A question about AEM performance, particularly workflow related
activities, including usability of the Workflow Admin interface.AEM 6.2
seems to disable by default the event.job.topic property index
(/oak:index/event.job.topic). The intent SEEMS to be to replace this
with the slingeventJob (/oak:index/slingeventJob) index.However, as the
volume of workflow data grows in a high volume environment the
slingeventJob index under performs the old property index by an order of
magnitude. Cost of the p...
AEM 6.2...I would like to identify nodes in the DAM which may have been
missed by a disabled DAM Update Asset workflow. These nodes would be
missing a child metadata and/or renditions.Executing the following two
queries returns identical results:1) SELECT * FROM [dam:Asset] AS s
WHERE ISDESCENDANTNODE([/content/dam/])2) SELECT * FROM [dam:Asset] AS s
WHERE ISDESCENDANTNODE([/content/dam/]) and [jcr:content/metadata] is
nullThe null check does not appear to be respected.The following query
return...
We have an AEM 6.1 SP1 sites deployment where we are seeing
exceptionally poor search performance using the OOTB
/libs/foundation/components/search component.The initial search is slow
against a very small hierarchy of pages. It returns less than 20 results
(in one example) which is paged 10/page. When navigating / paging back
and forth between page 1 and 2 of the results we see significant spikes
in CPU usage from 5% up to over 150%. Additionally, memory usage seems
to climb slightly with each ...
Just curious if you found a resolution to these messages in the log.
Seeing the same with an AEM 6.2 environment upgraded from AEM 6.1 with
Solr.Thanks.
I did also find this thread:
http://stackoverflow.com/questions/33129042/how-to-hide-hidden-files-in-aem-crxde-liteWhich
suggests using an OSGi scheduler to simply delete them regularly from
AEM.package com.foo.bar; import
org.apache.felix.scr.annotations.sling.SlingServlet; import
org.apache.sling.api.SlingHttpServletRequest; import
org.apache.sling.api.SlingHttpServletResponse; import
org.apache.sling.api.resource.Resource; import
org.apache.sling.api.resource.ResourceResolver; import
org.apac...
Thanks Mark.This looks promising if the users are willing to make this
configuration change. I don't have a MAC to test on either... Your
response is appreciated.
These files get created at each level in the repository when a user
creates content (uploads files) via WebDAV to the DAM. There are
actually two files I think... .DS_Store and _.DS_Store...Is there a way
to avoid this / correct it?Thanks, Rob.
Does anyone from Adobe know why the Offloading documentation is
truncated?https://docs.adobe.com/docs/en/aem/6-2/deploy/configuring/offloading.html
Any chance of getting it fixed? Thanks,Rob.
Thanks very much for this helpful response. I have confirmed this
re-enables this functionality on my environment.In my particular case I
want to serve these files as an interim step in a migration effort so I
am not concerned about the security concerns. I will recheck the box
once completed.Cheers.
I have the same issue. I have a package containing static HTML files.I
import this into AEM 6.1 and it is rendered to the browser.Importing the
exact same package into AEM 6.2 results in the files being downloaded
instead of served into the browser.I attempted to add /content/path and
also /content/path/somefile.html to the Apache Sling Content Disposition
Filter as suggested without seeing any difference in behavior.I'm
curious if you had made any other changes to get this working. Thanks.
Thanks for the response.I had already corrected dependency issues,
replaced deprecated WCMUse class, etc. That all went well.The issue was
finally resolved after root cause was traced to
TooManyCallsExceptions...The TooManyCallsException is thrown by the
Sling implementation if too many inclusions have been called for during
a single request.This setting can be configured from the Sling Main
Servlet configuration under Felix console.
http://:/system/console/configMgr/org.apache.sling.engine.impl...
We are experiencing issues with Sightly components in a custom solution
after upgrading to AEM 6.2. We can reproduce the issue on a clean AEM
6.2 server (not upgraded). Unfortunately there are no error messages so
we do not have much to go on... I am describing here in case someone has
encountered this previously and has some tips to resolve or debug.I'll
describe the non-upgraded path to reproduce...We installed AEM 6.2
quickstart (author). Then deployed the we.retail sample package.
Everything...
We have an environment which was recently upgraded from AEM 6.1 to 6.2.
Intermittently for periods of 15mins to an hour the server becomes
nearly unresponsive. During these periods there are no errors in the
logs.However, I've noticed warnings related to idle sessions popping up
every 20-30 seconds which seems excessive. Is this related? Any other
thoughts?28.06.2016 15:33:40.203 *INFO* [10.240.1.3 [1467141961252] GET
/mnt/overlay/granite/ui/content/shell/header/actions/pulse.data.json
HTTP/1.1]...
Thanks for the response. I'm aware of the brand portal... however this
is a SAAS offering in the Cloud. I'm looking for a solution which can be
hosted in the AEM publish instance.
