Instead of a boutique message, use a "real" channel_announcement for
private channels (with fake sigs and pubkeys). This makes it far
easier for gossmap to handle local channels.
Backwards compatible update, since we update old stores.
We also fix devtools/dump-gossipstore to know about the tombstone markers.
Since we increment our channel_announce count for local channels now,
the stats in the tests changed too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
There's a few structs/wire calls that only exist under experimental features.
These were in a common file that was shared/used a bunch of places but
this causes problems. Here we move one of the problematic methods back
into `openingd`, as it's only used locally and then isolate the
references to the `witness_stack` in a new `common/psbt_internal` file.
This lets us remove the iff EXP_FEATURES inclusion switches in most of
the Makefiles.
1. Rename memleak_enter_allocations to memleak_find_allocations.
2. Unify scanning for pointers into memleak_remove_region / memleak_remove_pointer.
3. Document the functions.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
dual funding needs the max-witness-len and utxo fields set for every
input. we should add them when we create a 'fundpsbt', so that every
psbt that c-lightning generates is dual-funding ready
There's a lot of it, and it means we can't `make check-source` on
these files.
Also bring bolt quotes up-to-date.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This avoids overwriting the ones in git, and generally makes things neater.
We have convenience headers wire/peer_wire.h and wire/onion_wire.h to
avoid most #ifdefs: simply include those.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're going to make experimental versions of these completely separate files.
Also remove the dependency on the Makefile itself: it simply causes
unnecessary churn. We can always force-rebuild when we change a rule.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We create ALL_PROGRAMS, ALL_TEST_PROGRAMS, ALL_C_SOURCES and
ALL_C_HEADERS. Then the toplevel Makefile knows which are
autogenerated (by wildcard), so it can have all the rules to clean
them or check the source as necessary.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We are logging way too much from gossipd, causing noisy logs. This PR reduces
logs for incoming messages to those that actually caused a change in our
internal state (duplicate and old messages are just dropped silently now).
Changelog-Changed: gossipd: The `gossipd` is now a lot quieter, and will log only when a message changed our network topology.
Note that other directories were explicitly depending on the generated
file, instead of relying on their (already existing) dependency on
$(LIGHTNINGD_HSM_CLIENT_OBJS), so we remove that.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
See https://github.com/lightningnetwork/lightning-rfc/pull/767
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: Protocol: channels now pruned after two weeks unless both peers refresh it (see lightning-rfc#767)
It's not all that rare to do these operations, and requiring annotations
for it is a little painful.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
There were no channel updates in my log; because sendonion doesn't know the
actual node_ids or channel_ids, we can't tell gossipd what node/channel it was
so it can no longer remove them on PERM errors.
However, we can tell it the error message so it can apply the update.
Fixes: #3877
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The main change here is that the previously-optional open/accept
fields and reestablish fields are now compulsory (everyone was
including them anyway). In fact, the open/accept is a TLV
because it was actually the same format.
For more details, see lightning-rfc/f068dd0d8dfa5ae75feedd99f269e23be4777381
Changelog-Removed: protocol: support for optioned form of reestablish messages now compulsory.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Since we now over-write the wally malloc/free functions, we need to do
so for tests as well. Here we pull up all of the common setup/teardown
logic into a separate place, and update the tests that use libwally to
use the new common_setup core
Changelog-None
This saves us keeping it in memory (so far, no channels have features), but
lets us optimize that case so we don't need to hit the disk for most of the
channels in listchannels.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
When we have only a single member in a TLV (e.g. an optional u64),
wrapping it in a struct is awkward. This changes it to directly
access those fields.
This is not only more elegant (60 fewer lines), it would also be
more cache friendly. That's right: cache hot singles!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This will be used when we want to specify these in a route. But for now, they
only alter gossipd, which always sets them to NULL.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Use `LC_ALL=C sort` instead of `sort` so that mocks get sorted in
the same way on all developers' environments.
Re-record the result of `make update-mocks`.
Changelog-None
It's almost always "their_features" and "our_features" respectively, so
make those names clear.
Suggested-by: @cdecker
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Turns out that unnecessary: all callers can access the feature_set,
so make it much more like a normal primitive.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is to prepare for dynamic features, including making plugins first
class citizens at setting them.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
```
clang10 -DBINTOPKGLIBEXECDIR="\"../libexec/c-lightning\"" -Wall -Wundef -Wmissing-prototypes -Wmissing-declarations -Wstrict-prototypes -Wold-style-definition -Werror -std=gnu11 -g -fstack-protector -I ccan -I external/libwally-core/include/ -I external/libwally-core/src/secp256k1/include/ -I external/jsmn/ -I external/libbacktrace/ -I external/libbacktrace-build -I . -I/usr/local/include -DCCAN_TAKE_DEBUG=1 -DCCAN_TAL_DEBUG=1 -DCCAN_JSON_OUT_DEBUG=1 -DSHACHAIN_BITS=48 -DJSMN_PARENT_LINKS -DBUILD_ELEMENTS=1 -DBINTOPKGLIBEXECDIR="\"../libexec/c-lightning\"" -c -o gossipd/routing.o gossipd/routing.c
gossipd/routing.c:651:10: error: implicit conversion from 'unsigned long' to 'double' changes value
from 18446744073709551615 to 18446744073709551616 [-Werror,-Wimplicit-int-float-conversion]
if (r > UINT64_MAX)
~ ^~~~~~~~~~
```
It is ok to change the values because they are approximate anyway. Thus,
explicitly typecast to `double` to silence the warning without changing
behavior.
Changelog-None
Even without optimization, it's faster to walk all the channels than
ping another daemon and wait for the response.
Changelog-Changed: Forwarding messages is now much faster (less inter-daemon traffic)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The idea is that gossipd can give us the cupdate we need for an error, and
we wire things up so that we ask for it (async) just before we send the
error to the subdaemon.
I tried many other things, but they were all too high-risk.
1. We need to ask gossipd every time, since it produces these lazily
(in particular, it doesn't actually generate an offline update unless
the channel is used).
2. We can't do async calls in random places, since we'll end up with
an HTLC in limbo. What if another path tries to fail it at the same time?
3. This allows us to use a temporary_node_failure error, and upgrade it
when gossipd replies. This doesn't change any existing assumptions.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is a common thing to do, so create a macro.
Unfortunately, it still needs the type arg, because the paramter may
be const, and the return cannot be, and C doesn't have a general
"(-const)" cast.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
common should not include specific per-daemon files. Turns out this
caused a lot of indirect includes to be exposed.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Before this patch we used to send `double`s over the wire by just
copying them. This is not portable because the internal represenation
of a `double` is implementation specific.
Instead of this, multiply any floating-point numbers that come from
the outside (e.g. JSONs) by 1 million and round them to integers when
handling them.
* Introduce a new param_millionths() that expects a floating-point
number and returns it multipled by 1000000 as an integer.
* Replace param_double() and param_percent() with param_millionths()
* Previously the riskfactor would be allowed to be negative, which must
have been unintentional. This patch changes that to require a
non-negative number.
Changelog-None
I hadn't realized that lightningd asks gossipd every time we forward
a payment. But I'm going to abuse it here to get the latest channel_update,
otherwise (as lightningd takes over error message generation) lightningd
needs to do an async request at various painful points.
So have gossipd tell us the lastest update (stripped so compatible with
the strange in-onion-error format).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now that we have json_stream in common/, we can move all the related
helpers from lightningd/json to common/json. This way everyone can
benefit of them (including libplugin, the plugins themselves,
potentially lightning-cli), not lightningd alone!
Note that the Makefile of the common/test/ had to be modified, because
the new helpers make use of common/wireaddr... Which turns out to
\#include <lightingd/lightningd.h> ! So we couldnt just include the .c
and add mocks if we redefined some structs (hello run-param).
