[Gluster-devel] Logstash (was Re: GlusterFlow (GlusterFS Analytics) 0.02)

[Joe Julian]

On 02/05/2014 04:13 AM, Niels de Vos wrote:
> On Wed, Feb 05, 2014 at 04:15:09AM +0100, Justin Clift wrote:
>> Hi all,
>>
>> Feeling inspired, so did a marathon coding
>> session and created the next version of
>> GlusterFlow.  It's pretty much "Analytics
>> for GlusterFS".
>>
>>    https://github.com/justinclift/glusterflow
>>
>> No longer using Django & PostgreSQL, it's
>> now using LogStash, ElasticSearch, and
>> Kibana.
>>
>> If you like shiny things, this is a _very
>> shiny thing_. ;) (screenshots attached)
>>
>> Note - This is still initial prototype code,
>> working on a single node, but it shouldn't
>> be too long until it's properly multi-node
>> capable.
>>
>> Thoughts / suggestions? :)
> The screenshots look very promising! Have you thought of getting some
> things from the logs under /var/log/glusterfs? For me (troubleshooting
> issues in disconnected mode) that would be a more useful tool. Something
> that makes it possible to identify errors with specific bricks, volumes,
> clients etc...
>
That's already there with logstash. Just feed your logs into it. Voila! 
Now you have all your logs in one place with search tools, filtering, 
time-stamps, etc. If you're not already aggregating your logs, you don't 
even know how much work you're doing that you just don't have to.