I am trying to debug some rspec tests that just hang at random without any apparent reason (Rails 6.1 Ruby 3 Rspec 4 Capybara), so I need access to Chrome logs so I can spend another 2 weeks looking for possible errors/problems.
Chrome is currently logging under /tmp/.com.google.Chrome.<random>/Default/chrome_debug.log (for some reason the datadir isn't ~/.chrome but goes under /tmp).
I'd like to get them to save under /app/log/chrome.log but for some reason none of the documented parameters work.
Based on the 1390745 possibilities people wrote about, I have tried:
--log-path=/app/log
--log-path=/app/log/chrome.log
--log-path=stdout
--log-file=/app/log/chrome.log
--log-file=stdout
--enable-logging=/app/log
--enable-logging=/app/log/chrome.log
--enable-logging=stdout
No luck, at most it disabled logging completely, despite the path being accessible and 777'ed.
Does anyone know how to get Chrome to log to a specific file?
Related
I just started looking into using the intern test framework to write functional tests for our mobile web apps. I'm looking to run these tests on actual iOS devices.
I setup Appium as the selenium server talking to the intern client. It works as expected on the simulator (although I get 'method not implemented' for many of the touch related methods).
I then decided to configure everything to work on my iPhone 6 plus, running iOS 9.0.2
After much wrangling with instructions from here, I got all the various pieces working in tandem (ios-webkit-debug-proxy, appium-server from the repo, etc.)
Finally, I decided to run my tests. When I run the test runner, Safari on my device starts up (via SafariLauncher App), and ios-webkit-debug-proxy and appium-server seem to be talking to each other. However, intern simply posts a 'created session on...' message, and then does nothing.
To ensure that the server was running properly, I decided to use a rest client to send it POST and GET requests using the JSONWire Protocol methods, and it seems to be loading the web-page url and responding to JSONWire requests. So clearly, something is going wrong with Appium talking to intern, or vice-versa.
Any help/insight on this would be greatly appreciated. Thanks!
I'm also in the unique position of having to run these locally, so using SauceLabs, BrowserStack, etc. is out of the question.
Got it working!!!
So the trick was to specify the proxyUrl config option.
My original conjecture was that there might have been some issues with basePath, etc., but I noticed my console.log statements that were outside of test cases were running properly.
That's when I noticed that the Safari page tried to navigate to localhost right after the test had started. This is my first time ever working with Intern or Appium or any Selenium stuff for that matter. Because you cannot access localhost (which in this case was meant for accessing the Intern server on my computer, since you're running the tests on your computer, and not on an external device), from the iPhone, I specified proxyUrl and proxyPort in the config, and specified my computer's local IP (192.168.x.x) and the Intern server port (9000). It still didn't work. I looked at the URL in Safari and noticed that it wasn't using the proxyPort field - so that the URL looked like: 192.168.x.x/__intern/clie... So I just tried to enter the port in the proxyUrl option - as that option's a string - 192.168.x.x:9000. And voila - it worked!!!
So hopefully this helps some other soul looking to have a native mobile test working with this setup!
I am using homestead as my development environment, I turned on the hhvm option for the site
sites:
- map: homestead.app
to: /home/vagrant/Code/wheremyprojectis
hhvm: true
I found that when there is an exception, everything is fine, but If I forgot to use namespace, got syntax error in the blade templates, I got nothing, blank page. I go check the logs and still nothing, the debug option is true. It's quite frustrating until I turned off the hhvm option.
I know it is not a big deal, but I still want to know is there any way to fix this?
I experienced the same problem. I searched around and found that it seems to be intentional:
https://github.com/facebook/hhvm/issues/4818
https://github.com/facebook/hhvm/issues/2571
Now you can poke through the github issues mentioned above, as well as these stackoverflow questions:
Display fatal/notice errors in browser
hhvm-fastcgi + nginx how to make it display fatal errors in the browser
laravel 5 show blank page on server error (and no laravel log), running with hhvm and nginx
For the time being, it ultimately boils down to writing your own handler, which isn't too bad. You can also tail the errors at /var/log/hhvm/error.log. Any errors that you intentionally want going to the browser you could of course handle using Laravel's error handling and logging.
UPDATE:
I reported this issue(and a fix) to the Laravel github here:
https://github.com/laravel/framework/issues/8744
I'm new to all this programming stuff, but trying to learn. I've had moderate success, but feel like it's more accidental than intentional. I'd appreciate help learning and understanding AppleScript in particular and networking, automation and programing in general.
Here's the environment: Almost all Apple with two Minis, a couple of laptops and various iOS devices including an AppleTV. The network consists of an Apple Airport Extreme, and an older D-Link switch I rescued (with permission) from the recycle bin at work. There are three Foscam wi-fi cameras as well. One of the Mini's (named Dora) is acting as a file server running Yosemite and the OS X Server application with a growing RAID 1 array for media storage.