This makes it clear we're dealing with a message which is a wrapped error
reply (needing unwrap_onionreply), not an already-wrapped one.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This lets us do more flexible filtering in the next patch. But it also
keeps some weird logic out of gossipd.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We've been sending them errors for invalid replies; instead, this works
around it.
Changelog-Added: Workaround LND's reply_channel_range issues instead of sending error.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is ignored in subdaemons which are per-peer, but very useful for
multi-peer daemons like connectd and gossipd.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It really, really doesn't matter. But we were dramatically reducing
our view of the network:
In my gossip_store (mainnet):
channel_announcement: 30349
channel_update: 55119
node_announcment: 1783
Changelog-Fixed: No longer discard most node_announcements (fixes#3194)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The flat feature PR changes the rules so these are OK to propagate.
That makes sense: the unsupported features means there's something
unsupported about the *node* or *channel*, not the msg itself
(for that we'd use a different message type).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This prevents a gratuitous lookup of we get a late channel_announce,
but even better, it suppresses the "bad gossip" messages in case of
a late channel_update, which have plagued Travis (especially since we
got aggressive in pushing our own updates).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is a better fix than doing it manually, which turned out
to do it in the wrong order (node_announcement followed by
channel_announcement) anyway.
Should fix many "Bad gossip" messages.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I was wondering why TAGS was missing some functions, and finally
tracked it down: PRINTF_FMT() confuses etags if it's at the start
of a function, and it ignores the rest of the file.
So we put PRINTF_FMT at the end, but that doesn't work for
*definitions*, only *declarations*. So we remove it from definitions
and add gratuitous declarations in the few static places.1
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Asking for the last few blocks was logical, but my node is missing
most gossip in practice.
For the moment, simply ask a peer for every channel it knows, once
we're started up.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
When probing, no point probing for before lightning became cool. Current
logic means we often probe below block 500,000, and think things are OK
because there are no short_channel_ids.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We always ended up sending an empty query when we had stale scids!
And it turns out we consider such a query invalid:
Bad query_short_channel_ids query_flags 010506226e46111a0b59caaf126043eb5bbf28c34f3a5e332a1fc7b2b73cf188910f000100010100
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We only chose 3 peers to gossip with us (down from 8 last release).
There's no justification for this number, or reason to believe that
it is sufficient to keep us in sync.
Be more conservative for now; we can always decrease it later once
we have more data.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
polling the last 32 is fairly useless in practice, since they tend to
be recent nodes; it won't detect long-forgotten ones.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If we get a channel_update while we're still verifying the channel_announcement
we didn't set the peer pointer, so it didn't get credit. As a result, the
seeker tended to think we were done gossiping sooner than we were.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I had a report of a 0.7.2 user whose node hadn't appeared on 1ml. Their
node_announcement wasn't visible to my node, either.
I suspect this is a consequence of recent version reducing the amount of
gossip they send, as well as large nodes increasingly turning off gossip
altogether from some peers (as we do). We should ignore timestamp filters
for our own channels: the easiest way to do this is to push them out
directly from gossipd (other messages are sent via the store).
We change channeld to wrap the local channel_announcements: previously
we just handed it to gossipd as for any other gossip message we received
from our peer. Now gossipd knows to push it out, as it's local.
This interferes with the logic in tests/test_misc.py::test_htlc_send_timeout
which expects the node_announcement message last, so we generalize
that too.
[ Thanks to @trueptolmy for bugfix! ]
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is mainly an internal-only change, especially since we don't
offer any globalfeatures.
However, LND (as of next release) will offer global features, and also
expect option_static_remotekey to be a *global* feature. So we send
our (merged) feature bitset as both global and local in init, and fold
those bitsets together when we get an init msg.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I thought LND had a bug, but turns out it doesn't like out-of-order
short_channel_ids: in fact, the spec says they have to be in order!
This means we use uintmap instead of a htable for unknown_scids and
stale_scids so they're nicely ordered.
