I am running into an issue with the Grails Pdf plugin which uses Flying Saucer. Everything works as expected until I deploy onto an Ubuntu server running Tomcat6. Then references in my gsp's to css and images fail, though I still get the PDF to render.
I have tried two different approaches to building the PDF
ITextRenderer renderer = new ITextRenderer()
renderer.setDocument(url)
renderer.setDocumentFromString(content, baseUri)
Running a war with 'grails prod run-war' works, running and a dummy app with no security works locally, but fails when I deploy it on the server as well. (though none of the content I am trying to render is secured anyway), the URL's of the images are correct. (I have tried both absolute and relative URL's) neither gets rendered in the PDF, but if you request those resources from a browser they are there. References to images not hosted on the server do work.
All this leads me to believe that the tomcat6 that gets installed with ubuntu when you do sudo apt-get install tomcat6 is configured funny somehow. I know that it runs with user 'tocat6' instead of 'root' as many installations do. Could that be causing Flying Saucer to somehow not have the right access to get at the files being referenced?
Since everything except the images/css is working, I guess your baseURI is not correct?! I have this code on a production system and it is working. All the images are referenced absolut:
renderer.setDocument(doc, request.getScheme() + "://" + request.getServerName() + ":" + request.getServerPort());
What is your baseURI set to?
Related
I have an Expo managed react native application. I created my .env file in the root of my project, installed react-native-dotenv and set up babel to use it. After a while I managed to get it to work.
I have my environment variable
ENDPOINT=http://127.0.0.1:8000/api
and i use it with
process.env.ENDPOINT
After a while I decided to test the android version of the app, so i changed the endpoint url to my LAN ip and restarted the server. The problem is that even after restarting the server, the cache and the computer, when I call process.env.ENDPOINT it keeps the first url I set.
Here's a list of the things i tried:
restarting the server
restarting the server and the cache
restarting the whole computer
change the variable name to REACT_APP_ENDPOINT as many suggested (I get undefined, it's still stuck to ENDPOINT)
empty expo cache
The strange thing is that I already changed that same variable twice (from 127.0.0.1:8000 to 127.0.0.1:8080 and back for a problem with backend) and had the same problem, but it went away by itself after a couple of minutes (and server restarts).
This time I've been trying to get it to work for 7 hours and nothing has changed.
Any idea?
I had the same issue and managed to run the app with .env changes after using the following command.
expo r -c
reference: https://github.com/goatandsheep/react-native-dotenv/issues/75#issuecomment-728055969
After a couple hundred more tests I gave up and implemented a "custom" solution, without any external library:
Switched .env files to TypeScript files (E.g. .env.development -> env.development.ts)
Set up an object named env that has all environmental variables as properties
export const env = {
VAR1: 'foo',
...
}
Imported this constant inside the application entry point (in my case App.tsx)
Inside the main constructor assign env to global.env
Use global.env instead of process.env
Not sure if this is the best practice, but solved my problem for now, works like a charm and doesn't require me to reload my application at every change. I'm a bit concerned by the security aspect of having the environment in a global variable inside a js project, so any suggestion is still welcome
I work still view month on a project with RapidClipse 4.0
I deployed on the production server several versions of the project war files. Everything worked fine.
After the last deployment I got a blank screen after loading the application URL
For the server I use a docker container with following setting:
Apache Tomcat/8.5.43, JVM: 1.8.0_222-b10, 3.10.105, amd64
My first thought was: "ok you did something wrong in your code.. turn back and every thing is fine.... :-((
It wasn't !!
I used several versions which runs fine befor.
I stopped the application, redeployed it and deleted it.
Then I deployed an older version....and once again a version older..a.s.o
Non of the versions which worked fine befor did work again.
I got every time the same result: after loading the application a blank white screen.
So far so bad:
I tried to look into ../conf/server.xml if deployment parameter is set correctly:
<Host name="localhost" appBase="webapps"
unpackWARs="true" autoDeploy="true">
looked fine!
I enhanced the cache by:
$CATALINA_BASE/conf/context.xml added following code:
<Resources cachingAllowed="true" cacheMaxSize="100000" />
also without success.
