Upgrade VMware data collection components
To collect VMware data in SAI, you have to install components in the Splunk VMware Add-on for ITSI from the Splunk IT Service Intelligence (ITSI) package. You can't use earlier versions of the Splunk Add-on for VMware.
The Splunk VMware Add-on for ITSI sends metrics to the vmware-perf-metrics
metrics index. Previous versions of the Splunk Add-on for VMware send data to the vmware-perf
events index. If you want to search historical event data in the vmware-perf
events index, include the vmware-perf
index in indexes.conf
. For more information, see Search historical data after you upgrade.
These steps show you how to upgrade the VMware data collection components from an earlier version of vmware_ta_itsi
and the Splunk Add-on for VMware version 3.4.x.
Stop the DCS before upgrading the DCS, DCN, and each indexer. After you upgrade everything, start the DCS.
Upgrade a Data Collection Scheduler
Replace the required components and delete old components that aren't necessary for VMware data collection in SAI on the Data Collection Scheduler (DCS).
Upgrade a DCS running an earlier version of vmware_ta_itsi
Follow these steps to upgrade a DCS that's running an earlier version of vmware_ta_itsi.
- If you haven't already, download the ITSI package from Splunkbase.
- In the ITSI user interface, click the Add Data tab and select VMware vSphere.
- Select the DCN tab and disable Enable Data Collection.
- Stop splunkd.
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-Hydra
andSplunk_TA_vmware
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. - Start splunkd.
$SPLUNK_HOME/bin/splunk start
Upgrade a DCS running the Splunk Add-on for VMware version 3.4.x
Follow these steps to manually upgrade a DCS that's running the Splunk Add-on for VMware version 3.4.x.
- If you haven't already, download the ITSI package from Splunkbase.
- Stop the DCS from the Collection Configuration page in the Splunk Add-on for VMware version 3.4.x.
- Stop splunkd.
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-Hydra
andSplunk_TA_vmware
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. - If any of these directories exist in
$SPLUNK_HOME/etc/apps
, remove them:SA-VMNetAppUtils
SA-VMWIndex
TA-VMW-FieldExtractions
Splunk_TA_vcenter
Splunk_TA_esxilogs
- In
$SPLUNK_HOME/etc/apps/SA-Hydra
, remove theappserver
directory. - In
$SPLUNK_HOME/etc/apps/SA-Hydra/bin/hydra
, remove theuuid.py
anduuid.pyc
files. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware
, remove theappserver
directory. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/bin
, remove thesuds
directory, and theuuid.py
anduuid.pyc
files. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/default
, remove thedata
directory. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/local
, removeinputs.conf
. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/local
, createpool.conf
if it doesn't exist already. - Open
pool.conf
. - Create a
[Global pool]
stanza. - Move only task-related parameters from the
[default]
stanza inta_vmware_collection.conf
to the[Global pool]
stanza you just created. This includetask
,atomic_tasks
,<task>_expiration
, and<task>_interval
. - Replace
otherperf
withclusterperf
. For example, renameotherperf_interval
toclusterperf_interval
. - Remove
hierarchyinv
andrpinv
parameters from thetask
andatomic_tasks
settings if they exist. - Remove
interval
andexpiration
parameters forhierarchyinv
andrpinv
if they exist. - Keep collection parameters other than task-related in
ta_vmware_collection.conf
. - If your
ta_vmware_collection.conf
file looks like this:[default] task = vminv, hostinv, otherperf, rpinv, hierarchyinv vminv_interval = 900 vminv_expiration = 900 hostinv_interval = 900 hostinv_expiration = 900 otherperf_interval = 1800 otherperf_expiration = 1800 hierarchyinv_interval = 1800 hierarchyinv_expiration = 1800 rpinv_interval = 1800 rpinv_expiration = 1800 deployment_type = Custom perf_format_type = normal host_instance_whitelist = .*ahegd.*
Move these parameters to
pool.conf
:[Global pool] task = vminv, hostinv, clusterperf vminv_interval = 900 vminv_expiration = 900 hostinv_interval = 900 hostinv_expiration = 900 clusterperf_interval = 1800 clusterperf_expiration = 1800
These are the remaining parameters in
ta_vmware_collection.conf
:[default] deployment_type = Custom perf_format_type = normal host_instance_whitelist = .*ahegd.*
- In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/local
, make these changes: - Open
hyrda_node.conf
and addpool_name = Global pool
to every stanza. - Open
ta_vmware_collection.conf
and addpool_name = Global pool
to every stanza except thedefault
stanza. - Start splunkd.
$SPLUNK_HOME/bin/splunk start
Upgrade a Data Collection Node
Replace the required components and delete old components that aren't necessary for VMware data collection in SAI on the Data Collection Node (DCN).
Upgrade a DCN running an earlier version of vmware_ta_itsi
Follow these steps to upgrade a DCN that's running an earlier version of vmware_ta_itsi.
- If you haven't already, download the ITSI package from Splunkbase.