But our nodes-missing-announcements probe is harder since they can
also contain duplicates: we switch that to iterate through channels
rather than nodes.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We weren't supposed to do any gossiping until we were synced (and thus
knew blockheight), but our seeker_check() didn't wait for it! Move the
waiting all into seeker.c, so it can handle it all consistently.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
On testing, I found a node which would hang up every time we asked it
for query_short_channel_ids (despite it offering features 0x81, meaning
it should handle this message).
Then it would reconnect, and we'd choose it again as our
PROBING_NANNOUNCES peer!
Instead, leave finding another peer to the once-a-minute
seeker_check() function.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
By combining set_state() with selected_peer() we can give a single log
line describing what we're asking for, from whom.
We also add more verbosity to a few key areas, such as gossip rotation
and when gossipd tells a peer to send an error. And move a comment which
was above the wrong function (due to rebase?).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It usually means we're missing something, but there's no way to ask what.
Simply start a broad scid probe.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This should give more reliable results, though it risks us getting
suckered into always consulting the same peer.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We assume that the time for gossip propagation is < 10 minutes, so by
going back that far from last gossip we won't miss anything,
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's simple: if we wouldn't accept the timestamp we see, don't put
the channel in the stale_scid_map.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Just try to choose another. Under Travis, this causes many failures due
to slowness (they only get 10 seconds in -dev mode).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We eliminate the "need peer" states and instead check if the
random_peer_softref has been cleared.
We can also unify our restart handlers for all these cases; even the
probe_scids case, by giving gossip credit for the scids as they come
in (at a discount, since scids are 8 bytes vs the ~200 bytes for
normal gossip messages).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Build up a map of short_channel_ids which we have old info for (only
if peer supports gossip_query_ex).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This asks peers to append the timestamps or checksums: if it has
gossip_query_ex support, it will, otherwise it's ignored.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If the peer supports `gossip_query_ex` we can use query_flags to simply
request the node_announcements when probing for nodes, rather than
getting everything. If a peer doesn't support `gossip_query_ex` then
it's harmless to add it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We pick some nodes which don't seem to have node_announcements and we
ask a channel associated with them. Again, if this reveals more
node_announcements, we probe for twice as many next time.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If we have any unknown short_channel_ids, we ask a random peer for
those channels. Once it responds, we probe again for a small random
range in case more are missing, again enlarging if we find some.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Instead of a linear array which is fairly inefficient if it turns out
we know nothing at all.
We remove the gossip_missing() call by changing the api to
remove_unknown_scid() to include a flag as to whether the scid turned
out to be real or not.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Once we've finished streaming gossip from the first peer, we ask a
random peer (maybe the same one) for all short_channel_ids in the last
6 blocks from the latest channel we know about.
If this reveals new channels we didn't know about, we expand the probe
by a factor of 2 each time.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The seeker starts by asking a peer (the first peer!) for all gossip
since a minute before the modified time of the gossip store.
This algorithm is enhanced in successive patches.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Since we have to validate, there can be a delay (and peer might
vanish) between receiving the gossip and actually confirming it, hence
the use of softref.
We will use this information to check that the peers are making progress
as we start asking them for specific information.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is the modified-time of the file. We have to store it internally
since we overwrite the gossip file with compaction on startup.
This means the "are we behind on gossip?" heuristic is no longer inside
gossip_store.c, which is cleaner.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We now have a pointer to chainparams, that fails valgrind if we do anything
chain-specific before setting it.
Suggested-by: Rusty Russell <@rustyrussell>
We do this by keeping a current and an old map, and moving the current to old
every hour or 10,000 entries.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This encoding scheme is no longer just used for short_channel_ids, so make
the names more generic.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I was seeing some accidental pruning under load / Travis, and in
particular we stopped accepting channel_updates because they were 103
seconds old. But making it too long makes the prune test untenable,
so restore a separate flag that this test can use.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The only real change is dump_gossip() used to call
maybe_create_next_scid_reply(), but now I've simply renamed
that to maybe_send_query_responses() and we call it directly.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now we queue them, we should place a limit. It's not the worst thing in
the world if we discard them (we'll catch up eventually), but we should
try not to in case we're just a bit behind.
