This was already fixed but we can do better by using SecuredUpload::isValidText
Also grabs some tokens to security.properties::deniedWebShellTokens from
DataEvents.java, and adds "alert" even it's harmless as is, as it's often used
by attackers to demonstrate issues, so will earn time.
Though an improvement, I backport as it's related to security
Conflicts handled by hand in data/DataEvents.java
By inserting malicious content in the “Text” field from
“/content/control/updateLayoutSubContent” -> “Templates”, an attacker may
perform SSTI (Server-Side Template Injection) attacks, which can leverage
FreeMarker exposed objects to bypass restrictions and obtain RCE (Remote Code
Execution).
This fixes it by calling SecuredUpload::isValidText on the “Text” field content.
I'll check that there are no other attack opportunities...
Thanks: Mal Aware <awaremal@gmail.com> for reporting this post-auth vulnerabily
Conflicts handled by hand in ContentManagementServices.java
Adds some tokens in security.properties::deniedWebShellTokens
Updates SecurityUtilTest::webShellTokensTesting accordingly
# Conflicts handled by hand
framework/security/config/security.properties
framework/security/src/test/java/org/apache/ofbiz/security/SecurityUtilTest.java
Trivial change in DataServices.java: uses the SecuredUpload import in 2 places
it was missing.
Adds some tokens in security.properties::deniedWebShellTokens
Removes a duplicated htaccess in security.properties::deniedFileExtensions
Last commits put in an issue in DataServices.java and GroovyBaseScript.groovy.
I should have used SecuredUpload::isValidFileName with dataResourceName and
SecuredUpload::isValidFile with objectInfo. This fixes that.
Also fixes formatting checkstyle issues in SecuredUpload class and move allow
all and check line length in right places (it's about content)
In security.properties, adds some deniedFileExtensions and improves few comments
(easier to merge later)
In DataServices::createFileNoPerm filters file upload to verify the filename
before the content is checked if necessary
Manually merges SecuredUpload.java based on trunk
In previous comment, to prevent linked images inside SVG I used xlink:href
That's no good, it must be followed by <<="http">>, this does it. Else a lot
of SVP images would not be loadable. Because xlink:href can be used in other
cases.
I reopened, forgot that you need to set -PjvmArgs, as I commented before:
<<For a 2GB+ file you need to set Xmx to more than the double of the file size,
eg: gradlew ofbiz -PjvmArgs="-Xms1024M -Xmx5048M">>
This also adds information in security.properties:
<<By default (OOTB) OFBiz is protected against Large File Denial of Service>>
After reviewing https://docs.oracle.com/cd/B14099_19/web.1012/b14014/jspxml.htm,
since we use "<jsp:" this removes "scriptlet>", "declaration>" and "expression>".
They are redundant with "<jsp:"; "<jsp:" covers those already.
Also adds eval( for js and shrinks base64_decode to decode, and processbuilder
to process because of OFBIZ 12571
I always test the denied tokens against a lib of images with 33 600 elements (
3.99 GB). If an image contains the token it can't be in the denied list.
Simpler for now, later OFBIZ 12573 needs to be done
Also formatting improvement in DataResourceWorker.java: removes a blank lines
Thanks: Tim Allison <tallison@apache.org> for the nice exchanges :)
The 1st issue was due to use of processbuilder token. It has been added to
deniedWebShellTokens in security.properties by f2cf262 commit for OFBIZ 11948
The tokens function (for js) and class have been added since while browsing
https://github.com/tennc/webshell
As mention the related deniedWebShellTokens TODO comment: "TODO.... to be continued
with known webshell contents... a complete allow list is impossible anyway...
So, later a deeper review of Groovy sandbox possibilities will be done..
Thanks: Y4er for report
This fixes and improves file upload.
Refactor a bit SecuredUpload class and fixes:
Prevent double extensions
Check extensions
Improves getMimeTypeFromFileName() by checking is file exists
In isValidImageFile() finally use isValidText() bypassing encoding
Completely refactor deniedWebShellTokens in security.properties by re-adding
removed tokens in last commit and adding several new ones. I have still to check
encoded, encrypted and obfuscated webshells)
Modifies SecurityUtilTest::webShellTokensTesting accordingly
While at it better format commonsImagingSupportedFormats, deniedFileExtensions
and imagejSupportedFormats properties for legibility.
