Open sclawrenc opened 3 years ago
Having the same issue on my Pi 4.
Exact same problem on my pi 4.
Having same issue on Ubuntu 18.04
I believe I found a workaround on this error by turning off Boot to Desktop via sudo raspi-config
(Option 1, then Option S5).
Upon reboot, and restart of the processes, I don't get the Chrome error anymore.
Of course, this is only useful if you don't plan on using the desktop.
I believe I found a workaround on this error by turning off Boot to Desktop via
sudo raspi-config
(Option 1, then Option S5). Upon reboot, and restart of the processes, I don't get the Chrome error anymore. Of course, this is only useful if you don't plan on using the desktop.
Thanks for your help. :)
On a similar topic, will the Inventory Hunter docker run with the Lite Pi OS? I could just run that one with less resources anyway, but I wasn't sure initially so I just installed the full Pi OS with desktop.
Eric, thank you for sharing this tool.
In my case (Ubuntu 16.04 PC) I fixed it by adding a new Chrome option in src/driver.py. This option should start Chrome without a graphical interface. Here is the patch:
diff --git a/src/driver.py b/src/driver.py
index 2f33da4..dd29718 100644
--- a/src/driver.py
+++ b/src/driver.py
@@ -66,6 +66,7 @@ class SeleniumDriver(Driver):
self.options.page_load_strategy = 'eager'
if getpass.getuser() == 'root':
self.options.add_argument('--no-sandbox') # required if root
+ self.options.add_argument('--headless')
self.options.add_argument('--disable-blink-features=AutomationControlled')
self.options.add_argument('--disable-dev-shm-usage')
self.options.add_argument(f'--user-agent="{user_agent}"')
After that, if you want to use Docker, rebuild the image and pass it to to the docker_run.bash script (-i option). Hope it helps.
I've experienced the Chrome issue and 100% disk usage issue until I switched to ubuntu server 20.04; before I was using desktop 20.04. Seems like it's related to GUI somehow.
Thanks for sharing Inventory Hunter.
I am experiencing some issues getting data from BestBuy and Amazon to pull correctly on Pi 4 running the standard PI OS with desktop , but Newegg seems to work fine. The logs are attached from all 3 sites. Strangely, the BestBuy scraper seems to work fine when running in Windows 10. The Amazon scraper runs in Windows 10, but very slowly displaying an error "W2020-12-19 21:32:09,244 [root] unable to save screenshot of webpage: Message: timeout: Timed out receiving message from renderer: 10.000 (Session info: chrome=83.0.4103.116)"
I have attached some logs below.
inventory hunter logs.zip