You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@nifi.apache.org by Kevin Verhoeven <Ke...@ds-iq.com> on 2016/12/21 23:18:14 UTC
Remote Processor Group error: Unable to communicate with remote NiFi
cluster
I am running a cluster of twelve nodes on four VMs, each node runs on a different port. This configuration worked great on version 0.7. However, after an update from 0.7 to 1.0 we started seeing errors and flowfiles stopped moving into the Remote Processor Group. We are using the Remote Processor Group to send flowfiles, generated on the Primary Node, back into the same cluster but distributed among the other nodes.
The error we see on the Remote Processor Group:
Unable to refresh Remote Group’s peers due to Unable to communicate with remote NiFi cluster in order to determine which nodes exist in the remote cluster.
When I attempt to change the Remote Processor I see the following error:
Unable to find remote process group with id '87f96bb0-0158-1000-ffff-ffffc65b55ce'.
Do you have any suggestions?
Thanks,
Kevin
Re: Remote Processor Group error: Unable to communicate with remote
NiFi cluster
Posted by Koji Kawamura <ij...@gmail.com>.
Hi Kevin,
I hope the latest version would solve your problems.
Just a thought, I wonder if there's any round robin stuff is going on
behind the DNS.. It could produce intermittently failure.
Please let us know if the issue persists after update.
Thanks,
Koji
On Fri, Dec 23, 2016 at 3:14 AM, Kevin Verhoeven
<Ke...@ds-iq.com> wrote:
> Thanks Koji,
>
> The VM ip addresses have not changed and the target URL is accessible. However, I am using a DNS alias which allows me to change the name if the ip changes. One thing that is odd is that flowfiles have moved through the RemoteProcessGroup successfully, but intermittently it will stop.
>
> I will try the 1.1.0 update to test, or wait for the 1.1.1 update. This might just solve my problem.
>
> Thanks for your advice,
>
> Kevin
>
> -----Original Message-----
> From: Koji Kawamura [mailto:ijokarumawak@gmail.com]
> Sent: Wednesday, December 21, 2016 10:53 PM
> To: users@nifi.apache.org
> Subject: Re: Remote Processor Group error: Unable to communicate with remote NiFi cluster
>
> Hello Kevin,
>
> Sorry to hear that you got bitten by the issue, I think the 'Unable to find remote process group with id ...' is the same one I've encountered with NiFi 1.0.0.
> The issue has been solved since 1.1.0.
> https://issues.apache.org/jira/browse/NIFI-2687
>
> So, if you're able to upgrade the cluster to 1.1.0, you'll be able to update RemoteProcessGroup.
> Please also note that bug fix release 1.1.1 is under voting process right now. It seems to be released in next few days.
> I'd recommend to wait 1.1.1 if you're going to update your cluster again.
>
> However, I remember that although I couldn't update RemoteProcessGroup due to NIFI-2687, NiFi was able to transfer data using RemoteProcessGroup. So, it seems there's another issue.
>
> Have the VMs ip addresses (or hostnames) been changed? Is the target URL used by the RemoteProcessGroup still accessible?
>
> Thanks,
> Koji
>
> On Thu, Dec 22, 2016 at 8:18 AM, Kevin Verhoeven <Ke...@ds-iq.com> wrote:
>> I am running a cluster of twelve nodes on four VMs, each node runs on
>> a different port. This configuration worked great on version 0.7.
>> However, after an update from 0.7 to 1.0 we started seeing errors and
>> flowfiles stopped moving into the Remote Processor Group. We are using
>> the Remote Processor Group to send flowfiles, generated on the Primary
>> Node, back into the same cluster but distributed among the other nodes.
>>
>>
>>
>> The error we see on the Remote Processor Group:
>>
>>
>>
>> Unable to refresh Remote Group’s peers due to Unable to communicate
>> with remote NiFi cluster in order to determine which nodes exist in
>> the remote cluster.
>>
>>
>>
>> When I attempt to change the Remote Processor I see the following error:
>>
>>
>>
>> Unable to find remote process group with id
>> '87f96bb0-0158-1000-ffff-ffffc65b55ce'.
>>
>>
>>
>> Do you have any suggestions?
>>
>>
>>
>> Thanks,
>>
>>
>>
>> Kevin
RE: Remote Processor Group error: Unable to communicate with remote
NiFi cluster
Posted by Kevin Verhoeven <Ke...@ds-iq.com>.