What I’d like to be able to do is to run a single AppleScript from my laptop (Adam’s Macbook) that does the following:
Launch the application iCamSource on Adam’s Macbook
Change “Require Password to Wake” to true on Adam’s Macbook
Quit iTunes on Adam’s Macbook if it’s running
Start the current ScreenSaver locking Adam’s Macbook
Quit iTunes on Dora if it’s running
Start the current ScreenSaver on Dora if it’s not already running thereby locking Dora
I’d also like to have a single script on Adam’s Macbook that reverses these actions.
I have been able to launch iCamSource, set “Require Password to Wake” to true and start the ScreenSaver on Adam’s Macbook with a lot of Google-Fu, much trial and error, and quite a bit of frustration. I’ve not yet tried to quit iTunes if it’s running. That’s a new idea/requirement, but I think it should be pretty simple to add to the functioning script.
Where I’m running into trouble is controlling Dora from a script on Adam’s Macbook. I run into access errors (905 and 927 generally) about half to two thirds of the time I run the script. I’ve tried every syntax option and many combinations of syntax that I found via Google searches, but this is the one that has the most (but still inconsistent) success.
set remoteMachine to "eppc://remoteMachineadam:slajdlll#10.0.1.2"
tell application "iTunes" of machine remoteMachine
quit
end tell
When reading various articles explaining how to do this, sometimes the scripts use eppc, others use epps and I’ve not seen a consistent result with either one. Some of the articles use the actual IP address of the remote machine, others use password#name.local, so that syntax would be slajdlll#dora.local. That change also seems inconsistent in its results.
This seems to be a pretty basic task, and I’ve found lots of articles that do this very thing, but their methods don’t work in my environment. Many of the articles I’ve read are quite old, so the syntax may be out of date. I’m kind of at my wits end, but feel I should be able to make this work. Any help y’all could provide would be VERY appreciated.
Update - I had the scripts working last night in all regards except controlling the remote machine. This morning, I tried to open the "Away" script to again tackle the remote control issue and I get a Syntax Error Unable To Authenticate User message and the script won't compile. Here's the code:
tell application "System Events"
try
tell application "iTunes" to quit
end try
try
tell application "iTunes" of machine "eppc://Dora.local" to quit
end try
tell security preferences
set require password to wake to true
tell application "iCamSource" to activate
tell application "System Events" to start current screen saver
end tell
end tell
Both "Away" and "Home" scripts run from the AppleScript menu. When I unlock the ScreenSaver after running the "Away" script I get a dialog box stating "There was a problem connecting to the server "Dora's Airport Extreme. The share does not exist on the server. Please check the share name, and then try again." I'm beginning to think I may have a Permissions error on one or more of the machines, or if there's another problem that's making the scripts act inconsistently.
Thanks for all the help so far!
UPDATE Ok, it seems that there was a problem with iPhoto and Screen Saver interacting with the files on the networked hard drive. Opening iPhoto and slowly opening each image and re-linking those that had lost their location seems to have solved that problem. I am still encountering the access error code 905. In order to troubleshoot this process, I'm stripping the script down to one task, to quit iTunes on the remote machine named "Dora" Here's the code:
set Dora to "epps://adam:slajdlll#10.0.1.2"
try
tell application "iTunes" of machine Dora to quit
end try
Here's the event log that running that script generates:
tell application "iTunes" of machine "eppc://epps/adam%253kslajdlll%4010.0.1.2.local"
quit
--> error number -905
end tell
Is it possible/probable that the OS X Server application is mis-configured and is preventing the script from controlling the remote machine? Any other thoughts or things to try?
Thanks!
UPDATE 2
In the Groups section of the OS X Server application running on the remote machine, I found that there is a group called Apple Remote Events ACL. When I click the Edit button to view it's settings I see the actual account name is "com.apple.access_remote_ae-disabled". I wonder if this is telling us something. I searched several libraries for "com.apple.access" thinking it may be something that could be edited, but found nothing. I'm a member of that group. I tried removing myself from the group and tried the script with no change in the results so I added myself back into it.
UPDATE 3
As suggested I'm trying to eliminate OSX Server from the equation. To do this I'm going to try to launch TextEdit (since it's faster to launch than iTunes) on a third machine, known as Office Mini that does not have the Server application on it. Here's the script:
set OfficeMini to "eppc://10.0.1.196"
try
tell application "System Events" of machine OfficeMini
open "/Applications/TextEdit.app"
end tell
end try
When I ran the script initially I got two dialog boxes asking for username and password. I supplied them and checked the "Save in Keychain" checkbox. The script didn't open TextEdit, but I'm not getting the access errors. I am getting different errors though, -600 or -10010 depending on if I use System Events or the Finder. Using the list on this page: Nicholas Reilingh AppleScript Error Codes I find that with System Events I get an error -600 which means the application isn't running. That's no surprise, I'm trying to open it. When I use the Finder instead of System Events, I get an error -10010 which the list indicates is "Handler can’t handle objects of this class." So, there's progress, but I'm still stumped.
