[heketi-devel] Why Boltdb ?

Raghavendra Talur rtalur at redhat.com
Wed Apr 17 20:28:09 UTC 2019


On Wed, Apr 17, 2019 at 3:20 PM Sudheer Singh
<sudheer.singh at stonybrook.edu> wrote:
>
> Hi,
> I have been exploring glusterfs for a while now. I wanted to understand what lead to choosing BoltDb? Can we disable BoltDb? can you please provide some doc where I can get more information?.

Gluster itself does not use boltdb and does not require heketi.

If you choose to use heketi to administer your Gluster Cluster(s),
then you also end up using BoltDB.

Why did the original author choose BoltDB?
I don't know all the reasons, but the slide[1] should tell the
features that lead to the choice.

Can we disable BoltDB?
No, as of today, heketi relies on BoltDB to store the information
about Gluster/Devices/Nodes.
It is possible to refactor the code to store the information in any
other key value store but we have not found any reason to.

[1] https://talks.godoc.org/github.com/lpabon/go-slides/2016/fosdem-writing-go-rest.slide#36


>
> Thanks,
> Sudheer
>
> _______________________________________________
> heketi-devel mailing list
> heketi-devel at gluster.org
> https://lists.gluster.org/mailman/listinfo/heketi-devel


More information about the heketi-devel mailing list