Closed Bug 1115134 Opened 10 years ago Closed 10 years ago

[Raptor] Port mozversion to Node.js for test metadata

Categories

(Firefox OS Graveyard :: Gaia::PerformanceTest, defect)

ARM
Gonk (Firefox OS)
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: Eli, Assigned: Eli)

References

Details

(Keywords: perf)

Attachments

(4 files)

In order to have access to the same test metadata that Datazilla has, e.g. Gaia and Gecko revisions, we need to port mozversion over to Node.js.
Porting device interactions from Raptor and orng into mozdevice. Upcoming PR for integration into Raptor.
Attachment #8553292 - Flags: review?(rwood)
This PR removes the device interactions from Raptor and re-integrates it through the previous MozDevice PR.
Attachment #8553294 - Flags: review?(rwood)
Blocks: 1115138
Comment on attachment 8553292 [details] Link to Github pull-request: https://github.com/eliperelman/node-mozdevice/pull/1 Awesome patch! Just a couple of issues and nits (see comments in PR).
Attachment #8553292 - Flags: review?(rwood) → review-
(In reply to :Eli Perelman from comment #2) > Created attachment 8553294 [details] > Link to Github pull-request: https://github.com/mozilla-b2g/raptor/pull/7 > > This PR removes the device interactions from Raptor and re-integrates it > through the previous MozDevice PR. The patch LGTM. When testing it out, I notice that if I have the gaia/.env file (with the RAPTOR_DATABASE vars) but don't have the actual DB / raptor-docker container running, it attempts to write to the DB anyway and results in this error: [Test] Run 1 complete mozdevice:util Killing process 4301 +207ms mozdevice:command [Executing] ANDROID_SERIAL=3567d06c adb shell 'kill 4301' +0ms raptor:reporter Writing report results to file +50ms raptor:reporter Writing report results to database +1ms raptor:reporter Error writing report results to database: {} +9ms Error: No host available at InfluxRequest._request (/home/rwood/raptor/node_modules/influx/lib/InfluxRequest.js:103:21) at InfluxRequest._parseCallback (/home/rwood/raptor/node_modules/influx/lib/InfluxRequest.js:122:19) at Request._callback (/home/rwood/raptor/node_modules/influx/lib/InfluxRequest.js:113:10) at self.callback (/home/rwood/raptor/node_modules/influx/node_modules/request/request.js:373:22) at Request.EventEmitter.emit (events.js:95:17) at Request.onRequestError (/home/rwood/raptor/node_modules/influx/node_modules/request/request.js:971:8) at ClientRequest.EventEmitter.emit (events.js:95:17) at Socket.socketErrorListener (http.js:1547:9) at Socket.EventEmitter.emit (events.js:95:17) at net.js:441:14 mozdevice:logging Stopping logging process +61ms Should it check first if the DB is even running before attempting to write? Maybe in the stdout it should just say something like "raptor database not found, reporting to log file only"? Or does it matter? I guess we are assuming if the .env exists then the user intends to write to the DB.
Comment on attachment 8553292 [details] Link to Github pull-request: https://github.com/eliperelman/node-mozdevice/pull/1 Relevant nits and changes made.
Attachment #8553292 - Flags: review- → review?(rwood)
(In reply to Robert Wood [:rwood] from comment #4) > Should it check first if the DB is even running before attempting to write? > Maybe in the stdout it should just say something like "raptor database not > found, reporting to log file only"? Or does it matter? I guess we are > assuming if the .env exists then the user intends to write to the DB. The connection to the Raptor DB is HTTP, so stateless. The only way to know if the connection is good is to hit it with the ping route: http://raptor:8086/ping. I'd hate to do a ping before each write, so I think what we have in place is fine. If the environment specifies the connection information to write to, and the environment doesn't actually have that connection, I think it's legit to fail hard and fast.
Comment on attachment 8553292 [details] Link to Github pull-request: https://github.com/eliperelman/node-mozdevice/pull/1 Tested in conjunction with the other patch in this bug. Ran the raptor performance launch_test on both the emulator and the flame device, single and multiple times, and verified output in raptor.log and the visualization. Also ran the boot_test on the flame device, single and multiple times.
Attachment #8553292 - Flags: review?(rwood) → review+
Attachment #8553294 - Flags: review?(rwood) → review+
Awesome! I'll merge and start the process of getting this integrated back into Gaia.
mozdevice published to npm: https://www.npmjs.com/package/mozdevice orng is now npm deprecated.
Could either of you rubber-stamp this for bumping gaia-raptor?
Attachment #8555515 - Flags: review?(kgrandon)
Attachment #8555515 - Flags: review?(jlal)
Attachment #8555515 - Flags: review?(kgrandon)
Attachment #8555515 - Flags: review?(jlal)
Attachment #8555515 - Flags: review+
One more? I will await the green run before landing.
Attachment #8555520 - Flags: review?(kgrandon)
Attachment #8555520 - Flags: review?(kgrandon) → review+
Blocks: 1127019
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: