Diego Ongaro
Diego Ongaro
Seems like a good idea to be able to update the list of hosts on running clients.
scripts/logcabinctltest.sh sometimes hangs on "ServerControl info get" when the server has failed to start. The end of the server's log looks like this: ``` 2016-06-08 18:09:28.051544 Server/Main.cc:345 in main() NOTICE[5578:evloop]:...
LogCabin currently uses a mixture of Travis CI, pre-commit hooks, and a hookmatrix script. The pre-commit hooks: - \+ are pretty extensive - \- must be set up and configured...
I see that @siddontang added a Jepsen test for LogCabin a few months ago. It uses the CLI client. https://github.com/aphyr/jepsen/commit/90720bf6dfcaf8b9e7cf613783c0dddb4df6547c @siddontang: What's the state of this test? Did you find...
It came up in talking with @astigsen and @tnachen today that we should highlight where LogCabin is used in production. Maybe just prominently on the README?
ClientImpl.cc misses a call to doneWithRPC() for keepalives that have to be retried: https://github.com/logcabin/logcabin/blob/v1.1.0/Client/ClientImpl.cc#L397 This caused the client's firstOutstandingRPC to get stuck, causing servers to accumulate session state, write a...
This would help log replication performance. I started to implement this for my dissertation (see #95 / a7ce12d / nasty-thesis-wip tag for code) but never cleaned it up. See the...
The pre-commit hooks are difficult for people to run right now, mostly for historical reasons/I'm lazy. The best docs are presently buried here: https://github.com/logcabin/logcabin/pull/184#issuecomment-132903180 It should be entirely possible to...