Advice if Flapjack is right for me.

[Andy Lax]

Hi

I’m looking for a central notification solutions as I have a number of monitoring tools (Nagios is one of them) and it would makes life a lot easier to be able to define notification rules in one place, some of the tools may raise different alerts for the same server so it would save repetition, and also some of the tools are not very flexible with the notification rules you can define so it would add flexibility.

I think Flapjack is a good fit, my user case would be to have multiple tools sending alerts with the solution then routing the alerts and either sending emails, SMS messages, update twitter feeds, or raise tickets with our service desk.

I have some questions but if I’m barking up the wrong tree feel free to let me know.

i)                    Is Flapjack the right tool for my requirement?

ii)                   I can’t see documentation on how I send alerts to Flapjack from tools other than Nagios, Sensu, Incinga, is it possible?

iii)                 How do I build in other notification gateways (for updating twitter updates and raising tickets)?

iv)                 When will v2.0 be available?

v)                  Looking at a previous post (https://groups.google.com/forum/#!topic/flapjack-project/dUO8aZpDhKM ) I’m just looking to route alerts so what version will the back port https://github.com/flapjack/flapjack/pull/748 be included?

vi)                 I have situations where a contact will need to receive any and all alerts for a server (which I know Flapjack can do), but there are also contacts that will need to receive notifications only about specific checks that will apply to multiple hosts, is this possible?

vii)               More a comment than a question, in the post above (https://groups.google.com/forum/#!topic/flapjack-project/dUO8aZpDhKM) there was a comment “We’re planning on incorporating additional information into notifications, whether that information is already looked up and included in the event flapjack receives, or whether flapjack engages an external lookup service. Watch this space.” this is something I would be really interested in as we would also like to enrich our alerts with data from our CMDB.

 Apologies for the amount of questions, and many thanks in advance for any replies or advice.

 Andy

[Jesse Reynolds]

> On 15 Apr 2015, at 7:24 am, Andy Lax <andy...@gmail.com> wrote:
>
> Hi
>
> I’m looking for a central notification solutions as I have a number of monitoring tools (Nagios is one of them) and it would makes life a lot easier to be able to define notification rules in one place, some of the tools may raise different alerts for the same server so it would save repetition, and also some of the tools are not very flexible with the notification rules you can define so it would add flexibility.
>
> I think Flapjack is a good fit, my user case would be to have multiple tools sending alerts with the solution then routing the alerts and either sending emails, SMS messages, update twitter feeds, or raise tickets with our service desk.
>
> I have some questions but if I’m barking up the wrong tree feel free to let me know.
>
> i) Is Flapjack the right tool for my requirement?

Probably :-)

>
> ii) I can’t see documentation on how I send alerts to Flapjack from tools other than Nagios, Sensu, Incinga, is it possible?

Yes it’s possible. Currently the integration requires a connection to flapjack’s redis db and you then add events as json blobs to the events list. This is how flapjackfeeder (nagios / iginga event broker) work. See the Events format definition at the top of http://flapjack.io/docs/1.0/development/DATA_STRUCTURES/

There is also the possibility of POSTing events to the http event broker, which has some additional features also, eg it can be used for non-heartbeat style event streams, where you send along a TTL and the http event broker will then cache the last state and repeatedly emit events for this check until the TTL expires, whereupon it’ll start sending out UNKNOWN events for that check.

I’ve just added some basic documentation for the http broker: http://flapjack.io/docs/1.0/usage/Http-Broker/

>
> iii) How do I build in other notification gateways (for updating twitter updates and raising tickets)?

At the moment it involves adding it to the flapjack source code in the same way the existing notification gateways are implemented. This sucks. We’re slating fixing this with a cleanly pluggable system in a future major release, probably Flapjack 3. For now, the situation is less than ideal. We would accept code contributions for a twitter dm and/or public tweet gateway. As for raising a ticket, that depends I guess.

>
> iv) When will v2.0 be available?

In some months. We’re getting pretty close.

>
> v) Looking at a previous post (https://groups.google.com/forum/#!topic/flapjack-project/dUO8aZpDhKM ) I’m just looking to route alerts so what version will the back port https://github.com/flapjack/flapjack/pull/748 be included?

It was included in the 1.x series since 1.3 see https://github.com/flapjack/flapjack/releases/tag/v1.3.0 … it’s properly exposed in the api in 1.5.

>
> vi) I have situations where a contact will need to receive any and all alerts for a server (which I know Flapjack can do), but there are also contacts that will need to receive notifications only about specific checks that will apply to multiple hosts, is this possible?

Yes, you can assign those contacts to the ALL entity and then create a notification rule for each specific tag-set they want to be notified for, and disable the notifications from their default general (catch-all) rule.

>
> vii) More a comment than a question, in the post above (https://groups.google.com/forum/#!topic/flapjack-project/dUO8aZpDhKM) there was a comment “We’re planning on incorporating additional information into notifications, whether that information is already looked up and included in the event flapjack receives, or whether flapjack engages an external lookup service. Watch this space.” this is something I would be really interested in as we would also like to enrich our alerts with data from our CMDB.

Good to know. As far as I’m aware nobody is actively working on this but it’s certainly something that’s planned.

>
> Apologies for the amount of questions, and many thanks in advance for any replies or advice.

You’re welcome!