On Mon, 2019-02-25 at 18:53 -0500, marc.davenp...@oracle.com wrote:
> I've been trying to rationalize the hit or miss nature of this
> problem.
> In my mind, I'm looking for a race condition. I'm going to bet that
> this is a configuration issue on our end, but I'm not sure how this
> would
Julian,
Thanks for even the wildest of speculation. I can use all the help I can
get right now. It looks like the slingevent namespace we don't even
need. I inherited this project as a kind of mess. It looks like whoever
was migrating this from Sling 6 to Sling 10 had just included the
Hi Marc
>From your log, I suspect that the "sling" and "slingevent" Namespaces
are (not yet?) registered when the repository is being initialized.
During repository initialization all installed OSGi bundles are
inspected in order to see whether they contribute namespaces or
node-type definitions
Robert,
Thanks for thinking about this issue with me.
You asked: do you have any custom code that plugs into the repository
initialisation?
I can not find any custom classes that are implementing RepositoryInitilizer.
We have a custom class which provides the SlingRepository service and a
On Thu, 2019-02-14 at 16:02 -0500, marc.davenp...@oracle.com wrote:
> 14.02.2019 12:28:29.190 *WARN*[CM Event Dispatcher (Fire
> ConfigurationEvent:
> pid=org.apache.jackrabbit.oak.segment.SegmentNodeStoreService)]
> org.apache.sling.commons.scheduler.impl.QuartzScheduler Scheduler
> job
>
Hi Marc,
On Thu, 2019-02-21 at 18:18 -0500, marc.davenp...@oracle.com wrote:
> Hello Robert et all,
>
> Thank you for your reply. I have not been able to pair down to a
> sample
> project that isolates this issue. I'm currently deploying my
> application
> over and over until I hit the
Hello Robert et all,
Thank you for your reply. I have not been able to pair down to a sample
project that isolates this issue. I'm currently deploying my application
over and over until I hit the failure mode. I haven't identified the
code that would have created the missing primary types.
Hi Marc,
On Thu, 2019-02-14 at 16:02 -0500, marc.davenp...@oracle.com wrote:
> I've been moderately successful. However, about
> 1/2 the time, our application will fail to deploy. I can not parse
> if
> this is a Oak failure or a sling10 failure, or a configuration
> between
> the two issue.
Hello,
We are in the process of migrating from Sling6 to Sling10 (Sling11 came
out after we started the initiative). One of the upgrades is moving from
Jackrabbit 2 to Jackrabbit Oak. We've modeled ourselves after the
Sling10 starter war. The starter war uses Oak 1.6.8 so we started with