Our behaviour here is also O(n^2) so we don't want a massive queue
anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The first one means we don't discard channels just because we're not
synced, and the second is implied by the spec: don't accept
channel_announcement if the channel isn't 6 deep. Since LND defers in
such cases, we do too (unless it's newer than the current block, in
which case we simply discard). Otherwise there's a risk that a slow
node might discard valid gossip.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This will let gossipd be more intelligent about gossiping before we're
synced, and also it might know how far behind we are.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Happened under Travis with --dev-fast-gossip (90 second prune time), but can
happen anyway if gossip is almost 2 weeks old when we receive it:
2019-09-20T19:16:51.367Z DEBUG lightning_gossipd(20972): Received node_announcement for node 022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59
2019-09-20T19:16:51.376Z DEBUG lightning_gossipd(20972): Ignoring node_announcement timestamp 1569006918 for 022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59
2019-09-20T19:16:51.669Z **BROKEN** lightning_gossipd(20972): pending node_announcement 01013094af771d60f4de69bb39ce045e4edf4a06fe6c80078dfa4fab58ab5617d6ad4fa34b6d3437380db0a8293cea348bbc77f714ef71fcd8515bfc82336667441f00005d852546022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59022d2253494c454e544152544953542d633961313734610000000000000000000000000000 malformed? (version c9a174a)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's generally clearer to have simple hardcoded numbers with an
#if DEVELOPER around it, than apparent variables which aren't, really.
Interestingly, our pruning test was always kinda broken: we have to pass
two cycles, since l2 will refresh the channel once to avoid pruning.
Do the more obvious thing, and cut the network in half and check that
l1 and l3 time out.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If you send a message which simply changes timestamp and signature, we
drop it. You shouldn't be doing that, and the door to ignoring them
was opened by by option_gossip_query_ex, which would allow clients to
ignore updates with the same checksum.
This is more aggressive at reducing spam messages, but we allow refreshes
(to be conservative, we allow them even when 1/2 of the way through the
refresh period).
I dropped the now-unnecessary sleep from test_gossip_pruning, too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Make update_local_channel use a timer if it's too soon to make another
update.
1. Implement cupdate_different() which compares two updates.
2. make update_local_channel() take a single arg for timer usage.
3. Set timestamp of non-disable update back 5 minutes, so we can
always generate a disable update if we need to.
4. Make update_local_channel() itself do the "unchanged update" suppression.
gossipd: clean up local channel updates.
5. Keep pointer to the current timer so we override any old updates with
a new one, to avoid a race.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Normally we'd put a pointer into struct half_chan for local
information, but it would be NULL on 99.99% of nodes. Instead, keep a
separate hash table.
This immediately subsumes the previous "map of local-disabled
channels", and will be enhanced further.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Write helpers to split it into non-timestamp, non-signature parts,
and simply compare those. We extract a helper to do channel_update, too.
This is more generic than our previous approach, and simpler.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
For memory-usage reasons, struct chan doesn't use a tal destructor, in
favor of us calling free_chan in the right places.
In DEVELOPER mode, we should check that is the case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We've been slack, but it's going to be important for testing
ratelimiting. And it currently has a minor memory leak.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Rather than reaching into data structures, let them register their own
callbacks. This avoids us having to expose "memleak_remove_xxx"
functions, and call them manually.
Under the hood, this is done by having a specially-named tal child of
the thing we want to assist, containing the callback.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
`make update-mocks` is usually run in DEVELOPER mode, but then it includes
definitions for functions which aren't declared in non-DEVELOPER mode.
We hacked this in a few places, but it's fragile, and worst, now we
have EXPERIMENTAL_FEATURES as well, it's complex.
Instead, declare developer-only functions (but don't define them).
This is a bit more awkward if you accidentally use one in
non-DEVELOPER code (link error rather than compile error), but makes
autogenerating test mocks much easier.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fortunately, again, only happens with EXPERIMENTAL_FEATURES.
If the query causes us not to actually send anything, we won't
get called again. This can validly happen if they only asked for
the node_announcements, for example.
(Found by protocol tests).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Our "are we finished?" logic was wrong: it tested if there are no more
node_announcements, but it's possible that there were no node_announcements
for either end of the channel whose information we sent.
This is actually quite unusual on the real network: looking at mainnet
statis from last May, 4301 of 4337 nodes have node_announcements.
However, with query flags it's much more likely, since they might not
ask for node announcements at all.
(Found by gossip protocol tests)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
These both allow us to reproduce the test vectors in the next patch. But
using Z_DEFAULT_COMPRESSION is a reasonable idea anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Make the TLV element a simple array. This is a bit neater, in fact, and
makes the test vectors in that 557 PR work.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In fact, we always generate them, we only send them if asked. And we set
the flags to 0 if not --enable-experimental-features, so we never send in
that case.
Generating checksums involves pulling the channel_update from the
gossip_store, which is suboptimal: there's a FIXME to store the
checksum in memory.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're about to use the for gossip extended info too, which *don't* put
the encoding byte at the beginning of the data stream. So this removes
some "scids" from function names and separates out the "prepend a byte"
case from the "external encoding_type" case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
These indicate what fields we are to return. If there's now TLV, or we
haven't got --enable-experimental-features, it's set to all 1s so behaviour
is unchanged.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We ignored this before, which meant that the DEVELOPER-mode check that we
delete the correct record didn't check that it wasn't already deleted.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We always know the length, so we don't need it. It causes much extra work
when we want to delete a record, which I suspect may cause issues amongst
some users who've been seeing gossip_store corruption.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We added a random channel to the list, but we can just free it immediately
(since traversal of a uintmap isn't altered by deletion).
This was introduced in d1f43d993a where we explicitly call free_chan
rather than relying on destructors.
Fixes: #2837
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
==1503== Use of uninitialised value of size 8
==1503== at 0x566786B: _itoa_word (_itoa.c:179)
==1503== by 0x566AF0D: vfprintf (vfprintf.c:1642)
==1503== by 0x569790F: vsnprintf (vsnprintf.c:114)
==1503== by 0x156CCB: do_vfmt (str.c:66)
==1503== by 0x156DB1: tal_vfmt_ (str.c:92)
==1503== by 0x1289CD: status_vfmt (status.c:141)
==1503== by 0x128AAC: status_fmt (status.c:151)
==1503== by 0x118E05: route_prune (routing.c:2495)
==1503== by 0x11DE2D: gossip_refresh_network (gossipd.c:1997)
==1503== by 0x1292B8: timer_expired (timeout.c:39)
==1503== by 0x12088C: main (gossipd.c:3075)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
updates the bolt version to 6639cef095a2ecc7b8f0c48c6e7f2f906fbfbc58.
this requires us to use the new bolt parser at generate-bolt.py
and updates to all of the type specifications (ie. from u8 -> byte)
Rewriting the gossip_store is much more trivial when we don't have
any pointers into it, so add some simple offline compaction code
and disable the automatic compaction code.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The crashes in #2750 are mostly caused by us trying to partially truncate
the store. The simplest fix for release is to discard the whole thing if
we detect a problem.
This is a workaround: it'd be far nicer to try to recover.
Fixes: #2750
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We hit the timestamp assert on #2750; it shouldn't happen, but crashing
doesn't leave much information.
Reported-by: @m-schmook
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If something went wrong and there was an old one, we were
appending to it!
Reported-by: @SimonVrouwe
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We might have channel_announcements which have no channel_update: normally
these don't get written into the store until there is one, but if the
store was truncated it can happen. We then get upset on compaction, since
we don't have an in-memory representation of the channel_announcement.
Similarly, we leave the node_announcement pending until after that
channel_announcement, leading to a similar case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We can't continue, since we've moved the indexes. We'll just crash
anyway, as seen from bugs #2742 and #2743.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We catch node_announcements for nodes where we haven't finished
analyzing the channel_announcement yet (either because we're still
checking UTXO, or in this case, because we're waiting for a channel_update).
