In this commit, we address a number of edge cases that were unaccounted
for when responding to errors that can be sent back due to an HTLC
routing failure. Namely:
* We’ll no longer stop payment attempts if we’re unable to apply a
channel update, instead, we’ll log the error, prune the channel and
continue.
* We’ll no remember which channels were pruned due to insufficient
fee errors. If we ever get a repeat fee error from a channel, then we
prune it. This ensure that we don’t get stuck in a loop due to a node
continually advertising the same fees.
* We also correct an error in which node we’d prune due to a
temporary or permanent node failure. Before this commit, we would prune
the next node, when we should actually be pruning the node that sent us
the error.
Finally, we also add a new test to exercise the fee insufficient error
handling and channel pruning.
Fixes#865.
In this commit, we add a new field to the LightningPayment struct:
PayAttemptTimeout. This new field allows the caller to control exactly
how much time should be spent attempting to route a payment to the
destination. The default value we’ll use is 60 seconds, but callers are
able to specify a diff value. Once the timeout has passed, we’ll
abandon th e payment attempt, and return an error back to the original
caller.
In this commit, we fix a goroutine leak that could occur if while we
were loading an error occurred in any of the steps after we created the
channel object, but before it was actually loaded in to the script. If
an error occurs at any step, we ensure that we’ll stop toe channel.
Otherwise, the sigPool goroutines would still be lingering and never be
stopped.
Since https://github.com/lightningnetwork/lnd/pull/506, `defaultchanconfs` is nested under `bitcoin`. Using this parameter at the top level generates a warning message:
```
2018-03-17 23:07:08.273 [WRN] LTND: /home/bitcoin/.lnd/lnd.conf:7: unknown option: defaultchanconfs
```
and using it nested under `bitcoin` does not.
In this commit, fix the inability of some users to connect to the DNS
seed using our direct TCP fallback. We do this as some resolvers filter
out our large SRV requests due to their size (they also include public
keys). Instead, we’ll use a direct TCP resolution in this case.
However, after a recent change, we forgot the period at the end of the
target DNS host. This is an issue as the domain needs to be fully
qualified.
The fix is easy, add a period within our string formatting to target
the proper sub-domain and SRV target.
Fixes#854.
The port numbers used in the ZMQ examples are inconsistent between:
- The initial explanation of ZMQ
```
They must be combined in the same ZMQ socket
address (e.g. `--zmqpubrawblock=tcp://127.0.0.1:28332` and
`--zmqpubrawtx=tcp://127.0.0.1:28332`).
```
- The sample bitcoin.conf
```
zmqpubrawblock=tcp://127.0.0.1:18501
zmqpubrawtx=tcp://127.0.0.1:18501
```
- The lnd command-line example:
```
lnd --bitcoin.active --bitcoin.testnet --debuglevel=debug --bitcoin.node=bitcoind --bitcoind.rpcuser=REPLACEME --bitcoind.rpcpass=REPLACEME --bitcoind.zmqpath=tcp://127.0.0.1:28332 --externalip=X.X.X.X
```
I changed the sample bitcoin.conf port numbers to 28332 so they match the rest of the examples.
In this commit, we update the DNS bootstrapper to match the new query
semantics expected by the new DNS server. We no longer hard code the
target DNS host, and instead, we’ll re-use the same target endpoint as
we only need the soaShim in order to establish a direct TCP connection
for the queries.
This commit removes two unnecessary disconnect/reconnect attempts
from the switch itest. Both occur after the primary test has been
completed, and were being executed before doing a final sanity check
that the path is still usable. Nothing about the test behavior should
change.
This commits adds a complement to the existing WaitPredicate
helper function, ensuring that a boolean statement remains
true for the duration of the provided timeout. This expands
our ability to do simple sanity checks where the wait-until
semantics of WaitPredicate may not be as-well suited.
This commits removes many of the timeout-based consistency
checks, which tend to be less dependable on Travis. They
have been replaced with stronger WaitPredicates, and the new
WaitInvariant where appropriate. This should give provide a
much greater degree of reliability in our highly-variant
testing environments.
