Hi Avi,
These are internal files to RockDb. Depending on your load in the system I
suppose they could contain quite a bit of data. How large was the load in the
system these past two weeks so we can calibrate? Otherwise I'm not sure if
1-2GB is a lot or not (sounds like not that big to make the disk full, was
there something else as well that ate up space?)
Thanks
Eno
> On 29 Jun 2016, at 16:33, Avi Flax <avi.f...@parkassist.com> wrote:
>
> Hello all,
>
> I’ve been working on a Streams app and so far it’s going quite well. I’ve had
> the app up and running in a staging environment for a few weeks, processing
> real data.
>
> Today I logged into the server to check in on some things, and I found the
> disk was full.
>
> I managed (with ncdu) to find that most of the unexpected usage was in:
>
> /tmp/kafka-streams/my_app/0_0/rocksdb/my_intermediate_aggregate_topic
>
> and specifically in files named LOG.old.{timestamp}. The files were 2.6 GB
> and 118 MB. Also, LOG was 1.1 GB.
>
> I checked the docs and searched the list (and Google) for info on Kafka
> Streams and/or RocksDB disk usage, cleanup, etc, but didn’t find much.
>
> So:
>
> * Is this normal, expected behavior?
>
> * Any recommendations or suggestions for dealing with this?
>
> Thanks!
> Avi
