundee_should_forget(struct lightningd *ld,
struct channel *channel,
u32 block_height)
{
u32 max_funding_unconfirmed = ld->max_funding_unconfirmed;
/* BOLT #2:
*
* A non-funding node (fundee):
* - SHOULD forget the channel if it does not see the
* correct funding transaction after a reasonable timeout.
*/
/* Only applies if we are fundee. */
if (channel->funder == LOCAL)
return false;
/* Does not apply if we already saw the funding tx. */
if (channel->scid)
return false;
/* Not even reached previous starting blocknum.
* (e.g. if --rescan option is used) */
if (block_height < channel->first_blocknum)
return false;
/* Timeout in blocks not yet reached. */
if (block_height - channel->first_blocknum < max_funding_unconfirmed)
return false;
/* Ah forget it! */
return true;
}
/* Notify all channels of new blocks. */
void channel_notify_new_block(struct lightningd *ld,
u32 block_height)
{
struct peer *peer;
struct channel *channel;
struct channel **to_forget = tal_arr(NULL, struct channel *, 0);
size_t i;
list_for_each (&ld->peers, peer, list) {
list_for_each (&peer->channels, channel, list)
if (is_fundee_should_forget(ld, channel, block_height)) {
tal_arr_expand(&to_forget, channel);
}
}
/* Need to forget in a separate loop, else the above
* nested loops may crash due to the last channel of
* a peer also deleting the peer, making the inner
* loop crash.
* list_for_each_safe does not work because it is not
* just the freeing of the channel that occurs, but the
* potential destruction of the peer that invalidates
* memory the inner loop is accessing. */
for (i = 0; i < tal_count(to_forget); ++i) {
channel = to_forget[i];
/* Report it first. */
log_unusual(channel->log,
"Forgetting channel: "
"It has been %"PRIu32" blocks without the "
"funding transaction %s getting deeply "
"confirmed. "
"We are fundee and can forget channel without "
"loss of funds.",
block_height - channel->first_blocknum,
type_to_string(tmpctx, struct bitcoin_txid,
&channel->funding_txid));
/* FIXME: Send an error packet for this case! */
/* And forget it. */
delete_channel(channel);
}
tal_free(to_forget);
}
static struct channel *find_channel_by_id(const struct peer *peer,
const struct channel_id *cid)
{
struct channel *c;
list_for_each(&peer->channels, c, list) {
struct channel_id this_cid;
derive_channel_id(&this_cid,
&c->funding_txid, c->funding_outnum);
if (channel_id_eq(&this_cid, cid))
return c;
}
return NULL;
}
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
/* Since this could vanish while we're checking with bitcoind, we need to save
* the details and re-lookup.
*
* channel_id *should* be unique, but it can be set by the counterparty, so
* we cannot rely on that! */
struct channel_to_cancel {
struct node_id peer;
struct channel_id cid;
};
static void process_check_funding_broadcast(struct bitcoind *bitcoind,
const struct bitcoin_tx_output *txout,
void *arg)
{
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
struct channel_to_cancel *cc = arg;
struct peer *peer;
struct channel *cancel;
/* Peer could have errored out while we were waiting */
peer = peer_by_id(bitcoind->ld, &cc->peer);
if (!peer)
return;
cancel = find_channel_by_id(peer, &cc->cid);
if (!cancel)
return;
if (txout != NULL) {
for (size_t i = 0; i < tal_count(cancel->forgets); i++)
was_pending(command_fail(cancel->forgets[i], LIGHTNINGD,
"The funding transaction has been broadcast, "
"please consider `close` or `dev-fail`! "));
tal_free(cancel->forgets);
cancel->forgets = tal_arr(cancel, struct command *, 0);
return;
}
const char *error_reason = "Cancel channel by our RPC "
"command before funding "
"transaction broadcast.";
/* Set error so we don't try to reconnect. */
cancel->error = towire_errorfmt(cancel, NULL, "%s", error_reason);
subd_send_msg(cancel->owner,
take(towire_channel_send_error(NULL, error_reason)));
}
struct command_result *cancel_channel_before_broadcast(struct command *cmd,
const char *buffer,
struct peer *peer,
const jsmntok_t *cidtok)
{
struct channel *cancel_channel;
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
struct channel_to_cancel *cc = tal(cmd, struct channel_to_cancel);
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
cc->peer = peer->id;
if (!cidtok) {
struct channel *channel;
cancel_channel = NULL;
list_for_each(&peer->channels, channel, list) {
if (cancel_channel) {
return command_fail(cmd, LIGHTNINGD,
"Multiple channels:"
" please specify channel_id");
}
cancel_channel = channel;
}
if (!cancel_channel)
return command_fail(cmd, LIGHTNINGD,
"No channels matching that peer_id");
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
derive_channel_id(&cc->cid,
&cancel_channel->funding_txid,
cancel_channel->funding_outnum);
} else {
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
if (!json_tok_channel_id(buffer, cidtok, &cc->cid))