UPDATE 4 To further clarify/simplify, I tried running a simple script to open iTunes on the Office Mini locally on that machine and it worked fine.
tell application "iTunes" to activate
Substituting the "quit" for "activate", and trying this syntax with several different applications all worked quite well and very fast. I tried the same script and variations on the Dora machine with the same, successful results.
I then tried to make that syntax work remotely with this code:
set OfficeMini to "eppc://10.0.1.196"
tell application "iTunes" of machine OfficeMini to quit
I got this error message in the event log:
tell application "iTunes" of machine "eppc://10.0.1.196"
quit
--> error number -927
Result:
error "iTunes got an error: Unable to authenticate user." number -927
So, since OSX Server isn't running on the Office Mini, this seems to eliminate that as a culprit for the error. I tried altering the code with another variant I saw online that uses "epps" in place of "eppc" which results in a similar error message: "iTunes got an error: Remote access isn’t allowed." number -905. I did not get dialog boxes requesting credentials. When I've seen the "epps" string used online, it was formatted as "epps://username:password#location" so I used "epps://adam:slajdlll#10.0.1.196" and received the same error message. I even tried "Adam" in place of "adam" with no change in the results.
Yes, step 1 is to try to make your script as simple as possible to narrow down the point of failure.
I have copied your minimal script onto my machine & gotten it to work:
set Dora to "minipuppy.local"
try
tell application "iTunes" of machine Dora to quit
end try
A few notes:
I used "minipuppy.local" (the Bonjour linklocal name) rather than your eppc stuff, because I know what the former is, but not the latter. (You can get the link local name from the Sharing prefs panel* where it says "Computers on your local network can access your computer at:XXXXX"
An authentication dialog popped up saying that I needed to enter the username and password for the target machine - but it did have
an eppc address in the dialog (I didn't take a screenshot :-( )
After entering the username/password my script did nothing. Interestingly, I could not get it to print out a log as yours has?
I fixed this by going to the target Mac's System Preferences & enabling "Remote Apple Events" in the Sharing panel*. After turning that on, my script worked.
I'm hoping enabling "Remote Apple Events" is the ultimate solution to your simplified problem.
*Apple Menu -> System Preferences -> Sharing
The following issue is using the following:
AWS
PostgreSQL
Grails 2.3.3
Redis
On our account creation page, we are having some REALLY obscure functionality with the Grails withForm{}.invalidToken{} closure.
Upon hitting the page for the first time, everything works fine. You can post back to the server fine as long as you do not leave this page.
Upon leaving this page, either through navigation links or logging off, returning to the page (Again through navigation links or logging on and heading there), we can no longer submit, it hits the invalidToken closure every time.
I know AWS is involved as we took the project and deployed it to local machines both with IntelliJ and Tomcat by itself and both work fine. This issue only occurs upon deploying the WAR to AWS. (This occurs both with local builds and automated builds. They work locally but not on AWS)
We have spent almost a week on this issue trying to figure out why this is occuring, and all we have to show for it is we know AWS is somehow involved, but that's as far as we have gotten.
Does anyone have any insight into what would be causing our session to act like this?
After a LOT of searching about this issue, me and my team finally figured it out. Taken directly from our JIRA:
"This issue is caused by the implementation of tomcat-redis-session-manager used on AWS. As per their documentation (https://github.com/jcoleman/tomcat-redis-session-manager#session-change-tracking), there are "unintended consequence of hiding writes if you implicitly change a key in the session or if the object's equality does not change even though the key is updated." Specifically, the "useToken" implementation is Grails 2.3.8 is: "String generateToken(String url) { final UUID uuid = UUID.randomUUID() getTokens(url).add(uuid) return uuid }" The combination of these native implementations are there for incompatible.
The tomcat-redis-session-manager does support a manual dirty tracking mode by setting: RedisSession.setManualDirtyTrackingSupportEnabled(true); but this would require a forked build of SynchronizerTokensHolder in grails-core."
I am having trouble in understanding how can I check the logs for a test inside of an engine.
So My setup is the typical example, with a main app and engine. Inside of the engine I have a dummy application(inside the rspec folder) that I use to test my engine.
Now I am construction a lib file and would like to test it. This lib is a communication system to an API and I am getting problems when communicating but it's weird since I have another application making the same requests to the same API and it works. So I wanted to check the logs for the tests(so I can see the request and the messages being passed to the API).
My first thought was to look for a log folder at the engine, which does not exist???
second I thought maybe the main application(just to clear that out of the way), obviously the logs there were not logging anything.
Then I thought it could make sense if the dummy application had some logs and it does. The test.log, but that does not log anything.
Now just as detail, I am running my tests with guard & rspec.
let me know if you have some light on this or any other thoughts that could help.
all the best,
Andre