But we reference count the pending_node_announce, so if we have
multiple channels pending, we might try to insert it twice. Clear it
so this doesn't happen.
There's a second bug where we continue to catch node_announcements
until *all* the channel_announcements are no longer pending; this is fixed
by removing it from the map.
Fixes: #2735
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We seek a certain number of peers at each level of gossip; 3 "flood"
if we're missing gossip, 2 at 24 hours past to catch recent gossip, and
8 with current gossip. The rest are given a filter which causes them
not to gossip to us at all.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The first sign that we're missing gossip is that we get a channel_update
for an unknown channel. The peer might be wrong (or lying), but if it turns
out to be a real channel, we were definitely missing something.
This patch does two things: queries when we get an unknown channel_update,
and then notes that a channel_announcement was from such an update when
it's finally processed.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In particular, we'll need to know the short_channel_id if a
channel_update is unknown (implies we're missing a channel), and whether
processing a pending channel_announcement was successful (implies that
the channel was real).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Up until now we only generated these in dev mode for testing. Hoist
into common code, turn counter into a flag (we're only allowed one!)
and note if query is internal or not.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
I decided to try a faster implementation, only to find our crc32c was
not correct! Ouch.
I removed the crc32c functions from ccan/crc, and added a new crc32c
module which has the Mark Adler x86-64-optimized variants.
We bump gossip_store version again, since csums have changed.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means there's now a semantic difference between the default `fromid`
and setting `fromid` explicitly to our own node_id. In the default case,
it means we don't charge ourselves fees on the route.
This means we can spend the full channel balance.
We still want to consider the pricing of local channels, however:
there's a *reason* to discount one over another, and that is to bias
things. So we add the first-hop fee to the *risk* value instead.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(Or, if we crashed before we got to write out the channel_update).
It's a corner case, but one reported by @darosior and reproduced
on my test node (both with bad gossip_store due to previous iterations
of this patchset!).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Triggered by a previous variant of this PR, but a goo1d idea to simply
discard the store in general when we get a duplicate entry.
We crash trying to delete old ones, which means writing to the store.
But they should have already been deleted.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This clarifies things a fair bit: we simply add and remove from the
gossip_store directly.
Before this series: (--disable-developer, -Og)
store_load_msec:20669-20902(20822.2+/-82)
vsz_kb:439704-439712(439706+/-3.2)
listnodes_sec:0.890000-1.000000(0.92+/-0.04)
listchannels_sec:11.960000-13.380000(12.576+/-0.49)
routing_sec:3.070000-5.970000(4.814+/-1.2)
peer_write_all_sec:28.490000-30.580000(29.532+/-0.78)
After: (--disable-developer, -Og)
store_load_msec:19722-20124(19921.6+/-1.4e+02)
vsz_kb:288320
listnodes_sec:0.860000-0.980000(0.912+/-0.056)
listchannels_sec:10.790000-12.260000(11.65+/-0.5)
routing_sec:2.540000-4.950000(4.262+/-0.88)
peer_write_all_sec:17.570000-19.500000(18.048+/-0.73)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We have a problem: if we get halfway through writing the compacted store
and run out of disk space, we've already changed half the indexes.
This changes it so we do nothing until writing is finished: then we
iterate through and update indexes. It also weans us off broadcast
ordering, which we can now eliminated.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We didn't count some records before, so we could compare the two counters.
This is much simpler, and avoids reliance on bs.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means we intercept the peer's gossip_timestamp_filter request
in the per-peer subdaemon itself. The rest of the semantics are fairly
simple however.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
(We don't increment the gossip_store version, since there are only a
few commits since the last time we did this).
This lets the reader simply filter messages; this is especially nice since
the channel_announcement timestamp is *derived*, not in the actual message.
This also creates a 'struct gossip_hdr' which makes the code a bit
clearer.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Keeping the uintmap ordering all the broadcastable messages is expensive:
130MB for the million-channels project. But now we delete obsolete entries
from the store, we can have the per-peer daemons simply read that sequentially
and stream the gossip itself.
