Open GoogleCodeExporter opened 9 years ago
This may help as a quick solution:
https://code.google.com/p/chromedriver/issues/detail?id=838&q=memory&colspec=ID%
20Status%20Pri%20Owner%20Summary
Original comment by stefan.l...@gmail.com
on 9 Sep 2014 at 8:36
I have already tried that and it does not help here.
Original comment by gopalakb...@gmail.com
on 9 Sep 2014 at 8:42
can reproduce as well - our long weekend tests also ended with a lot of
consumed memory of the ChromeDriver.exe (>1GB).
We even started Chrome with --silent mode and all log levels set to off.
ChromeDriver v2.10
Chrome v37.0.2062.103 m
Selenium v2.41
Original comment by stefan.l...@gmail.com
on 9 Sep 2014 at 10:14
Attachments:
Here the issue is with chrome process rather than chromedriver
Original comment by gopalakb...@gmail.com
on 9 Sep 2014 at 4:30
Original comment by klundberg@chromium.org
on 11 Sep 2014 at 7:11
#3:
Would you mind providing portion of the chromedriver.log and maybe a test html
page as well?
The --silent mode only applies to the logging output of chromedriver, but if
Chrome itself produces a lot of logging entries, those will be cached in memory
by chromedriver until a session is deleted by command driver.quit
Original comment by st...@chromium.org
on 11 Sep 2014 at 7:22
Original comment by st...@chromium.org
on 16 Sep 2014 at 5:09
Attached you can find the logs.
The file was quite big, so I splitted it.
Original comment by stefan.l...@gmail.com
on 16 Sep 2014 at 12:04
Attachments:
Second Log
Original comment by stefan.l...@gmail.com
on 16 Sep 2014 at 12:09
Attachments:
Third Log
Original comment by stefan.l...@gmail.com
on 16 Sep 2014 at 12:09
Attachments:
Fourth Log
Original comment by stefan.l...@gmail.com
on 16 Sep 2014 at 12:11
Attachments:
Unfortunately I cannot provide you with a test page.
But we actually not doing anything special.
We are clicking a button continuously, which sends out a WebSocket request.
When the response is received, the buttons state should change, which is what
we query after clicking via CSS Selectors.
Original comment by stefan.l...@gmail.com
on 16 Sep 2014 at 12:12
Issue 927 has been merged into this issue.
Original comment by samu...@chromium.org
on 10 Oct 2014 at 6:22
Issue 926 has been merged into this issue.
Original comment by samu...@chromium.org
on 10 Oct 2014 at 6:22
I've just uploaded a possible fix, based on Alp Soyupak's investigation. I'll
publish this CL for review once I've run a few longer-running tests on it.
https://codereview.chromium.org/643003003
Original comment by samu...@chromium.org
on 10 Oct 2014 at 6:29
Actually it looks like there are two separate issues being discussed here. The
original issue is with memory use in chrome, and the other issue is a memory
issue in chromedriver. I'm going to continue putting updates on the
chromedriver leak to issue 926, and keep this issue open for the chrome memory
usage issue.
Original comment by samu...@chromium.org
on 11 Oct 2014 at 1:40
Original comment by st...@chromium.org
on 14 Oct 2014 at 4:06
Ran following set up for more than 10,000 times, and the memory within 60 MB
Google Chrome 39.0.2171.65 (Official Build) m
ChromeDriver 2.12
if you can try the newer version.
Original comment by andrewch...@chromium.org
on 21 Nov 2014 at 10:11
Original comment by samu...@chromium.org
on 21 Feb 2015 at 12:18
Original issue reported on code.google.com by
gopalakb...@gmail.com
on 8 Sep 2014 at 9:21Attachments: