I have a Laravel7 app to which I send large data files (mp3/wav) and a CSV file via XHR to a POST method.
Generally, the stuff works like a charm, but for one of my testers (slower connection), it works randomly.
The thing is that I see in my Apache logs that the request is accepted, and responded with a 200 status code. The Laravel log doesn't show anything (I put some Log::debug at the start on my controller action).
If I test by myself, the log show what it has to.
In my /etc/php.ini, I set higher values for max_execution_time (3600), max_input_time(3600), post_max_size(4G), and upload_max_filesize(4G).
In my virtualhost conf, I set a TimeOut value to 3600.
Also, it seems, that when my tester sends the csv file as raw text file (.txt), the app responds, but I cannot figureout why, as we have no blocking concerning the non-audio file ...
Have you any idea ?
THanks in advance.
Related
I tried to run a POST method for API testing in JMeter, I have sent the parameters as needed in body using a csv file. It works with https url(domain url) but not working in localhost. In localhost, requested body is showing blank. Please help me to resolve this issue.
There should be no any difference between local and non-local URLs so most probably you made a configuration mistake somewhere. We cannot state where exactly without seeing the test plan so I would suggest:
Inspect request and response details using View Results Tree listener
Ensure that the variables originating from the CSV Data Set Config really exist and have their respective values by adding a Debug Sampler and seeing what variables are there.
Check jmeter.log file for any suspicious entries. If it doesn't say anything meaningful you can try increasing logging verbosity to DEBUG level.
Jmeter 2.12.
I used a scenario fully functional in front of a reverse proxy Apache. Recently we 've replaced the reverse proxy with the F5 BIGIP technology and now my scenario hangs.
The problem is for a particular ajax POST request the HTTP response is truncated : i receive a 200 OK but the HTML content is not full (no html tags for example). When i post the same request with Firefox the full content is ok.
Note that i don't receive the http header Transfer-Encoding: chunked.
In this case what can be the difference between Firefox and JMETER ?
Anyone have an idea on how could i get the full html response ?
Thanks for any reply.
That completely depends on the settings on your F5 and what exactly you mean with "response is truncated" and "no html tags". Do you get the correct response but the html tags are stripped out? or is the response just truncated so you i.e. only get the first n bytes?
The best way to find out what is actually going wrong is to use something like fiddler in between and try to find the real difference between the responses, especially regarding the response headers (Content-Length, Transfer-Encoding, etc).
When you found the actual difference please post here so we can help you further.
On a sidenote, by any chance do you have some custom coding on the f5 (iRules) which react to different user-agent settings?
Given you send identical requests you should be receiving identical responses.
Use JMeter's View Results Tree listener to inspect request details, or even better compare requests which are being sent by Firefox and JMeter using a lower level network sniffer tool like Wireshark, detect the differences and configure JMeter accordingly to send the same request(s) as Firefox does.
The other reason might be JMeter truncating large response, by default JMeter displays "only" first 10 megabytes in the View Results Tree listener, if this is the case - you can add the next line to user.properties file:
view.results.tree.max_size=0
and restart JMeter to pick the property up - it will suppress response truncation and you will be able to view the full response data.
Alternative way of setting the property is passing it via -J command line argument like:
jmeter -Jview.results.tree.max_size=0 ....
References:
Full list of command-line options
Apache JMeter Properties Customization Guide
Facing the issue while developing the performance script for salesforce application. The issue is, one of the apexremote post call is too large in size. Jmeter is getting hanged if i am trying to click that particular request. Even if I manage to enhance the post request somehow the cursor is not going to end of the line. Hence at the end of the line there is one CSRF which I am not able to correlate. This is only happening for only one apex post request. Because of this my orders are getting rejected while provisioning. I tried by increasing my heapsize in jmeter.bat file but no help.Searched in google but didn't find related queries. I tried the same thing with neoload and I was successfully access and enhanced that particular request.
Has anyone experienced ever! How to resolve this.
Please HELP!
1.Basically for that you need to be Patience after clicking on the request and when the request populate properly copy that post data into notepad++for correlation purpose.
2.Or other solution is Open the .JMX directly into notepad and from there copy the post data into other notepad++ instance and form there with some manipulation you can do correlation.
3.Hope you already changed the property "view.results.tree.max_size=0" in "jmeter.properties" file because may be for next correlation you will get the message like "Response data is to Large".
I had also faced the same issue several times in past, so in that case just copy the complete post body and paste it into notepad++ and modify the content there instead of modifying in jmeter itself, then paste the content back to jmeter and execute it.
