[jira] [Comment Edited] (JCRVLT-341) https doesn't work anymore

["Tobias Bocanegra (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/JCRVLT-341?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16886697#comment-16886697 ] 

Tobias Bocanegra edited comment on JCRVLT-341 at 7/17/19 6:00 AM:
------------------------------------------------------------------

[~tripod]

If I change the url pattern as your proposing, I get the following error

{noformat}
{
"status": "error",
 "message": "Error while executing 'start': Login failed: Invalid workspace name ''."
}
{noformat}

I debugged the code a little bit and I think I found the error. It is during the login in the source repository. 

In the following jar "org.apache.jackrabbit.vault.rcp-3.2.8.jar" includes 

{noformat}
<!-- https://mvnrepository.com/artifact/org.apache.jackrabbit/jackrabbit-spi2dav -->
<dependency>
 <groupId>org.apache.jackrabbit</groupId>
 <artifactId>jackrabbit-spi2dav</artifactId>
 <version>2.16.3</version>
</dependency>
{noformat}

That jar has the class "org.apache.jackrabbit.spi2dav.RepositoryServiceImpl" and on line 231 of this class,  a HttpHost is instanciated. If the schema is not a parameter of the constructor, the constructor assumes "http". So the login fails for this url "*http*://@host:port". The login-address should be "*https*://@host:port".

{code}
try {
 URI repositoryUri = computeRepositoryUri(uri);
 httpHost = new HttpHost(repositoryUri.getHost(), repositoryUri.getPort()*, repositoryUri.getScheme()*);

 nsCache = new NamespaceCache();
 uriResolver = new URIResolverImpl(repositoryUri, this, DomUtil.createDocument());
 NamePathResolver resolver = new NamePathResolverImpl(nsCache);
 valueFactory = new ValueFactoryQImpl(qValueFactory, resolver);

} catch (URISyntaxException e) {
 throw new RepositoryException(e);
} catch (ParserConfigurationException e) {
 throw new RepositoryException(e);
}
{code}

Do you have the possibility to fix that and send me a patch so I can test it?

Thanks a lot in advance.

Regards

Walter


was (Author: ch00stw):
[~tripod]

If I change the url pattern as your proposing, I get the following error

{

"status": "error",
 "message": "Error while executing 'start': Login failed: Invalid workspace name ''."
}

I debugged the code a little bit and I think I found the error. It is during the login in the source repository. 

In the following jar "org.apache.jackrabbit.vault.rcp-3.2.8.jar" includes 

<!-- https://mvnrepository.com/artifact/org.apache.jackrabbit/jackrabbit-spi2dav -->
<dependency>
 <groupId>org.apache.jackrabbit</groupId>
 <artifactId>jackrabbit-spi2dav</artifactId>
 <version>2.16.3</version>
</dependency>

 

That jar has the class "org.apache.jackrabbit.spi2dav.RepositoryServiceImpl" and on line 231 of this class,  a HttpHost is instanciated. If the schema is not a parameter of the constructor, the constructor assumes "http". So the login fails for this url "*http*://@host:port". The login-address should be "*https*://@host:port".

<<<<<<<<<<<<<<

try {
 URI repositoryUri = computeRepositoryUri(uri);
 httpHost = new HttpHost(repositoryUri.getHost(), repositoryUri.getPort()*, repositoryUri.getScheme()*);

 nsCache = new NamespaceCache();
 uriResolver = new URIResolverImpl(repositoryUri, this, DomUtil.createDocument());
 NamePathResolver resolver = new NamePathResolverImpl(nsCache);
 valueFactory = new ValueFactoryQImpl(qValueFactory, resolver);

} catch (URISyntaxException e) {
 throw new RepositoryException(e);
} catch (ParserConfigurationException e) {
 throw new RepositoryException(e);
}

>>>>>>>>>>>>>>

Do you have the possibility to fix that and send me a patch so I can test it?

Thanks a lot in advance.

Regards

Walter

> https doesn't work anymore
> --------------------------
>
>                 Key: JCRVLT-341
>                 URL: https://issues.apache.org/jira/browse/JCRVLT-341
>             Project: Jackrabbit FileVault
>          Issue Type: Bug
>          Components: RCP, vlt
>    Affects Versions: 3.1.44, 3.2.6, 3.2.8
>         Environment: source instance is suse linux sles 12
> target instance is suse linux sles 12 and windows 10
> java 1.8
> AEM 6.4.4
>            Reporter: WalterSteiner
>            Priority: Major
>
> org.apache.jackrabbit.vault.rcp-3.1.24 introduced support of https. I now upgraded to the version 3.2.8. I want to transfer with the following path "/content/etc/robots.txt"  from one aem-author instance to another. I get the following log entry
> 16.07.2019 10:52:52.827 *ERROR* [Vault RCP Task - copy-contentetcrobots.txt] org.apache.jackrabbit.vault.util.RepositoryCopier Error while retrieving src node /content/etc/robots.txt: javax.jcr.PathNotFoundException: /content/etc/robots.txt
> In the source, this path exists. If I do the same with http. The path is found and the content transfered.
> I use the following json to create the tast in aem
> {code}
> {
>  "cmd":"create",
>  "id":"copy-contentetcrobots.txt",
>  "src":"https://user:password@host:port/crx/server/crx.default/jcr:root/content/etc/robots.txt",
>  "dst":"/content/etc/robots.txt",
>  "batchsize": 2048,
>  "update": true,
>  "onlyNewer": true,
>  "recursive": true,
>  "throttle": 1,
>  "excludes": []
> }
> {code}
> Please fix this issue, because we are no longer allowed to use http in our company.
>  
>  



--
This message was sent by Atlassian JIRA
(v7.6.14#76016)