Having isues with NiFi cluster

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Having isues with NiFi cluster

Kalisz, John T.
I set up a NiFi cluster based on the following Wiki:

https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/

It is a three node cluster with one node (server) set up as the cluster manager and two other servers set up as nodes.
When I start the cluster everything appears to work. There are no discernable errors in the log files and the two nodes are sending heartbeats to the master. When I try bring up the NiFi index page, I simply see a "loading" icon that never goes away.
I also noticed that the heartbeats from the nodes are being generated at a significantly higher rate than the default setting of 5 seconds.

Any guidance would be appreciated.

John T. Kalisz
General Dynamics Mission Systems

Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]<mailto:[hidden email]>

This message and/or attachments may include information subject to GD Corporate Policies 07-103 and 07-105 and is intended to be accessed only by authorized recipients.  Use, storage and transmission are governed by General Dynamics and its policies. Contractual restrictions apply to third parties.  Recipients should refer to the policies or contract to determine proper handling.  Unauthorized review, use, disclosure or distribution is prohibited.  If you are not an intended recipient, please contact the sender and destroy all copies of the original message.

Reply | Threaded
Open this post in threaded view
|

Re: Having isues with NiFi cluster

Matt Gilman
John,

Can you check both of the log files (nifi-app and nifi-user) on the NCM and
the Nodes. Specifically, the user log that shows the web requests into the
NiFi instance. Without knowing more, I would guess that the culprit is
going to be on one of the nodes. If there is an error in any of the user
logs, you can enable debug logging to get the entire stacktrace by editing
conf/logback.xml. Specifically set the level to DEBUG on the following
entry:

    <logger name="org.apache.nifi.web.api.config" level="INFO"
additivity="false">
        <appender-ref ref="USER_FILE"/>
    </logger>

There's likely something misconfigured and it failing and the UI is not
handling it well. Once we know what that is, we can address that and update
the UI appropriately.

Thanks.

Matt Gilman

On Thu, Mar 19, 2015 at 7:05 AM, Kalisz, John T. <[hidden email]>
wrote:

> I set up a NiFi cluster based on the following Wiki:
>
>
> https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/
>
> It is a three node cluster with one node (server) set up as the cluster
> manager and two other servers set up as nodes.
> When I start the cluster everything appears to work. There are no
> discernable errors in the log files and the two nodes are sending
> heartbeats to the master. When I try bring up the NiFi index page, I simply
> see a "loading" icon that never goes away.
> I also noticed that the heartbeats from the nodes are being generated at a
> significantly higher rate than the default setting of 5 seconds.
>
> Any guidance would be appreciated.
>
> John T. Kalisz
> General Dynamics Mission Systems
>
> Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]
> <mailto:[hidden email]>
>
> This message and/or attachments may include information subject to GD
> Corporate Policies 07-103 and 07-105 and is intended to be accessed only by
> authorized recipients.  Use, storage and transmission are governed by
> General Dynamics and its policies. Contractual restrictions apply to third
> parties.  Recipients should refer to the policies or contract to determine
> proper handling.  Unauthorized review, use, disclosure or distribution is
> prohibited.  If you are not an intended recipient, please contact the
> sender and destroy all copies of the original message.
>
>
Reply | Threaded
Open this post in threaded view
|

Re: Having isues with NiFi cluster

Corey Flowers
In reply to this post by Kalisz, John T.
Have you verified that all the management ports are open? It maybe waiting on a response from one of the nodes. If you would like, you could send me the app and wrapper log and I will take a look.

Sent from my iPhone

> On Mar 19, 2015, at 7:05 AM, Kalisz, John T. <[hidden email]> wrote:
>
> I set up a NiFi cluster based on the following Wiki:
>
> https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/
>
> It is a three node cluster with one node (server) set up as the cluster manager and two other servers set up as nodes.
> When I start the cluster everything appears to work. There are no discernable errors in the log files and the two nodes are sending heartbeats to the master. When I try bring up the NiFi index page, I simply see a "loading" icon that never goes away.
> I also noticed that the heartbeats from the nodes are being generated at a significantly higher rate than the default setting of 5 seconds.
>
> Any guidance would be appreciated.
>
> John T. Kalisz
> General Dynamics Mission Systems
>
> Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]<mailto:[hidden email]>
>
> This message and/or attachments may include information subject to GD Corporate Policies 07-103 and 07-105 and is intended to be accessed only by authorized recipients.  Use, storage and transmission are governed by General Dynamics and its policies. Contractual restrictions apply to third parties.  Recipients should refer to the policies or contract to determine proper handling.  Unauthorized review, use, disclosure or distribution is prohibited.  If you are not an intended recipient, please contact the sender and destroy all copies of the original message.
>
Reply | Threaded
Open this post in threaded view
|