This will resolve your issue but have patience while clicking on http request to copy the post body because it'll take some time to open the request properly
I have found out the solution for the very big request for which meter is getting hanged. The best solution is to take the request from the html body of the view result tree listener for the recorded original one and paste in the payload of a new sampler. Then add it in appropriate place accordingly. We can also keep a raw request and make the correlation changes there itself instead of messing around in the big request payload.
I attempted to record a file upload in JMeter, using port 8888, but the file failed to upload when using JMeter as the proxy ( Recording Controller and HTTP(S) Test Script Recorder ). I also tried to run this from within JMeter by choosing "Start" and viewing what I had just recorded in the View Results Tree listener. When I ran it this way, I received a 500 error for the file upload line in the View Results Tree listener.
So, I ran Charles network proxy while doing the file upload to see if the same thing happened. It did not. I was able to successfully upload the file going through the Charles network proxy on the same port 8888.
Furthermore, I took out all proxies and did the file upload successfully.
Only when I use JMeter as the proxy do I have a problem uploading the file.
In JMeter, after first running it with no changes, I then tried to change many settings like Implementation, Content Encoding, Browser-campatible headers, Retrieve All Embedded Resources, and many others to see if this would help, but it did not.
So, my gut tells me that there is a setting that is set incorrectly.
I found this post that talks about changing to use "Java" in the HTTP Request Defaults. ( Unable to upload image/file when using Jmeter HTTP Proxy server ) This did not help me though when I tried that.
I found this post also which mentions some other types of things ( Unable to record an upload file scenario in JMeter though it is possible by manual effort with specific browser settings ) . I did check the headers against Charles and they look the same.
Any ideas?
============= Edit below is adding the test plan screen shot===============
During recording you need to put the file you want to upload in jmeter/bin folder.
This is due to some limitations of browsers which do not transmit the full path.
So, apparently, I had to remove the parameters from the "Path" line and move them to the "Send Parameters With the Request:" for the line that I had the 500 error on ( the highlighted line in the screen shot I had attached above ).
In summary, here is what happened. When I recorded the file upload with JMeter using the Recording Controller and HTTP(S) Test Script Recorder, JMeter had put all of the parameters in the "Path" field for me on many of the lines that JMeter had recorded for me. I did not even think twice about what JMeter had done automatically for me. So, after I removed all text except "/" in the "Path" field and added each parameter one at a time under "Send Parameters With the request", I received a "200" status code on the line that had given me the problem. I also went into all other lines and made these same changes for the parameters. I still am not sure if the file upload is actually working, but I don't care about that right now since I did get a "200" on the line that was giving me a problem. That is enough progress to keep me moving ahead... :-).
There are plenty of examples of similar problems littered around the web but none of their solutions seem to fix this particular variation. Any suggestions would be appreciated.
Usually this problem occurs because a rogue link is causing a request for a resources like a favicon or css file to hit the dispatcher more than once, thus causing multiple dispatch processes and therefore multiple rows in your database.
I have checked that all the links on this very simple example page do actually resolve to the resource to which they point.
The session handler is setup as follows:
Zend_Db_Table_Abstract::setDefaultAdapter($db);
Zend_Session::setSaveHandler(new
Zend_Session_SaveHandler_DbTable($config->session->toArray()));
The db logging is setup as follows:
$writer = new Zend_Log_Writer_Db($db, $config->log->tableName,
$config->log->columnMap->toArray());
$logger = new Zend_Log($writer);
Both objects are correctly setup and can read and write to and from the database. Only everything happens twice. If I put a test log message anywhere in the application it is written into the database twice. If I increment three variables with every call to the index action - one stored in the session, one passed around via a Zend_Registry object and another local to the indexAction - only the session variable is incremented by 2. The Apache access log shows the correct amount of requests being fired from the page load and all have good response codes of either 200 or 304 (unchanged).
I have tried disabling all head links.
I have tried disabling the layout entirely.
I have localised everything to the dispatcher and exited before dispatch is run.
In all cases the extra write/increment takes place.
Any thoughts?
Thanks in advance for any help.
I seem to have found and fixed the issue. Chrome (and possibly all Webkit browsers) issues an additional HEAD request on top of the GET which means the application is hit twice and anything session based will be triggered as a result of both requests. My temporary solution is to put the following code near the start of my index.php file.
if ("HEAD" == $_SERVER['REQUEST_METHOD']) {
exit;
}
I hope that helps anyone with the same issue.
Google Chrome always asks for the favicon.ico by making annoying requests to the server. Take care about this in Chrome.
For more information:
http://framework.zend.com/issues/browse/ZF-11502?page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#issue-tabs
Thanks to the Sebastian Galenski contribution.