CODEWITHSUNDEEP
marklogicmarklogic-10
Arunjay
Arunjay

Reputation: 71

Marklogic start Issue after Upgrading MarkLogic 10.0-9.5 to MarkLogic 11.3.1 in RHEL 8.8

I'm unable to start MarkLogic 11.3.1 after upgrading from MarkLogic 10.0-9.5 in AWS, and I'm encountering the following error. But Marklogic 10.0-9.5 run perfectly in the same AWS instance.

OpenJDK 64-Bit Server VM warning: Option UseConcMarkSweepGC was deprecated in version 9.0 and will likely be removed in a future release.

Set configuration: JAVA_HOME="/usr/lib/jvm/java-11"

Set configuration: MARKLOGIC_MDB_TYPE=""

Set configuration: AWS_REGION="us-east-1"

Set configuration: AWS_DEFAULT_REGION="us-east-1"

Set configuration: MARKLOGIC_ZONE="us-east-1b"

Set configuration: MARKLOGIC_HOSTNAME="xx.xx.xx.xxx"

Initialize Configuration. AWS Region: us-east-1, ZONE: us-east-1b. INSTANCE: xxxxxxxxxx

OpenJDK 64-Bit Server VM warning: Option UseConcMarkSweepGC was deprecated in version 9.0 and will likely be removed in a future release.

Instance is not managed

Waiting for device mounted to come online : /dev/xvdf

Volume /dev/sdf has failed to attach - aborting

Warning: ec2-startup did not complete successfully

Check the error logs for details

this script is giving message and aborted ML startup: enter image description here

Steps taken to upgrade:

  1. Instance type: m4.xlarge
  2. No volume was attached at the time of upgrade; only MarkLogic was upgraded.
  3. We simply followed the steps to uninstall MarkLogic 10.0-9.5 and install MarkLogic 11.3.0.
  4. Start MarkLogic using the command /etc/init.d/MarkLogic start.
  5. We encountered the above mentioned error.

+++++++++++++++++++++++++++++++++

Also we are trying to upgrade another system but getting same issue.

Set configuration: JAVA_HOME="/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.382.b05-2.el8.x86_64"

Set configuration: MARKLOGIC_MDB_TYPE=""

Set configuration: AWS_REGION="us-east-1"

Set configuration: AWS_DEFAULT_REGION="us-east-1"

Set configuration: MARKLOGIC_ZONE="us-east-1a"

Set configuration: MARKLOGIC_HOSTNAME="ip-10-97-21-232.ec2.internal"

Initialize Configuration.

AWS Region: us-east-1, ZONE: us-east-1a. INSTANCE: i-00b57d0c9c5216bb4

Instance is not managed

Waiting for device mounted to come online : /dev/nvme1n1

UUID dev mapping:

Final dev mapping: /dev/nvme1n1

Attempting to mount EBS device /dev/sdf OS device /dev/nvme1n1 on

mountpoint /var/opt/MarkLogic

Directory /var/opt/MarkLogic has contents - skipping

Failed to mount EBS volume - aborting Warning: ec2-startup did not complete successfully

Check the error logs for details

Upvotes: 3

Views: 218

Answers (1)

Mads Hansen
Mads Hansen

Reputation: 66781

There is logic in the startup scripts that look to determine whether a host is running in EC2 and is part of a managed cluster.

In older versions of MarkLogic, it was only setting the EC2 flag if you were running Amazon Linux.

In order to prevent some of the managed node startup logic from firing, you can explicitly configure the host as NOT being a managed node.

Create a /etc/marklogic.conf file and set to false

export MARKLOGIC_MANAGED_NODE=0

That should get you past this startup issue. Although, when EC2 integration is enabled it will expect Java to be installed and will fail to start if it can't find Java. So, it could then hit on that startup issue. You can define JAVA_HOME in /etc/marklogic.conf:

export JAVA_HOME=/usr/lib/jvm/jre

Or you could disable EC2 integration by setting:

export MARKLOGIC_EC2_HOST=0

https://help.marklogic.com/Knowledgebase/Article/View/best-practice-editing-marklogic-server-environment-variables

Upvotes: 1

Related Questions