Check if createAnonFile service is used in GroovyBaseScript.groovy and if a
complete file name is used (file exist) check, using SecuredUpload, extensions
and prevent double extensions, actually check all the file and stop there.
Adds some "Check if a webshell is not uploaded" comments in
ContentManagementServices.java
DataServices.java
ScaleImage.java
FrameImage.java
ImageManagementServices.java
ProductServices.java
HttpRequestFileUpload.java
ProgramExport.groovy
Trivial comments fixes in catalog.properties
Also fixes OFBIZ 12571 (I use here a space because of pending INFRA 22843)
by simply adding processbuilder to deniedWebShellTokens
Conflicts handled by hand
framework/security/config/security.properties
framework/security/src/main/java/org/apache/ofbiz/security/SecuredUpload.java
framework/security/src/test/java/org/apache/ofbiz/security/SecurityUtilTest.java
framework/service/src/main/groovy/org/apache/ofbiz/service/engine/GroovyBaseScript.groovy
framework/webtools/groovyScripts/entity/ProgramExport.groovy (reverted)
As reported by Michael, secretRequired value must be false because secret value
is empty. Else a notifying message appears in log saying that AJP is not available.
This uncomment out secretRequired, so its value is now false, and document more
notably about that. I'll later add more information in the
https://cwiki.apache.org/confluence/display/OFBIZ/Keeping+OFBiz+secure wiki page
Thanks: Michael for report
In SecuredUpload::isValidImageFile I initially used isValidText() and thought
that decoding would be better so finally used isValidTextFile() instead. But
then valid images files are not passing. So this replaces isValidTextFile by
isValidText there.
Also while at it removes few other PHP tokens, now useless (hopefully, I have
still to check encoded and encrypted PHP webshells), from
security::deniedWebShellTokens. The less tokens we have the better the whole is
legible.
Improves related comments.
Modifies SecurityUtilTest::webShellTokensTesting accordingly
Conflicts handled by hand
framework/security/config/security.properties
framework/security/src/test/java/org/apache/ofbiz/security/SecurityUtilTest.java
Previous commit added one checkstyle issue to the already 39 still existing in
R18, not a big deal. This just allows checkstyle to pass (maxErrors from 38514
to 38515)
Lion Tree <liontree0110@gmail.com> has reported us that
"CVE-2020-1938 is not fully fixed".
Though it was fixed by OFBIZ-11407, it still possible for an authenticated user
to upload a webshell included in an image using one of the upload possibilities
in OFBiz. That is not new and covered by OFBIZ-12080 "Secure the uploads", but
was still incomplete.
This enforces the secured uploads by
* checking in SecuredUpload::isValidImageFile that a webshell is not embedded in
an image.
* Keeping only "<%" as a denied token for JSP webshells, instead of currently
"<%@ page"
* Adds "application/text/x-ruby" to SecuredUpload::isExecutable
Also
* Adds "<jsp", and "<?" for PHP. Even if OFBiz does not use PHP at all,
it's often installed on servers.
* Removes "import=\"java" and "runtime.getruntime().exec(". They are no
longer useful since "<%" and "<jsp" block them.
* Remove php token since I'll put "<?" in.
* Adds "#!", rather than adding other shebangs like perl,python and ruby
This will make deniedWebShellTokens more understandable.
But I'm conscious that despite SecuredUpload::isExecutableI I still need to
better handle encoded webshells. I'll do that soon in a second approach.
I'll also certainly more prune PHP related tokens.
Thanks: Lion Tree for report
When you define a widget form and wish populate hidden fields with the element auto-fields-service, OFBiz rendering each as edit field type.
<form name="ListPartyQuals" ...>
<auto-fields-service service-name="updatePartyQual" default-field-type="hidden"/>
</form>
The problem came from the function ModelFormFieldBuilder.induceFieldInfoFromServiceParam didn't manage the case "hidden"
The view definition for ViewBinaryDataResource is missing.
