Permalink
Newer
Older
100644 176 lines (136 sloc) 7.45 KB
1
# Contributing to Cockroach
2
3
## Prerequisites
4
5
Before you start contributing, review these [basic guidelines](https://www.cockroachlabs.com/docs/stable/contribute-to-cockroachdb.html) on finding a project, determining its complexity, and learning what to expect in your collaboration with the Cockroach Labs team.
6
7
## Getting and Building
9
1. Install the following prerequisites, as necessary:
10
- A C++ compiler that supports C++11. Note that GCC prior to 6.0 doesn't
11
work due to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48891
12
- A Go environment with a recent 64-bit version of the toolchain. Note that
13
the Makefile enforces the specific version required, as it is updated
14
frequently.
15
- Git 1.8+
16
- Bash (4+ is preferred)
17
- GNU Make (3.81+ is known to work)
18
- CMake 3.1+
19
- Autoconf 2.68+
20
- Optional: NodeJS 6.x and Yarn 0.22.0+. Required when compiling protocol
21
buffers.
23
Note that at least 4GB of RAM is required to build from source and run tests.
25
2. Get the CockroachDB code:
27
```shell
28
go get -d github.com/cockroachdb/cockroach
29
cd $GOPATH/src/github.com/cockroachdb/cockroach
30
```
32
3. Run `make build`, `make test`, or anything else our Makefile offers. Note
33
that the first time you run `make`, it can take some time to download and
34
install various dependencies. After running `make build`, the `cockroach`
35
executable will be in your current directory and can be run as shown in the
36
[README](README.md).
38
### Other Considerations
40
- The default binary contains core open-source functionally covered by the
41
Apache License 2 (APL2) and enterprise functionality covered by the
42
CockroachDB Community License (CCL). To build a pure open-source (APL2)
43
version excluding enterprise functionality, use `make buildoss`. See this
44
[blog post] for more details.
45
46
[blog post]: https://www.cockroachlabs.com/blog/how-were-building-a-business-to-last/
48
- If you edit a `.proto` or `.ts` file, you will need to manually regenerate
49
the associated `.pb.{go,cc,h}` or `.js` files using `make generate`.
51
- We advise to run `make generate` using our embedded Docker setup.
52
`build/builder.sh` is a wrapper script designed to make this convenient. You
53
can run `build/builder.sh make generate` from the repository root to get the
54
intended result.
56
- If you plan on working on the UI, check out [the UI README](pkg/ui).
58
- To add or update a Go dependency:
59
- See [`build/README.md`](build/README.md) for details on adding or updating
60
dependencies.
61
- Run `make generate` to update generated files.
62
- Create a PR with all the changes.
66
[Style Guide](STYLE.md)
68
## Code Review Workflow
70
- All contributors need to sign the [Contributor License Agreement](https://cla-assistant.io/cockroachdb/cockroach).
72
- Create a local feature branch to do work on, ideally on one thing at a time.
73
If you are working on your own fork, see [this tip](http://blog.campoy.cat/2014/03/github-and-go-forking-pull-requests-and.html)
74
on forking in Go, which ensures that Go import paths will be correct.
76
`git checkout -b update-readme`
77
78
- Hack away and commit your changes locally using `git add` and `git commit`.
79
Remember to write tests! The following are helpful for running specific
80
subsets of tests:
82
```shell
83
make test
84
# Run all tests in ./pkg/storage
85
make test PKG=./pkg/storage
86
# Run all kv tests matching '^TestFoo' with a timeout of 10s
87
make test PKG=./pkg/kv TESTS='^TestFoo' TESTTIMEOUT=10s
88
# Run the sql logic tests
89
make test PKG=./pkg/sql TESTS='TestLogic$$'
90
# or, using a shortcut,
91
make testlogic
92
# Run a specific sql logic subtest
93
make test PKG=./pkg/sql TESTS='TestLogic$$/select$$'
94
# or, using a shortcut,
95
make testlogic FILES=select
96
```
97
98
Logs are disabled during tests by default. To enable them, include
99
`TESTFLAGS="-v -show-logs"` as an argument the test command:
100
101
```shell
102
make test ... TESTFLAGS="-v -show-logs"
103
```
104
105
When you're ready to commit, be sure to write a Good Commit Message™. Consult
106
https://github.com/erlang/otp/wiki/Writing-good-commit-messages if you're
107
not sure what constitutes a Good Commit Message™.
