[webdriver] org.openqa.selenium.WebDriverException

[sqa tester]

I have been on 2.0a2 webdriver for ~3 months and it was stable.

However, today, at one point web driver is not able to launch browser
at all, and I got the following error:

Cause :org.openqa.selenium.WebDriverException: Unable to bind to
locking port 7054 within 45000 ms
System info: os.name: 'Windows XP', os.arch: 'x86', os.version: '5.1',
java.version: '1.6.0_17'
Driver info: driver.version: firefox
2010-05-21 09:54:51,905 INFO [loadSelenium]
Message :org.openqa.selenium.WebDriverException: Unable to bind to
locking port 7054 within 45000 ms
System info: os.name: 'Windows XP', os.arch: 'x86', os.version: '5.1',
java.version: '1.6.0_17'
Driver info: driver.version: firefox
System info: os.name: 'Windows XP', os.arch: 'x86', os.version: '5.1',
java.version: '1.6.0_17'
Driver info: driver.version: firefox

I upgraded my firefox to latest version and it 's now working fine.
(Additionally, I am also trying 2.0a4 since I think the certificate
error fixes I need are there).



What could have caused this issue?

--
You received this message because you are subscribed to the Google Groups "webdriver" group.
To post to this group, send email to webd...@googlegroups.com.
To unsubscribe from this group, send email to webdriver+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/webdriver?hl=en.

[Simon Stewart]

We serialize startup of firefox instances so that we guarantee that
each instance can be communicated with effectively. We wait up to 45
seconds for this port to be freed before timing out. Generally the
reason for the time out is because the machine is starved of resources
and can't start firefox instances quickly enough. Heavy disk IO can
sometimes cause this starvation, as can starting a large number of
firefox instances at the same time.

Regards,

Simon

[Vishnu]

Thanks. Is there a way to increase the timeout ? 

To unsubscribe from this group, send email to webdriver+unsubscribe@googlegroups.com.

[Simon Stewart]

Modify the timeout on the FirefoxBinary. We should really make it a
capability or expose it on the FirefoxProfile, but have never got
round to it.

Simon

>> webdriver+...@googlegroups.com.

>> For more options, visit this group at
>> http://groups.google.com/group/webdriver?hl=en.
>>

> --
> You received this message because you are subscribed to the Google Groups
> "webdriver" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/webdriver/-/E3BL-wOSLbQJ.

>
> To post to this group, send email to webd...@googlegroups.com.
> To unsubscribe from this group, send email to

> webdriver+...@googlegroups.com.

[Mike Riley]

I did a similar logic for my tests where I needed to succeed at some critical operation, such as setting up or removing data that would affect later tests.

I would try 5 times with a delay of either 5 or 10 seconds between each try.  It did make those critical method much more robust.  I also logged that it was doing a retry and what the count was, so I knew that it had been triggered.

Mike

On Wednesday, August 22, 2012 6:59:34 AM UTC-7, Alexander Zhovnuvaty wrote:

Good idea. Thanks! We have updated out test framework this way: when new browser instance is starting up and gets an exception, it waits for 1 second and makes one more attempt. Once it is reached max attempts amount it throws the exception, otherwise proceeds with testing. However, I assume 1 second is not enough, may be this value should be increased.

Понеділок, 7 червня 2010 р. 03:23:22 UTC+3 користувач Simon Stewart написав:

[Nischand]

Hi Alexander,

Could you share the logic code please. 

I have similar situation, few weeks back Selenium Tests are stable when running Parallel with 2-Firefox, Chrome & IE  through Grid2 > used new selenium V.2.35.0

and I faced this error today (Unable to bind to locking port 7054 within 45000 ms).

My set-up :

4 Virtual machines with diffrent versions of browsers

Using Grid2  > Web-driver V.2.35.0 > TestNG > Ant > Jenkins 

I updated all my FF in virtual machines to 23 version, tests run normally for few times but ideal solution is better to catch exception like what you implemented.

Please share Thank you.

On Wednesday, August 22, 2012 2:59:34 PM UTC+1, Alexander Zhovnuvaty wrote:

Good idea. Thanks! We have updated out test framework this way: when new browser instance is starting up and gets an exception, it waits for 1 second and makes one more attempt. Once it is reached max attempts amount it throws the exception, otherwise proceeds with testing. However, I assume 1 second is not enough, may be this value should be increased.

Понеділок, 7 червня 2010 р. 03:23:22 UTC+3 користувач Simon Stewart написав: