jMeter JSON parsing failing on CentOS - centos

I am running jMeter via Taurus (bzt) on CentOS. My first call is a login that needs to get the access tokens from the returned JSON. When I run this test on my Mac everything works as expected. But when I run it on my CentOS server I get the following error every time I try to extract data from a JSON path. I have tried reinstalling the plugin. I have also tried running it directly via jMeter. In both cases I still have the same problem. Any idea what is causing this error?
2016/06/08 14:34:05 WARN - com.atlantbh.jmeter.plugins.jsonutils.jsonpathextractor.JSONPathExtractor: Extract failed com.jayway.jsonpath.PathNotFoundException: Extracted array is empty
at com.atlantbh.jmeter.plugins.jsonutils.jsonpathextractor.JSONPathExtractor.process(JSONPathExtractor.java:104)
at org.apache.jmeter.threads.JMeterThread.runPostProcessors(JMeterThread.java:776)
at org.apache.jmeter.threads.JMeterThread.executeSamplePackage(JMeterThread.java:489)
at org.apache.jmeter.threads.JMeterThread.processSampler(JMeterThread.java:410)
at org.apache.jmeter.threads.JMeterThread.run(JMeterThread.java:241)
at java.lang.Thread.run(Unknown Source)

If you use Taurus for quite a while it might be the case it's using previously downloaded JMeter 2.13 which doesn't have JSON Path PostProcessor yet. So the following commands can help:
Upgrade Taurus to the most recent version:
sudo pip instal --upgrade bzt
Remove previous JMeter installation:
rm -rf ~/.bzt/jmeter-taurus
Run your test normally - Taurus should download JMeter 3.0
If you don't have possibility to upgrade JMeter and/or Taurus - use JSON Path Extractor plugin instead

It turns out my problem was a version mismatch. I downloaded the latest greatest (stable) jMeter when I started working with jMeter about a month ago. About a week after I downloaded jMeter they released 3.0. I didn’t start setting up my perf server until over a week later. When I installed jMeter on the sever it got 3.0, but my Mac had 2.13. So I was trying to run jmx files created with 2.13 on my 3.0 server. I never got any version mismatch errors just stuff like I shared above. After I upgraded my Mac and rewrote my jmx files via 3.0 everything worked fine.

Related

Creating clean Vapor project (using fluent) results in MongoKitten runtime error

I managed to setup Vapor with MongoDB and get it working but if I do a new clean project I get runtime error from MongoKitten. In detail:
I have an outdated vapor project that works fine from 27/07/2017.
I do the same exactly project now 24/10/2017.
When it compiles for the first time I get the following warning (don't know if is relevant to the following error):
Compilation finishes and I get the following error when I run the project:
I only know that this is caused by the preparation step in main.swift e.g. here:
config.preparations.append(User.self)
Without this step I dont get any runtime error (but fluent connection to mongodb isn't working properly either).
My config is pretty simple:
This happens in Swift 3.1.1 but same happens in Swift 4.0. Environment is Ubuntu 17.04.
How can I solve this?
MongoDB's $addFields stage requires MongoDB 3.4 or greater (we're almost at 3.6, so 3.4 is the latest release). I'm assuming you're running MongoDB 3.2 or lower. Ubuntu's official APT repository is and has been behind the official MongoDB releases for a long time. Some ubuntu versions will still install the deprecated 2.6 release of MongoDB.
The solution is to install the official latest MongoDB server as described here.