I tried to look in catalina.out: There is still nothing helpfull:
14-Aug-2019 20:29:21.087 INFO [http-nio-8080-exec-6] org.apache.catalina.startup.HostConfig.deployWAR Deploying web application archive [/usr/local/tomcat/webapps/RC_07.war]
14-Aug-2019 20:29:31.190 INFO [http-nio-8080-exec-6] org.apache.catalina.startup.HostConfig.deployWAR Deployment of web application archive [/usr/local/tomcat/webapps/RC_07.war] has finished in [10,102] ms
after debug in browser I got following:
257ms Processing time was 134ms
257msReferenced paintables: 6
283msEstablishing push connection
300msCould not load theme from http://myIP:8888/RC_07/VAADIN/themes//styles.css?v=7.7.13
310msPush connection established using long-polling
I searched also in the history of the docker container an found that this problem (300ms....) still persists from beginning, over all versions I deployed before.
Out of this, I assume, that this could not be the reason, too.
Or am I wrong?
I searched around this VAADIN Problem and found a lot, but I was not able to solve it. The styles.css file are still in place on the server.
I am wondering on ..../VAADIN/themes//styles.css...
the double slash in error message.
But in my code I couldn't find similar.
Also the buildpath in eclipse includes the folder structure like expected.
Now I am at the end!
I am confused, how I should go ahead to figure out the reason for this behavior, or much better to fix it.
Any idea/ help would be welcome!!
Thank you in advance
rgds
OpaHeinz
After long research, togehter with RapidClipse support we found the solution.
I had two issues:
1) unknown how, we assume, that there was an error in the MainUI xml file.
After resetting with and high of view and layout, and turning back to setting before,
the page elements were visible again in design view.
2) there is a parameter for the theme under MainUI properties - misc -
This parameter was set, but without content. This results in a code line: this.setTheme("");
After resetting this, a deployment was possible like before.
Everything is fine now.
Thanks again to RapidClipse support.
I have created an authentication module which was build against
ejabberd 16.02 and runs fine when auth_method is set. It also works against
16.03.
However, from 16.04 onwards it gives me the error "[error] ignoring
option 'auth_method' with invalid value: [jwt]"
I checked the code diff between those releases and the only change
seems to be to the mod_pubsub.erl file, specifically adding the
following:
ServerHost = serverhost(Host),
+ ejabberd_hooks:run(pubsub_subscribe_node, ServerHost,
+ [ServerHost, Host, Node, Subscriber, SubId]),
https://github.com/processone/ejabberd/commit/639c2fb6401391663206c0e4c946d1a699689ac7
I have tried disabling this module and even deleting the beam file as
i don't use it, but i can't seem ti get round it.
Does anyone have any insight as to why these changes will have broken
my authentication module?
My source is at the link below, but as i say has worked fine for a year:
https://github.com/ParamountVentures/ejabberd-auth-jwt
The answer is that from 16.04 onwards you need to drop the .erl file into the ejabberd src folder and compile it with the source. Dropping in the .beam file to use alternative authentication modules no longer works.
I have gone through this post to run an application in production, but facing an issue.
https://www.ibm.com/developerworks/community/blogs/dhuyvett/entry/the_dojo_library_in_worklight_studio_v6_0?maxresults=15&lang=en
I have created a Worklight 6 hybrid application, listed below is a series of tasks that I performed on it.
I built and deployed the project with the 'Provide Library Resources' checked and I get a list of missing files (a few stated below) in the dojo library requests console and the application works fine.
[[2013-11-14 11:02:48] Application 'SampleBankingApp' requested a
missing resource. Providing library resource:
/dojoLib/toolkit/dojo/dijit/form/DateTextBox.js
[2013-11-14 11:02:48] Application 'SampleBankingApp' requested a missing resource. Providing library resource:
/dojoLib/toolkit/dojo/dijit/form/FilteringSelect.js
[2013-11-14 11:02:48] Application 'SampleBankingApp' requested a missing resource. Providing library resource:
/dojoLib/toolkit/dojo/dijit/Calendar.js]
I copied the missing files into the www folder keeping the folder structure as required, then I built and deployed the app with the 'Provide Library Resources' checked and I still get the same list of missing files in the dojo library requests console and the application also works fine.
