// AddRawLink adds a link to this node
AddRawLink(name string, lnk *Link) error
// Return a copy of the link with given name
GetNodeLink(name string) (*Link, error)
The same keys + nonces in secio were being observed. As described in
https://github.com/ipfs/go-ipfs/issues/1016 -- the handshake must
be talking to itself. This can happen in an outgoing TCP dial with
REUSEPORT on to the same address.
reuseport is a hack. It is necessary for us to do certain kinds of
tcp nat traversal. Ideally, reuseport would be available in go:
https://github.com/golang/go/issues/9661
But until that issue is fixed, we're stuck with this. In some cases,
reuseport is strictly a detriment: nodes are not NATed. This commit
introduces an ENV var IPFS_REUSEPORT that can be set to false to
avoid using reuseport entirely:
IPFS_REUSEPORT=false ipfs daemon
This approach addresses our current need. It could become a config
var if necessary. If reuseport continues to give problems, we should
look into improving it.
GNU Make's wildcard function does not recurse into subdirectories when
passed the '**' glob, which results in adding a dependency only to .go
files in the first level of subdirectories under the source root.
We shell out to 'find' instead, which catches all .go files in the
given directory.
Running make -jN would result in the tests starting to execute
before the tests binaries were built, resulting in the error:
"Cannot find the tests' local ipfs tool"
Each test now depends on the deps. They also depend on a new
target for cleaning the test results, so that the tests can
write new clean results.
The aggregate target also needs to depend on the same test
results clean target, as well as the tests themselves, so
that the aggregation happens when all tests have finished
running.
By introducing a separate target for cleaning test results we
also ensure that we don't end up removing and rebuilding
the binary on each test run.
The result is that the tests *can* be run with with -jN > 1,
but individual tests may still not supports this, so to get
stable test results it's still recommended to run them in
sequence.
These did not work before, and had some unnecessary complexity.
Now the filters use only one hashing function, no bignum arithmetic, and gets the additional bit positions by repeatedly hashing the result of prior hash.
Since we're not concerned about crypto hashing here, this should be a win.
External interfaces unchanged.
