内容简介:I wanted to know what data Oculus Runtime sends home. To do that, I’veYou can find my findings below.I’ve captured HTTP packets using Wireshark, and decrypted them using SSL KEYLOG generated by
I wanted to know what data Oculus Runtime sends home. To do that, I’ve implemented a custom tool to extract TLS keys from a running Oculus Runtime instance.
You can find my findings below.
Methodology
I’ve captured HTTP packets using Wireshark, and decrypted them using SSL KEYLOG generated by oculus-tls-extractor .
Then I’ve manually inspected the captured packets over multiple sessions.
Limitations
- I was not able to inspect data from OculusClient, which is an Electron application, and there may be additional data being sent from there.
- I manually inspected the packets and it is certain that I’ve missed something.
Findings
The most used endpoints were: https://graph.oculus.com/logging_client_events
and https://graph.oculus.com/graphql
.
/graphql
endpoint was mostly used as a read-only endpoint, with few exceptions.
For example:
Request: POST /graphql?forced_locale=en_US HTTP/1.1 Host: graph.oculus.com ... access_token=<removed_access_token>&variables=%7B%7D&doc_id=<removed_doc_id> --- Response: HTTP/1.1 200 OK ... Access-Control-Allow-Origin: https://facebook.com Content-Type: text/html; charset="utf-8" ... {"data":{"viewer":{"user":{"current_party":null,"friends":{"count":0,"edges":[]},"id":"<my user id>"}}}}
There were a few exceptions.
POST /graphql Form data: method = POST forced_locale = en_US doc_id = <removed_doc_id> variables = { "input": { "client_mutation_id": "30166", "device_id": "<device uuid>", "device_name": "UNGNU", "device_serial": "<another uuid>", "hardware": [ { "battery": "DISCHARGING", "battery_percent": 0, "connection": "connected", "operational": "operable", "serial": "<device serial>", "type": "HMD_LAGUNA" }, { "battery": "DISCHARGING", "battery_percent": 0, "connection": "disconnected", "operational": "inoperable", "serial": "<device serial>", "type": "INPUT_LLCON" }, { "battery": "DISCHARGING", "battery_percent": 0, "connection": "disconnected", "operational": "inoperable", "serial": "<device serial>", "type": "INPUT_RLCON" } ], "meets_min_spec": "PASS" } }
/logging_client_events
this is a more interesting endpoint.
It has a stream of data being sent by the runtime.
The requests have the following format:
Request: POST /logging_client_events HTTP/1.1 Host: graph.oculus.com Content-Type: application/json ... { "message": { "app_id": 490451621120355, "app_ver": "16.0.0.118.452", "data": [ { "extra": { "client_session_id": 1, "hmd_firmware_version": "2.2.0", "hmd_os_build": "0", "hmd_type": "HMD KM", "id": "<user_id>", "in_staged_rollout": false, "is_core2": true, "is_dash_up": false, "left_lcon_firmware_version": "unavailable", "release_channel": "LIVE", "remote_firmware_version": "unavailable", "right_lcon_firmware_version": "unavailable", "timestamp_gaps": "8", "tracker_firmware_version": "", "valid_imu_sample": "57937" }, "name": "hal_imu", "time": 1587645768.38274 } ], "device_id": "<device_id>", "log_type": "client_event", "oculus_access_token": "<access_token>", "oculus_userid": "<user_id>", "seq": 3, "session_id": "1587645706.843546", "time": 1587645777.5674851, "uid": "0" } } Response: HTTP/1.1 200 OK {"checksum":"","config":"","app_data":"{}"}
This is a remote logging endpoint. I’ve dumped all logs, sorted them by types and inspected a few log entries of each type.
If you want to inspect the data yourself, I’ve uploaded a my (cleaned up) log entries in LJSON format here .
One of the events caught my eye:
{ "extra": { ... "level": "Error", "message": "c:\\cygwin\\data\\sandcastle\\boxes\\trunk-hg-ovrsource-win \\software\\oculussdk\\pc\\oaf\\liboaf\\steamparser\\parsing.cpp(163) : Error parsing Steam info (1971039)", ... }, "name": "oculus_application_framework_error", "time": 1587238393.064477 }
It looks like Oculus Runtime tries to parse Steam information, but fails to do so? I would appreciate if it didn’t.
OK, let’s look at oculus_service_start
entry: https://gist.github.com/m1el/373b4fa14a46b22f933b44d96dd26f7b
"CpuList": [ { ... "MaxFrequencyMhz": 3600, "Model": 0, "Name": "Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz", "NumberOfCores": 8, "NumberOfLogicalProcessors": 16, ... "UsbMatchedController": { "Compatibility": 3, "Compatibility_str": "UNKNOWN_USB3", "DriverDescription": "USB xHCI Compliant Host Controller", ... "monitors": [ { "horizontal_active_pixels": 3840, "horizontal_image_size_mm": 1872, "vertical_active_pixels": 2160, "vertical_image_size_mm": 1053 }, { "horizontal_active_pixels": 1920, "horizontal_image_size_mm": 518, "vertical_active_pixels": 1200, "vertical_image_size_mm": 324 }, { "horizontal_active_pixels": 1920, "horizontal_image_size_mm": 510, "vertical_active_pixels": 1080, "vertical_image_size_mm": 287 } ], ... "VideoCardList": [ { "LUID": "<LUID>", "Name": "AMD Radeon R9 290/390", ... "motherboard_manufacturer": "ASUSTeK COMPUTER INC.", "motherboard_product": "PRIME Z390-A", "motherboard_serial": "<motherboard_serial>", "motherboard_version": "Rev 1.xx", "os": "Windows 10 64-bit", "os_build": "18363", ...
That’s a lot of info about the system! All GPUs, all monitors, USB controllers, motherboard serial number, etc.
The rest of the logging events were rather boring. Most of them were:
- applications being launched in VR
- guardian area
- debug information about inside-out tracking
- lots and lots of debug information metrics
- some cryptic errors
- no information about room geometry or photos of the room
Conclusions
The findings were rather boring. The most egregious things I’ve found were:
- Sending a dump of system information on service startup.
- Trying to parse Steam information?
I would like it if Oculus didn’t send this information home, but this probably won’t happen.
My analysis of the logs is incomplete, if you’d like to do inspect the data yourself, feel free to take a look at the logs I’ve captured myself or capture your own logs using the tool I wrote .
以上就是本文的全部内容,希望本文的内容对大家的学习或者工作能带来一定的帮助,也希望大家多多支持 码农网
猜你喜欢:本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们。
SSA:用户搜索心理与行为分析
[美] 罗森菲尔德(Louis Rosenfeld) / 汤海、蔡复青 / 清华大学出版社 / 2014-4-1 / 59.00
何为站内搜索分析(SSA)?它如何帮助你挖掘用户搜索曰志,从中洞悉用户搜索心理和行为,从而有针对性地改善用户体验,提升网站价值?这些都可以从《SSA:用户搜索心理与行为分析》中找到答案。《SSA:用户搜索心理与行为分析》首先通过故事来说明SSA是如何使Vanguard集团起死回生的,简要介绍SSA并指导读者动手实践。其次,通过丰富的实例来介绍很多工具和方法,帮助读者着手分析用户查询数据,从中获得更......一起来看看 《SSA:用户搜索心理与行为分析》 这本书的介绍吧!