wso2iot 3.1.0 install problems

Clean install:
Downloaded wso2iot 3.1.0 from GitHub
Downloaded yajsw-stable-12.09
Put them in one file (wso):
Pointed jdk1.7.0_75 for JAVA_HOME
Pointed CARBON_HOME to my wso/wso2iot 3.1.0
This is what I downloaded only.
Run iot-server in folder bin.
After that nothing the system don't give me localhost why?
Is this installation correct for running wso2iot?
If not please advise me. Do I need to install something else?
First, you need to have java 8 on your system in order to run the WSO2 IoT 3.1.0. Secondly, if you need to run the WSO2 IoT 3.1.0 as a windows services, you need to have jajws 11.03.
Please refer the following document for more information
You can download the latest release candidate for WSO2 IoT 3.1.0 from here unzip it and start the pack in the following order.
Navigate to <IOTS_HOME>/bin and run ./broker.sh
./iot-server.sh &
./analytics.sh
Then you can login to the console from - https://localhost:9443/devicemgt
Additionally, to the above reply by jdk7. Make sure that you have JAVA 8. WSO2 IoT Server needs JAVA 8.
[1] https://docs.wso2.com/display/IoTS310/System+Requirements

Reverting to old Google Cloud SDK shell version

I am trying to deploy a Java project into Google App Engine from Eclipse and am blocked by JSPs.
Everything JSP makes my app not capable of deploying due to "'utf8' codec can't decode byte" error.
Not trying to insert any strange (not UTF-8) character, everything Eclipse is set to UTF-8.
Tried to send archives which were OK with the Google App Engine Tools for Eclipse (soon to be deprecated), won't deploy.
I've seen threads regarding reverting the Google Cloud Tools for Eclipse plugin to previous version (1.58 seemed to avoid a few problems) being a possible solution to these recent errors but I have a problem installing, the batch just tells me "Install will exit" ?!
Tried to uninstall previous (latest) versions before, made sure I was launching the bundled-python batch, still not capable of installing an old version of the Google Cloud SDK shell.
Would be glad if anybody had any suggestion at what makes my install fail.
Thanks in advance.
There are a few ways you can get older version of Google Cloud SDK.
Download versioned archive
(If you are on windows) Grab google-cloud-sdk-XXX.0.0-windows-x86_64-bundled-python.zip file.
Unzip it to some\dir
Add some\dir\google-cloud-sdk\bin directory to your system path
Restart your command prompt (or other apps which depend on gcloud) and run for example gcloud info, it should be fully functional installation, no need to run install.bat.
Alternatively, use existing SDK installation and gcloud component manager to go back to previous versions. For example
gcloud components update --version 158.0.0
target by version number using apt-get :
sudo apt-get install google-cloud-sdk=294.0.0-0

Unable to get GraphAware to run on Windows Community edition

I've tried installing and running from the Zip instead of EXE as other have mentioned.
Gives a 404 if authorization header is supplied, and 401 if not
GraphAware seems to have only a version for 2.2.2, whereas 2.2.3 is the latest on the neo4j site. Managed to track down 2.2.2 only to have it not work still. Any ideas?
It is correct the GraphAware modules are not compatible with Neo4j2.2.3 due to a bug in the Transaction data in neo4j itself (ref here https://github.com/neo4j/neo4j/issues/4910) which should be fixed in 2.2.4 or 2.3-RC1
If you installed via the zip, this should however work with 2.2.2, Please amend your question with the modules you use and their version, also all configuration you have set for the moduleS.

What terminal command stops webrick server besides Ctrl-C?

We are using Redmine. The mysql adapter segfaults in the mornings when the first Redmine operation is attempted, e.g. clicking on the Projects menu. Redmine works perfectly after Webrick is manually restarted.
The problem is that the mysql gem was compiled against MySQL 5.0, but we are using a later MySQL version. This is what I learned on the official Redmine issue tracker. For the moment, I am not going to attempt to recompile against our version.
I have a vbscript script that can restart the server during the very early hours of the morning. The vbscript script runs a batch file containing ruby script/server webrick -e production which can start webrick, but if webrick is already running, the script does not make a difference. I need a command that can stop the server.
I had a similar issue when installed on a windows system - after 8 hours of activity, the mysql adapter would segfault.
The fix was to place an other mysql binary in the ruby\bin.
The issue is described in the forum, and the link to an other dll is provided in the installation manual.
On a side note, but not really related to this issue: The question is tagged webrick, be aware that you should run redmine on mongrel instead of webrick. The above linked wiki page also says:
It is not recommended to use webrick for anything other than
development.