Re: Having isues with NiFi cluster

Oscar de la Pena
Hi,

I remember having the same problem

In my case, I had to assign correct hostnames for each node in the cluster

In conf/nifi.properties,

NCM node:
nifi.web.http.host=vbox1
nifi.web.http.port=7070

Slave node:
nifi.web.http.host=vbox2 (leaving this blank makes NiFi bind to localhost)
nifi.web.http.port=7272

Let me know how your setup is and I'll try to check also.

Owie
----- Original Message -----
From: "Corey Flowers" <[hidden email]>
To: [hidden email]
Sent: Friday, March 20, 2015 4:51:27 AM
Subject: Re: Having isues with NiFi cluster

Have you verified that all the management ports are open? It maybe waiting on a response from one of the nodes. If you would like, you could send me the app and wrapper log and I will take a look.

Sent from my iPhone

> On Mar 19, 2015, at 7:05 AM, Kalisz, John T. <[hidden email]> wrote:
>
> I set up a NiFi cluster based on the following Wiki:
>
> https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/
>
> It is a three node cluster with one node (server) set up as the cluster manager and two other servers set up as nodes.
> When I start the cluster everything appears to work. There are no discernable errors in the log files and the two nodes are sending heartbeats to the master. When I try bring up the NiFi index page, I simply see a "loading" icon that never goes away.
> I also noticed that the heartbeats from the nodes are being generated at a significantly higher rate than the default setting of 5 seconds.
>
> Any guidance would be appreciated.
>
> John T. Kalisz
> General Dynamics Mission Systems
>
> Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]<mailto:[hidden email]>
>
> This message and/or attachments may include information subject to GD Corporate Policies 07-103 and 07-105 and is intended to be accessed only by authorized recipients.  Use, storage and transmission are governed by General Dynamics and its policies. Contractual restrictions apply to third parties.  Recipients should refer to the policies or contract to determine proper handling.  Unauthorized review, use, disclosure or distribution is prohibited.  If you are not an intended recipient, please contact the sender and destroy all copies of the original message.
>
Reply | Threaded
Open this post in threaded view
|

Re: Having isues with NiFi cluster

Jennifer Barnabee
Yes, and you need to set the node address as well. We plan to update the documentation with this. Please keep the feedback coming! Thanks.

Sent from my iPhone

> On Mar 20, 2015, at 4:00 AM, Oscar dela Pena <[hidden email]> wrote:
>
> Hi,
>
> I remember having the same problem
>
> In my case, I had to assign correct hostnames for each node in the cluster
>
> In conf/nifi.properties,
>
> NCM node:
> nifi.web.http.host=vbox1
> nifi.web.http.port=7070
>
> Slave node:
> nifi.web.http.host=vbox2 (leaving this blank makes NiFi bind to localhost)
> nifi.web.http.port=7272
>
> Let me know how your setup is and I'll try to check also.
>
> Owie
> ----- Original Message -----
> From: "Corey Flowers" <[hidden email]>
> To: [hidden email]
> Sent: Friday, March 20, 2015 4:51:27 AM
> Subject: Re: Having isues with NiFi cluster
>
> Have you verified that all the management ports are open? It maybe waiting on a response from one of the nodes. If you would like, you could send me the app and wrapper log and I will take a look.
>
> Sent from my iPhone
>
>> On Mar 19, 2015, at 7:05 AM, Kalisz, John T. <[hidden email]> wrote:
>>
>> I set up a NiFi cluster based on the following Wiki:
>>
>> https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/
>>
>> It is a three node cluster with one node (server) set up as the cluster manager and two other servers set up as nodes.
>> When I start the cluster everything appears to work. There are no discernable errors in the log files and the two nodes are sending heartbeats to the master. When I try bring up the NiFi index page, I simply see a "loading" icon that never goes away.
>> I also noticed that the heartbeats from the nodes are being generated at a significantly higher rate than the default setting of 5 seconds.
>>
>> Any guidance would be appreciated.
>>
>> John T. Kalisz
>> General Dynamics Mission Systems
>>
>> Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]<mailto:[hidden email]>
>>
>> This message and/or attachments may include information subject to GD Corporate Policies 07-103 and 07-105 and is intended to be accessed only by authorized recipients.  Use, storage and transmission are governed by General Dynamics and its policies. Contractual restrictions apply to third parties.  Recipients should refer to the policies or contract to determine proper handling.  Unauthorized review, use, disclosure or distribution is prohibited.  If you are not an intended recipient, please contact the sender and destroy all copies of the original message.
>>
Reply | Threaded
Open this post in threaded view
|