Goto: /content/control/EditDataResource?dataResourceId=GZ-DIG
Click the "Download"-Button.
Thanks: Ingo Wolfmayr for the patch
Conflicts handled by hand
The manufacturing jobshop find screen has the seach date initialized to nowdate.
if one wants to see all production runs, one has to clear the date which should
be empty initially
The same problem exists on the MRP log find where the date is filled by default.
Thanks: Hans for report and patches
Severity: high
Description:
There's a vulnerability within the Apache Xerces Java (XercesJ) XML
parser when handling specially crafted XML document payloads. This
causes, the XercesJ XML parser to wait in an infinite loop, which may
sometimes consume system resources for prolonged duration. This
vulnerability is present within XercesJ version 2.12.1 and the
previous versions.
Mitigation:
Apache XercesJ users, should migrate to version 2.12.2
Credit:
This issue was discovered by Sergey Temnikov and Ziyi Luo, from Amazon
Corretto/JDK Team
References:
https://markmail.org/message/vcmhwbuorfgcdr6l
jleroux: I already removed the "apache-" prefix but for a reason did not commit
Severity: high
Description:
There's a vulnerability within the Apache Xerces Java (XercesJ) XML
parser when handling specially crafted XML document payloads. This
causes, the XercesJ XML parser to wait in an infinite loop, which may
sometimes consume system resources for prolonged duration. This
vulnerability is present within XercesJ version 2.12.1 and the
previous versions.
Mitigation:
Apache XercesJ users, should migrate to version 2.12.2
Credit:
This issue was discovered by Sergey Temnikov and Ziyi Luo, from Amazon
Corretto/JDK Team
References:
https://markmail.org/message/vcmhwbuorfgcdr6l
The issue exists also in the older version.
The field "requirementByDate" form "FindApprovedProductRequirements" in
RequirementForms.xml has been set wrongly and it should be "requiredByDate".
This causes the result could not be queried out properly as indicated as below:
The result is governed by the service "getRequirementsForSupplier",
where reads the list from the view "RequirementAndRole".
In the view entity definition, the field is defined as
<alias entity-alias="RQ" name="requiredByDate"/>,
but it is not the case in the form "FindApprovedProductRequirements" above.
After changing it to the correct name, the bug is solved.
jleroux: Same exists in OrderEntryForms::FindRequirements
Thanks: Schumann Ye for report Pierre Smits for creating the initial PR494
The fix for bug CVE-2020-9484 introduced a time of check, time of use
vulnerability that allowed a local attacker to perform actions with the
privileges of the user that the Tomcat process is using. This issue is only
exploitable when Tomcat is configured to persist sessions using the FileStore.
Adding testIntegration in the GH Gradle process does not work.
For a reason I'll not dig in, it ALWAYS fails at:
Cannot find service [indexProductsFromContent] location class
(org.apache.ofbiz.content.search.SearchServices)
This complements previous commit:
<<After another issue reported today by Steven Payne on user ML about the
execution policy of PowerShell7, rather than using the pwsh path I document in
README.adoc solutions for cases we crossed so far.>>
with also trivial formatting.
Thanks: David S Hunter and Steven Payne for reports on user ML
After another issue reported today by Steven Payne on user ML about the
execution policy of PowerShell7, rather than using the pwsh path I document in
README.adoc solutions for cases we crossed so far.
Moreover I noticed today that my previous fix was not enough. "pwsh" was used
twice in the cmd line!
Thanks: David S Hunter and Steven Payne for reports on user ML
This complete the previous Improvement, Rendering different HTML container types with ScreenRenderer (OFBIZ-10495), where the xml macro are forgotten.
In same time resynchronise also missing macro renderColumn*
Because of Windows EOL difference with *nix, despite the number for maxErrors
being more around 39600+ we temporarily put 50000 here to avoid issue when
releasing. Just increasing the number of lines is the problem.
The problem was already fixed in trunk by allowing all types of EOL, ie using
NewlineAtEndOfFile value="lf_cr_crlf"
Because of Windows EOL difference with *nix, despite the number for maxErrors
being more around 39600+ we temporarily put 50000 here to avoid issue when
releasing. Just increasing the number of lines is the problem.