return command_fail(cmd, JSONRPC2_INVALID_PARAMS,
"Invalid channel_id parameter.");
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
cancel_channel = find_channel_by_id(peer, &cc->cid);
if (!cancel_channel)
return command_fail(cmd, LIGHTNINGD,
"Channel ID not found: '%.*s'",
cidtok->end - cidtok->start,
buffer + cidtok->start);
}
/* Check if we broadcast the transaction. (We store the transaction type into DB
* before broadcast). */
enum wallet_tx_type type;
if(wallet_transaction_type(cmd->ld->wallet,
&cancel_channel->funding_txid,
&type))
return command_fail(cmd, LIGHTNINGD,
"Has the funding transaction been broadcast? "
"Please use `close` or `dev-fail` instead.");
if (channel_has_htlc_out(cancel_channel) ||
channel_has_htlc_in(cancel_channel)) {
return command_fail(cmd, LIGHTNINGD,
"This channel has HTLCs attached and it is "
"not safe to cancel. Has the funding transaction "
"been broadcast? Please use `close` or `dev-fail` "
"instead.");
}
tal_arr_expand(&cancel_channel->forgets, cmd);
/* Check if the transaction is onchain. */
/* Note: The above check and this check can't completely ensure that
* the funding transaction isn't broadcast. We can't know if the funding
* is broadcast by external wallet and the transaction hasn't been onchain. */
bitcoind_gettxout(cmd->ld->topology->bitcoind,
&cancel_channel->funding_txid,
cancel_channel->funding_outnum,
process_check_funding_broadcast,
lightningd: fix crash in cancel_channel This path is not atomic, so we can't assume channel stays around. It could be the peer closes, it could be we get a parallel fund_channel_cancel. test_funding_cancel_race caused this crash: FATAL SIGNAL 6 (version 5f0a18e) backtrace: common/daemon.c:45 (send_backtrace) 0x55c7c373a429 backtrace: common/daemon.c:53 (crashdump) 0x55c7c373a479 backtrace: (null):0 ((null)) 0x7f88ee6ddf5f backtrace: (null):0 ((null)) 0x7f88ee6dded7 backtrace: (null):0 ((null)) 0x7f88ee6bf534 backtrace: ccan/ccan/tal/tal.c:93 (call_error) 0x55c7c379427c backtrace: ccan/ccan/tal/tal.c:165 (check_bounds) 0x55c7c3794444 backtrace: ccan/ccan/tal/tal.c:174 (to_tal_hdr) 0x55c7c3794483 backtrace: ccan/ccan/tal/tal.c:186 (to_tal_hdr_or_null) 0x55c7c3794504 backtrace: ccan/ccan/tal/tal.c:421 (tal_alloc_) 0x55c7c3794c10 backtrace: ccan/ccan/tal/tal.c:466 (tal_alloc_arr_) 0x55c7c3794ded backtrace: ccan/ccan/tal/str/str.c:91 (tal_vfmt_) 0x55c7c3793560 backtrace: common/wire_error.c:22 (towire_errorfmtv) 0x55c7c3747f7b backtrace: common/wire_error.c:39 (towire_errorfmt) 0x55c7c37480a1 backtrace: lightningd/channel_control.c:635 (process_check_funding_broadcast) 0x55c7c37015bb backtrace: lightningd/bitcoind.c:558 (process_gettxout) 0x55c7c36f8e75 backtrace: lightningd/bitcoind.c:227 (bcli_finished) 0x55c7c36f8090 backtrace: ccan/ccan/io/poll.c:244 (destroy_conn) 0x55c7c37869fe backtrace: ccan/ccan/io/poll.c:250 (destroy_conn_close_fd) 0x55c7c3786a1e backtrace: ccan/ccan/tal/tal.c:235 (notify) 0x55c7c3794629 backtrace: ccan/ccan/tal/tal.c:397 (del_tree) 0x55c7c3794b18 backtrace: ccan/ccan/tal/tal.c:481 (tal_free) 0x55c7c3794ea4 backtrace: ccan/ccan/io/io.c:450 (io_close) 0x55c7c378521d backtrace: ccan/ccan/io/poll.c:449 (io_loop) 0x55c7c3787139 backtrace: lightningd/io_loop_with_timers.c:24 (io_loop_with_timers) 0x55c7c370b26d backtrace: lightningd/lightningd.c:837 (main) 0x55c7c3711661 backtrace: (null):0 ((null)) 0x7f88ee6c0b6a backtrace: (null):0 ((null)) 0x55c7c36f70b9 backtrace: (null):0 ((null)) 0xffffffffffffffff Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2019-09-19 07:44:24 +01:00
notleak(cc));
return command_still_pending(cmd);
}
#if DEVELOPER
static struct command_result *json_dev_feerate(struct command *cmd,
const char *buffer,
const jsmntok_t *obj UNNEEDED,
const jsmntok_t *params)
{
u32 *feerate;
struct node_id *id;
struct peer *peer;
struct json_stream *response;
struct channel *channel;
const u8 *msg;
if (!param(cmd, buffer, params,
p_req("id", param_node_id, &id),
p_req("feerate", param_number, &feerate),
NULL))
return command_param_failed();
peer = peer_by_id(cmd->ld, id);
if (!peer)
return command_fail(cmd, LIGHTNINGD, "Peer not connected");
channel = peer_active_channel(peer);
if (!channel || !channel->owner || channel->state != CHANNELD_NORMAL)
return command_fail(cmd, LIGHTNINGD, "Peer bad state");
msg = towire_channel_feerates(NULL, *feerate,
feerate_min(cmd->ld, NULL),
feerate_max(cmd->ld, NULL));
subd_send_msg(channel->owner, take(msg));
response = json_stream_success(cmd);
json_add_node_id(response, "id", id);
json_add_u32(response, "feerate", *feerate);
return command_success(cmd, response);
}
static const struct json_command dev_feerate_command = {
"dev-feerate",
"developer",
json_dev_feerate,
"Set feerate for {id} to {feerate}"
};
AUTODATA(json_command, &dev_feerate_command);
#endif /* DEVELOPER */