Splunk Enterprise

datamodel acceleration doesn't remove old guid

sarit_s
Communicator

Hello,
We are running DM acceleration, we saw that every time the acceleration is running the disk got full.

After investigation, we saw that the data of the old guid does not removed from disk and that cause our disk full. 

We are running Splunk using Docker image and using ansible. looks like it is an issue with ansible but im not sure. 

Any idea anyone ?

Thanks

Labels (2)
Tags (2)
0 Karma

scelikok
Champion

Unfortunately no, only deleted by the search head if it disables/rebuilds the acceleration.

You should manually delete the old ones. 

If this reply helps you an upvote is appreciated.
0 Karma

scelikok
Champion

Hi @sarit_s,

DM acceleration summary files are kept in indexers by using guid of the search head that has DM config. If you are starting a fresh Splunk Docker image every time, this guid will change and re-create a new acceleration files.

You should run Splunk by keeping /opt/splunk/etc path persistent on disk. This will prevent changing guid and also new summary files.

If this reply helps you an upvote is appreciated.
0 Karma

sarit_s
Communicator

Thanks!

what about that the old once kept on disk? Shouldnt they be removed once i have new once?

0 Karma
.conf21 CFS Extended through 5/20!

Don't miss your chance
to share your Splunk
wisdom in-person or
virtually at .conf21!

Call for Speakers has
been extended through
Thursday, 5/20!