Re: Having isues with NiFi cluster

Corey Flowers
Hey John, are still having this issue? If you would like we could
troubleshoot over chat, my email address is [hidden email]. I can
also send you the section out of my properties file for the cluster I have
on my system.

On Fri, Mar 20, 2015 at 4:18 AM, Jennifer Barnabee <
[hidden email]> wrote:

> Yes, and you need to set the node address as well. We plan to update the
> documentation with this. Please keep the feedback coming! Thanks.
>
> Sent from my iPhone
>
> > On Mar 20, 2015, at 4:00 AM, Oscar dela Pena <[hidden email]>
> wrote:
> >
> > Hi,
> >
> > I remember having the same problem
> >
> > In my case, I had to assign correct hostnames for each node in the
> cluster
> >
> > In conf/nifi.properties,
> >
> > NCM node:
> > nifi.web.http.host=vbox1
> > nifi.web.http.port=7070
> >
> > Slave node:
> > nifi.web.http.host=vbox2 (leaving this blank makes NiFi bind to
> localhost)
> > nifi.web.http.port=7272
> >
> > Let me know how your setup is and I'll try to check also.
> >
> > Owie
> > ----- Original Message -----
> > From: "Corey Flowers" <[hidden email]>
> > To: [hidden email]
> > Sent: Friday, March 20, 2015 4:51:27 AM
> > Subject: Re: Having isues with NiFi cluster
> >
> > Have you verified that all the management ports are open? It maybe
> waiting on a response from one of the nodes. If you would like, you could
> send me the app and wrapper log and I will take a look.
> >
> > Sent from my iPhone
> >
> >> On Mar 19, 2015, at 7:05 AM, Kalisz, John T. <[hidden email]>
> wrote:
> >>
> >> I set up a NiFi cluster based on the following Wiki:
> >>
> >>
> https://kisstechdocs.wordpress.com/2015/01/15/simple-tasks-in-nifi-creating-a-super-simple-cluster/
> >>
> >> It is a three node cluster with one node (server) set up as the cluster
> manager and two other servers set up as nodes.
> >> When I start the cluster everything appears to work. There are no
> discernable errors in the log files and the two nodes are sending
> heartbeats to the master. When I try bring up the NiFi index page, I simply
> see a "loading" icon that never goes away.
> >> I also noticed that the heartbeats from the nodes are being generated
> at a significantly higher rate than the default setting of 5 seconds.
> >>
> >> Any guidance would be appreciated.
> >>
> >> John T. Kalisz
> >> General Dynamics Mission Systems
> >>
> >> Office  413-494-3376  |  Cell  413-822-1883 |  [hidden email]
> <mailto:[hidden email]>
> >>
> >> This message and/or attachments may include information subject to GD
> Corporate Policies 07-103 and 07-105 and is intended to be accessed only by
> authorized recipients.  Use, storage and transmission are governed by
> General Dynamics and its policies. Contractual restrictions apply to third
> parties.  Recipients should refer to the policies or contract to determine
> proper handling.  Unauthorized review, use, disclosure or distribution is
> prohibited.  If you are not an intended recipient, please contact the
> sender and destroy all copies of the original message.
> >>
>



--
Corey Flowers
Vice President, Onyx Point, Inc
(410) 541-6699
[hidden email]

-- This account not approved for unencrypted proprietary information --