In general, the update of already installed SKOOR Engine and/or Collector packages from the repository are very simple:

dnf clean all

dnf update eranger'*'

Quick guide

The following procedure can be used as a checklist for a standard update of the software:

1. Review the required rpm repositories and Redhat subscriptions

2. Check if full and config backups are available. However, it is always a good idea to take a full backup before an update

3. Note the overall system state or take a screenshot of the state bar at the bottom of the SKOOR Engine UI

4. Set overall maintenance on all collectors related to the update (external collectors might be left unchanged or updated later)

5. On replicated systems (primary/standby):

   1. Check replication state and take action if state is not ok

   2. Stop replication

   3. Perform standby server update

6. Perform (primary-) server update

7. Start replication if not running

8. Update external collectors if configured

9. Update Dashboard Viewer if configured
   

10. Check for new configuration files with ending .rpmnew and merge the content with the original files if required
    This can be done using:
    rpmconf -a

11. Verification

    1. System overall state is the same as before or differences can be explained

    2. Replication is working as before

    3. Dashboards / Dashboard Viewer is accessible

    4. Alarms are sent (mail, SMS, etc.)

12. End maintenance

13. Create a new full backup

Quick guide details

Standard path to backups:

/opt/eranger/server/backups

Run full backup:

cd /var/lib/pgsql

sudo -u postgres /opt/eranger/bin/eranger-server-backup.sh full

Set overall maintenance on all collectors related to the update in /root /Collectors

Check replication state and stop standby:

/opt/eranger/bin/eranger-server-replication.pl status

<standby-ip>
<standby-ip> checking ssh for user reranger
<primary-ip> is running
Last read on master(<primary-ip>) 0 seconds ago

/opt/eranger/bin/eranger-server-replication.pl stopstandby

GPG fingerprints:

RPM-GPG-KEY-SKOOR:

RPM-GPG-KEY-PGDG:

RPM-GPG-KEY-TIMESCALEDB:

Update to SKOOR 7.2

Each SKOOR collector uses at least one directory to store temporary measurement files. With SKOOR 7.2, all collector temp directories will be cleaned up regularly:

• Files older than 72h are deleted

• Empty directories older than 24h are deleted

Since the collector sends all measurement data to the server after each job run, there is no reason to keep this files longer in the temporary files directory. However, if files such as scripts or other persistent data was created in one of these directories, it must be moved to a different location before upgrading to 7.2.

Use the following command to get a list of a collectors temporary directories (to be run on the local as well as all external collectors):

grep -oP "server\\d+_fetch_parse_dir\\s*=\\s*/var/opt/run/eranger/collector/.+" /etc/opt/eranger/eranger-collector.cfg | perl -pe 's/server\d+_fetch_parse_dir\s*=\s*//g'

Please make sure that all files in these directories can safely be cleaned up.

Update to SKOOR 7.0

The sdaas-webservice RPM package was replaced by the new package eranger-etl which is available also in the SKOOR RPM repository.

Please install eranger-etl after the upgrade to version 7.0 (erasing of sdaas-webservice must be allowed to avoid a conflict):

yum remove sdaas-webservice && yum install eranger-etl

Update to SKOOR 6.4

With version 6.4, the pgpool component is substituted with PgBouncer, a more lighweight connection pooler compared to the predecessor.

Please remove pgpool with the following command and install pgbouncer before or after the update:

dnf remove eranger-pgpool pgpool-II-13

dnf install eranger-pgbouncer