Is there a strait forward way to securely expose the /assets page on a
publish instance effectively creation a TouchUI version of an Asset
Share Page?The Asset Share Page documentation outlines an approach using
the classic UI which is not desirable.
https://docs.adobe.com/docs/en/aem/6-2/author/assets/using-asset-share-pages.htmlEssentially,
I'd like to expose an anonymous Asset search page on the publish
instance which mirrors the functionality of the /assets page (filter,
full text search, et...
Can someone confirm if this is a defect or a work-around for the
following special characters issue?Created a new MetadataSchema and
added a Dropdown object when building the form.The issue is that if you
source the values for the drop-down from a JSON object (set the JSON
Path in the field settings) then if the JSON contains certain special
characters it causes the form to fail.If the JSON source contains a
slash (/) or a colon (:) in the text property then the slash causes it
to create corrupt...
Customizing the "Assets Admin Search Rail" search form.Adding a new
options predicate with a "JSON Path" source results in a filter that
doesn't work.Specifically, add an "Options Predicate" and configure the
following settings:Field Label - whateverJSON Path - Path to json object
in repo (eg. /apps/appname/rights.txt) which as the following content://
Rights Management{"options": [ { "value": "royalty_free", "text":
"Royalty Free" }, { "value": "rights_managed", "text": "Rights Managed"
}, { "v...
Thanks for the response / confirmation.Is the implementation for
Metadata Schemas something you are familiar with? Just wondering if you
can save me some reverse engineering... what components / code should I
be looking at to implement this custom functionality?Any idea on level
of effort?We had already configured the custom metadata schema for our
specific folder in the DAM.Seems like something that should be available
OOTB, especially in multi-tenant scenarios where different groups with
vastl...
When you customize the "Assets Admin Search Rail" search forum the
changes you make (adding predicates for example) are visible in all
contexts in
Assets.https://docs.adobe.com/docs/en/aem/6-2/administer/operations/search-forms.htmlIs
there any way to have the custom predicates only show for specific paths
in the content tree? Eg. not visible in /content/dam/abc but visible in
/content/dam/defThanks.
Following documentation to add a custom search form (in my case for
Asset Admin Search Rail) I can create a custom configuration for a
search form. However, when it says to navigate to the "Search Panel" it
is no longer possible in 6.2 to access this panel.Configuring Search
FormsHow do you access the custom search form / search panel in the AEM
6.2 Touch UI? Thanks for any help. Rob
Thanks Opkar.I had just done some more reading and came to the same
conclusion. I was just about to answer my own question.Thanks again for
your reply.Rob.
Jason, Did you resolve your first issue?I have a situation where the
csrf token is empty ("{}") when I make an unauthenticated call to
/libs/granite/csrf/token.json. However, when making an authenticated
call I receive a token as expected: {"token":"ey....U0"} Thanks, Rob.
Looking for root cause of NoSuchNodeTypeExceptions which are preventing
AEM instances (two of them on the same server... Author/Publish) from
starting successfully.The exact pathology is a but unclear at this
point. The servers were shutdown, compaction was executed and failed.
Then when restarted these errors showed up and AEM was not
healthy.However, after restoring a backup from when the AEM instance was
healthy the same error occurs. And now the same shows up on the Publish
instance on the s...
More thoughts...I'm thinking this is actually related to the old version
of Apache (2.2.15) which is installed.I'm basing this on a thread here:
http://stackoverflow.com/questions/4390436/need-to-allow-encoded-slashes-on-apacheNot
sure if I need to just get Apache updated (I don't completely control
environment) or if I can just add some flavor of the following
directive...AllowEncodedSlashes NoDecodeAgain, input is welcome. Thanks.
On working Windows Dispatcher that line looks like this:10.11.19.54 - -
[12/Apr/2016:22:42:03 -0400] "GET
/bin/wcmcommand?cmd=open&path=%2Fcontent%2Fcountry-sites%2Fu%2Fen&_charset_=utf-8
HTTP/1.1" 302 -I can see the difference.... %2F vs %252F... but what is
the cause / resolution?
Additional Background:The offending environment:RedHatDispatcher
4.1.12Apache 2.2.15JDK 1.8_31 Attempted to reproduce
on:WindowsDispatcher 4.1.12Apache 2.2.25JDK 1.8_71After copying the
exact dispatcher configuration to a Windows platform and pointing to the
same AEM Author Instance (6.1 SP1 on Redhat) I was unable to
reproduce.Is the issue perhaps with the Dispatcher 4.1.12 on RedHat
specifically?
Thanks for the reply.The setup is a standard Author Dispatcher /filter
with allow all then deny filters. I believe /bin is permitted with this
filter. Additionally, the fact that the error mentioned is showing in
the AEM log would suggest it is getting through the Dispatcher./filter {
# allow everything /0001 { /type "allow" /glob "*" } # block admin tools
/0010 { /type "deny" /glob "* /admin/*" } /0011 { /type "deny" /glob "*
/admin *" } /0012 { /type "deny" /glob "* /system/*" } /0013 { /type
...