- Stop splunkd.
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-Hydra
,Splunk_TA_vmware
,Splunk_TA_esxilogs
, andSplunk_TA_vcenter
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. - Start splunkd.
$SPLUNK_HOME/bin/splunk start
Upgrade a DCN running the Splunk Add-on for VMware version 3.4.x
Follow these steps to manually upgrade a DCN that's running the Splunk Add-on for VMware version 3.4.x.
- If you haven't already, download the ITSI package from Splunkbase.
- Stop splunkd.
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-Hydra
,Splunk_TA_vmware
,Splunk_TA_esxilogs
, andSplunk_TA_vcenter
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. - If any of these directories exist in
$SPLUNK_HOME/etc/apps
, remove them:SA-VMNetAppUtils
SA-VMWIndex
TA-VMW-FieldExtractions
- In
$SPLUNK_HOME/etc/apps/SA-Hydra
, remove these directories:local
appserver
- In
$SPLUNK_HOME/etc/apps/SA-Hydra/bin/hydra
, remove theuuid.py
anduuid.pyc
files. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware
, remove theappserver
directory. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/bin
, remove thesuds
directory, and theuuid.py
anduuid.pyc
files. - In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/local
, remove these files:inputs.conf
ta_vmware_cache.conf
hydra_metadata.conf
hydra_session.conf
- In
$SPLUNK_HOME/etc/apps/Splunk_TA_vmware/default
, remove thedata
directory. - Start splunkd.
$SPLUNK_HOME/bin/splunk start
Upgrade an indexer
Replace the required components and delete old components that aren't necessary for VMware data collection in SAI on each indexer. You can follow these steps if you're running a standalone indexer, a distributed indexer, or a clustered indexer environment.
If you want to continue searching data you already collected from the Splunk Add-on for VMware version 3.4.x, add the old index to the new indexes.conf
in SA-VMWIndex
during the upgrade process. If you don't include the old index before you start splunkd after upgrading, the indexer deletes the index. For more information, see Search historical data after you upgrade.
For more information about installing the add-on across an indexer cluster, see Update common peer configurations and apps in the Managing Indexers and Clusters of Indexers guide.
Upgrade an indexer running an earlier version of vmware_ta_itsi
Follow these steps to upgrade an indexer that's running an earlier version of vmware_ta_itsi.
- If you haven't already, download the ITSI package from Splunkbase.
- Stop splunkd:
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-VMWIndex
,Splunk_TA_esxilogs
, andSplunk_TA_vcenter
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. If you want to continue monitoring historical data, add the old index to the newindexes.conf
inSA-VMWIndex
. If you don't include the old index, the indexer deletes the index. See Search historical data after you upgrade. - Start splunkd:
$SPLUNK_HOME/bin/splunk start
Upgrade an indexer running the Splunk Add-on for VMware version 3.4.x
Following these steps to update an indexer that's running the Splunk Add-on for VMware version 3.4.x.
- If you haven't already, download the ITSI package from Splunkbase.
- Stop splunkd:
$SPLUNK_HOME/bin/splunk stop
- Copy the
SA-VMWIndex
,Splunk_TA_esxilogs
, andSplunk_TA_vcenter
directories from thevmware_ta_itsi
parent directory in the ITSI package to$SPLUNK_HOME/etc/apps
. If you want to continue monitoring historical data, add the old index to the newindexes.conf
inSA-VMWIndex
. If you don't include the old index, the indexer deletes the index. See Search historical data after you upgrade. - If any of these directories exist in
$SPLUNK_HOME/etc/apps
, remove them.SA-VMNetAppUtils
SA-VMWIndex
TA-VMW-FieldExtractions
- Start splunkd:
$SPLUNK_HOME/bin/splunk start
Search historical data after you upgrade
You can continue searching historical data that the Splunk Add-on for VMware version 3.4.5 indexed. To do so, you have to add the vmware-perf
stanza to $SPLUNK_HOME/etc/apps/SA-VMWIndex/local
.
Follow these steps to search historical data after you upgrade to the Splunk VMware Add-on for ITSI.
- Stop splunkd:
$SPLUNK_HOME/bin/splunk stop
- In
$SPLUNK_HOME/etc/apps/SA-VMWIndex/local
, createindexes.conf
if it doesn't already exist. - Open
indexes.conf
and add this stanza:[vmware-perf] homePath = $SPLUNK_DB/vmware-perf/db coldPath = $SPLUNK_DB/vmware-perf/colddb thawedPath = $SPLUNK_DB/vmware-perf/thaweddb repFactor = auto
- Start splunkd:
$SPLUNK_HOME/bin/splunk start
Deploy a Data Collection Scheduler | Update a Data Collection Node after migrating Python versions |
This documentation applies to the following versions of Splunk® App for Infrastructure (Legacy): 2.0.1, 2.0.2, 2.0.3, 2.0.4, 2.1.0, 2.1.1 Cloud only
Feedback submitted, thanks!