Monitor RIAK process with Supervisord
mkessler at basho.com
Mon Oct 31 04:59:54 EDT 2016
On 29 October 2016 at 19:59, vmalhotra <varun.malhotra at minjar.com> wrote:
> We run 8 nodes RIAK cluster in our Prod environment. Lot of time, RIAK
> process stops and we also noticed out of memory issues. Typically, we run
> restart the affected node to recover from the issue. I thought of using
> Supervisor to control the RIAK processes so the idea is if any of the
> process crash SupervisorD daemon will automatically restart that process on
> a crash.
> Wanted to know what you guys think? Can it cause any other issue or it
> should work fine?
I would recommend against blindly restarting Riak nodes, in particular if
these were shut down uncleanly, as may happen in out of memory situations.
There is a risk that an unclean shutdown leaves behind corrupted files and
that a subsequent restart is unsuccessful.
You should instead investigate why Riak stops being responsive.
Please have a look at the documentation, in particular memory
requirements, and OS tuning.
> Thanks in advance.
> View this message in context: http://riak-users.197444.n3.
> Sent from the Riak Users mailing list archive at Nabble.com.
> riak-users mailing list
> riak-users at lists.basho.com
Client Services Engineer
Basho Technologies Limited
Registered Office - 8 Lincoln’s Inn Fields London WC2A 3BP Reg 07970431
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the riak-users