This is the most primitive version, where all gossip is streamed;
successive patches will bring back proper handling of timestamp filtering
and initial_routing_sync.
We add a gossip_state field to track what's happening with our gossip
streaming: it's initialized in gossipd, and currently always set, but
once we handle timestamps the per-peer daemon may do it when the first
filter is sent.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
There's a corner case where otherwise a reader could see the header and
not the body of a message. It could handle that in various ways,
but simplest (and most efficient) is to avoid it happening.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
They already send *us* gossip messages, so they have to be distinct anyway.
Why make us both do extra work?
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We use the high bit of the length field: this way we can still check
that the checksums are valid on deleted fields.
Once this is done, serially reading the gossip_store file will result
in a complete, ordered, minimal gossip broadcast. Also, the horrible
corner case where we might try to delete things from the store during
load time is completely gone: we only load non-deleted things.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're about to bump version again, and the code to upgrade it was
quite hairy (and buggy!). It's not worthwhile for such a
poorly-tested path: I will just add code to limit how much incoming
gossip we get to avoid flooding when we upgrade, however.
I also use a modern gossip_store version in our test_gossip_store_load
test, instead of relying on the upgrade path.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
They're really gossipd-internal, and we don't want per-peer daemons
to confuse them with normal updates.
I don't bump the gossip_store version; that's coming with another update
anyway.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Now we handle node_announcements properly, we have a failure case where we
try to move them when a channel is deleted while loading the store.
We're going to remove this soon, in favor of in-place delete, so
workaround this for now to avoid an assert() when we try to write to
the store while loading.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
When we first receive a channel_update, we write both the
channel_announcement and that channel_update to the store: we need
that first update so we can set the channel_announcement timestamp.
However, the channel_update can be replaced later. This means we can
have a channel_announcement, a node_update which relies on it, then
the channel_update later.
So move the "this applies to a pending announcement" check lower, where
gossip_store can use it too. Has a nice side-effect of avoiding
one lookup of the node id.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It turns out that we don't look at type when we return 0, but gcc isn't
quite smart enough for that. Initializing to -1 is good practice anyway
for the failure path.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fix a path where tal_free is called on an uninitialized variable
If the first `goto bad_total` executes, then that path has
uninitialized `short_route` but bad_total passes through to `out`
whose first call is tal_free(short_route).
This was noticed by a maybe-uninitialized heuristic on gcc 7.4.0:
gossipd/routing.c: In function ‘find_shorter_route’:
gossipd/routing.c:1096:2: error: ‘short_route’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]
tal_free(short_route);
Reported-by: @ZmnSCPxj <https://github.com/ElementsProject/lightning/pull/2674#issuecomment-495617253>
Signed-off-by: William Casarin <jb55@jb55.com>
Each destructor2 costs 40 bytes, and struct chan is only 120 bytes. So
this drops our memory usage quite a bit:
MCP bench results change:
-vsz_kb:580004-580016(580006+/-4.8)
+vsz_kb:533148
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This has two effects: most importantly, it avoids the problem where
lightningd creates a 800MB JSON blob in response to listchannels,
which causes OOM on the Raspberry Pi (our previous max allocation was
832MB). This is because lightning-cli can start draining the JSON
while we're filling the buffer, so we end up with a max allocation of
68MB.
But despite being less efficient (multiple queries to gossipd), it
actually speeds things up due to the parallelism:
MCP with -O3 -flto before vs after:
-listchannels_sec:8.980000-9.330000(9.206+/-0.14)
+listchannels_sec:7.500000-7.830000(7.656+/-0.11)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We now have a test blockchain for MCP which has the correct channels,
so this is not needed.
Also fix a benchmark script bug where 'mv "$DIR"/log
"$DIR"/log.old.$$' would fail if you log didn't exist from a previous run.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Rusty Russell
niftynei
Antoine Poinsot
Matt Whitlock
Christian Decker
Vasil Dimov
arowser
willcl-ark
trueptolemy
ZmnSCPxj
William Casarin