108
In addition to the general rules referenced above, please also prefix your
109
commit subject line with the affected package, if one can easily be chosen.
110
For example, the subject line of a commit mostly affecting the
111
`server/serverpb` package might read: "server/serverpb: made great again".
112
Commits which affect many packages as a result of a shared dependency change
113
should probably begin their subjects with the name of the shared dependency.
114
Finally, some commits may need to affect many packages in a way which does
115
not point to a specific package; those commits may begin with "*:" or "all:"
116
to indicate their reach.
118
- Run the linters, code generators, and unit test suites locally:
120
```
121
make pre-push
122
````
123
124
This will take several minutes.
125
126
- When you’re ready for review, groom your work: each commit should pass tests
127
and contain a substantial (but not overwhelming) unit of work. You may also
128
want to `git fetch origin` and run
129
`git rebase -i --exec "make lint test" origin/master` to make sure you're
130
submitting your changes on top of the newest version of our code. Next, push
131
to your fork:
132
133
`git push -u <yourfork> update-readme`
Sep 10, 2014
134
135
- Then [create a pull request using GitHub’s UI](https://help.github.com/articles/creating-a-pull-request). If you know of
136
another GitHub user particularly suited to reviewing your pull request, be
137
sure to mention them in the pull request body. If you possess the necessary
138
GitHub privileges, please also [assign them to the pull request using
139
GitHub's UI](https://help.github.com/articles/assigning-issues-and-pull-requests-to-other-github-users/).
140
This will help focus and expedite the code review process.
Sep 10, 2014
141
142
- Address test failures and feedback by amending your commits. If your change
143
contains multiple commits, address each piece of feedback by amending that
144
commit to which the particular feedback is aimed. Wait (or ask) for new
145
feedback on those commits if they are not straightforward. An `LGTM` ("looks
146
good to me") by someone qualified is usually posted when you're free to go
147
ahead and merge. Most new contributors aren't allowed to merge themselves;
148
in that case, we'll do it for you.
Oct 28, 2015
149
Oct 28, 2015
151
152
Peeking into a running cluster can be done in several ways:
153
154
- the [net/trace](https://godoc.org/golang.org/x/net/trace) endpoint at
155
`/debug/requests`. It has a breakdown of the recent traced requests, in
156
particularly slow ones. Two families are traced: `node` and `coord`, the
157
former (and likely more interesting one) containing what happens inside of
158
`Node`/`Store`/`Replica` and the other inside of the coordinator
159
(`TxnCoordSender`).
160
- [pprof](https://golang.org/pkg/net/http/pprof/) gives us (among other things)
161
heap and cpu profiles; [this golang blog post](http://blog.golang.org/profiling-go-programs) explains it extremely well and
162
[this one by Dmitry Vuykov](https://software.intel.com/en-us/blogs/2014/05/10/debugging-performance-issues-in-go-programs)
163
goes into even more detail.
Oct 28, 2015
164
165
An easy way to locally run a workload against a cluster are the acceptance
166
tests. For example,
Oct 28, 2015
167
168
```shell
Oct 28, 2015
169
make acceptance TESTS='TestPut$$' TESTFLAGS='-v -d 1200s -l .' TESTTIMEOUT=1210s
170
```
171
172
runs the `Put` acceptance test for 20 minutes with logging (useful to look at
173
the stack trace in case of a node dying). When it starts, all the relevant
174
commands for `pprof`, `trace` and logs are logged to allow for convenient
175
inspection of the cluster.