Thanks Koji,
The VM ip addresses have not changed and the target URL is accessible. However, I am using a DNS alias which allows me to change the name if the ip changes. One thing that is odd is that flowfiles have moved through the RemoteProcessGroup successfully, but intermittently it will stop.
I will try the 1.1.0 update to test, or wait for the 1.1.1 update. This might just solve my problem.
Thanks for your advice,
Kevin
-----Original Message-----
From: Koji Kawamura [mailto:ijokarumawak@gmail.com]
Sent: Wednesday, December 21, 2016 10:53 PM
To: users@nifi.apache.org
Subject: Re: Remote Processor Group error: Unable to communicate with remote NiFi cluster
Hello Kevin,
Sorry to hear that you got bitten by the issue, I think the 'Unable to find remote process group with id ...' is the same one I've encountered with NiFi 1.0.0.
The issue has been solved since 1.1.0.
https://issues.apache.org/jira/browse/NIFI-2687
So, if you're able to upgrade the cluster to 1.1.0, you'll be able to update RemoteProcessGroup.
Please also note that bug fix release 1.1.1 is under voting process right now. It seems to be released in next few days.
I'd recommend to wait 1.1.1 if you're going to update your cluster again.
However, I remember that although I couldn't update RemoteProcessGroup due to NIFI-2687, NiFi was able to transfer data using RemoteProcessGroup. So, it seems there's another issue.
Have the VMs ip addresses (or hostnames) been changed? Is the target URL used by the RemoteProcessGroup still accessible?
Thanks,
Koji
On Thu, Dec 22, 2016 at 8:18 AM, Kevin Verhoeven <Ke...@ds-iq.com> wrote:
> I am running a cluster of twelve nodes on four VMs, each node runs on
> a different port. This configuration worked great on version 0.7.
> However, after an update from 0.7 to 1.0 we started seeing errors and
> flowfiles stopped moving into the Remote Processor Group. We are using
> the Remote Processor Group to send flowfiles, generated on the Primary
> Node, back into the same cluster but distributed among the other nodes.
>
>
>
> The error we see on the Remote Processor Group:
>
>
>
> Unable to refresh Remote Group’s peers due to Unable to communicate
> with remote NiFi cluster in order to determine which nodes exist in
> the remote cluster.
>
>
>
> When I attempt to change the Remote Processor I see the following error:
>
>
>
> Unable to find remote process group with id
> '87f96bb0-0158-1000-ffff-ffffc65b55ce'.
>
>
>
> Do you have any suggestions?
>
>
>
> Thanks,
>
>
>
> Kevin
Re: Remote Processor Group error: Unable to communicate with remote
NiFi cluster
Posted by Koji Kawamura <ij...@gmail.com>.
Hello Kevin,
Sorry to hear that you got bitten by the issue, I think the 'Unable to
find remote process group with id ...' is the same one I've
encountered with NiFi 1.0.0.
The issue has been solved since 1.1.0.
https://issues.apache.org/jira/browse/NIFI-2687
So, if you're able to upgrade the cluster to 1.1.0, you'll be able to
update RemoteProcessGroup.
Please also note that bug fix release 1.1.1 is under voting process
right now. It seems to be released in next few days.
I'd recommend to wait 1.1.1 if you're going to update your cluster again.
However, I remember that although I couldn't update RemoteProcessGroup
due to NIFI-2687, NiFi was able to transfer data using
RemoteProcessGroup. So, it seems there's another issue.
Have the VMs ip addresses (or hostnames) been changed? Is the target
URL used by the RemoteProcessGroup still accessible?
Thanks,
Koji
On Thu, Dec 22, 2016 at 8:18 AM, Kevin Verhoeven
<Ke...@ds-iq.com> wrote:
> I am running a cluster of twelve nodes on four VMs, each node runs on a
> different port. This configuration worked great on version 0.7. However,
> after an update from 0.7 to 1.0 we started seeing errors and flowfiles
> stopped moving into the Remote Processor Group. We are using the Remote
> Processor Group to send flowfiles, generated on the Primary Node, back into
> the same cluster but distributed among the other nodes.
>
>
>
> The error we see on the Remote Processor Group:
>
>
>
> Unable to refresh Remote Group’s peers due to Unable to communicate with
> remote NiFi cluster in order to determine which nodes exist in the remote
> cluster.
>
>
>
> When I attempt to change the Remote Processor I see the following error:
>
>
>
> Unable to find remote process group with id
> '87f96bb0-0158-1000-ffff-ffffc65b55ce'.
>
>
>
> Do you have any suggestions?
>
>
>
> Thanks,
>
>
>
> Kevin