Now I built and deployed by unchecking 'Provide Library Resources', when I tried to use the app it doesn't work, and the chrome console (where i use the mobile browser simulator) gives a series of errors like Failed to load resource: the server responded with a status of 404 (Not Found).
Whatever missing files it states in the dojo library requests console is in place inside the www folder, don't know whats going wrong.
I am using Eclipse Juno + Worklight6 + Dojo, any help would really be appreciated.
I think you may be failing on copying the resources to the correct path in the "www" folder.
For example, if you are missing /dojoLib/toolkit/dojo/dijit/form/DateTextBox.js, then you should copy that file into your_project/www/dijit/form
I am implementing the bundling and minification support in MVC4 and it appears as though it is making my javascript files bigger than if they weren't bundled/minified. I am using the latest build available in nuget (pre-release option on). I have the following bundle set up in my RegisterBundles class.
bundles.Add(new ScriptBundle("~/bundles/baseJS").Include(
"~/Scripts/jquery-1.7.1.js",
"~/Scripts/jquery.cookie.js",
"~/Scripts/jquery-ui-1.8.11.js",
"~/Scripts/bootstrap.js",
"~/Scripts/jquery.pjax.js",
"~/Scripts/kendo/2012.1.515/js/kendo.all.min.js",
"~/Scripts/jquery.jstree.js",
"~/Scripts/jquery.unobtrusive-ajax.js",
"~/Scripts/jquery.validate.js",
"~/Scripts/jquery.validate.unobtrusive.js",
"~/RIS.Scripts/PostJson.js"));
And I am loading it into my _Layout.cshtml using
#Scripts.Render("~/bundles/baseJS")
When I add up the bytes received in Fiddler for these scripts in debug mode I get the following
Name Size(bytes)
jquery 98013
jquery cookie 1455
jquery ui 124704
bootstrap 52378
pjax 8138
kendo.all 219751
jstree 55045
unobtrusive-ajax 2492
validate 13323
validate-unobtrusive 5138
postjson 634
Total 581071
And when I run it on my production server I get the following from fiddler for the entire js bundle.
Bytes Received: 999,396
What is going on here? Most of the files are minified to some extent, but it shouldn't almost double the size of my payload.
Additional details-
When I download the js files off my local dev box (fiddler reported size 379kb) and the server (fiddler reported size 999kb) and put them in kdiff they are binary identical. When I look in Chrome's developer tools network tab, the local server downloads 379kb, but the 'Parser' value is 975kb. What is this parser value. Could it be that there is some IIS compression setting that is not set in my server but is on my local IIS server? The only difference I note is the fact that the IIS Express I am running on my dev machine is 8.0 where the server is IIS 7.5.
Most likely what you are seeing here is some of the debug/release 'magic' that comes from the FileExtensionReplacementList.
Let's take jQuery for example. Typically in your scripts folder you will see two copies of each file, i.e. jquery-1.6.2.js and jquery-1.6.2.min.js.
By default optimization will use the min.js version when debug=false, and use the regular jquery-1.6.2.js when debug=true, since typically that makes debugging easier (no bundling and no minification of the bundle).
This file selection 'magic' is controlled via the FileExtensionReplacementList on BundleCollection.
In the next release (RTM), there will be a bit more granularity in this list, as typically developers will want to target when these are should be used, i.e.
list.Add("min", OptimizationMode.WhenEnabled);
list.Add("debug", OptimizationMode.WhenDisabled);
You have the bundling option working, but the minification is done by an BundleTable.EnableOptimizations = true setting and some "transform" options that you've haven't engaged. See CssMinify and JsMinify.
Something along the lines of:
var b1 =new ScriptBundle("~/bundles/jquery").Include(
"~/Scripts/jquery-1.*");
b1.Transforms.Add(new JsMinify());
bundles.Add(b1);
- and -
BundleTable.EnableOptimizations = true;