da82378313
This makes it easier for users to communicate the trace ID to the
Gerrit support team. E.g. users can simply use a bug ID as trace ID and
then there is no need to find the generated trace ID and post it in the
bug. This is nice because finding the generated trace ID can be tricky,
e.g. for REST calls it is returned as a response header that is only
visible with curl -v or in the browser network tooling.
This slightly changes the way how tracing is enabled:
1. REST API:
* old syntax:
a) ?trace -> enabled trace
b) ?trace=true -> enabled trace
c) ?trace=false -> disabled trace
* new syntax:
a) ?trace -> enabled trace with generated trace ID
b) ?trace=foo -> enabled trace with trace ID ‘foo’
2. Git API:
* old syntax:
a) -o trace -> enabled trace
b) -o trace=true -> enabled trace
c) -o trace=false -> disabled trace
* new syntax:
a) -o trace -> enabled trace with generated trace ID
b) -o trace=foo -> enabled trace with trace ID ‘foo’
3. SSH API:
* old syntax:
a) --trace -> enabled trace
* new syntax:
a) --trace -> enabled trace with generated trace ID
b) --trace --traceId=foo -> enabled trace with trace ID ‘foo’
c) --traceId=foo -> error, cannot use --traceId without --trace
Change-Id: I4f2c3b5748678bbdf8820463d8374b854200743a
Signed-off-by: Edwin Kempin <ekempin@google.com>
|
||
|---|---|---|
| .. | ||
| gerrit | ||
| gwtexpui/safehtml | ||