nifi-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Conrad Crampton <>
Subject Re: Nifi versions and upgrades
Date Fri, 06 May 2016 12:20:29 GMT
Ok, will do

On 06/05/2016, 13:18, "Joe Witt" <> wrote:

>Glad the upgrade is back on track.
>As for the kite/avro and locale case probably best to start another
>thread as it will have a better chance of catching the eye of folks
>with expertise there.
>On Fri, May 6, 2016 at 8:05 AM, Conrad Crampton
><> wrote:
>> Hi Joe,
>> Thanks for your quick response (again).
>> You prompted me to look at every node in the cluster with ps aux | grep nifi and
found that two of my nodes were still running 0.5.1. I do remember now that I was experiencing
an odd thing with these in that everytime I did a service nifi stop, I would see nifi start
up again (using ps) which was a bit odd but I had thought I had sorted that.
>> However, having now sorted out these two nodes all appears well again. Correct version
reported and all processors available. ;-)
>> However…
>> I am seeing
>> java.lang.IllegalArgumentException: Invalid locale format: default
>>         at org.apache.commons.lang.LocaleUtils.toLocale( ~[na:na]
>>         at org.apache.nifi.processors.kite.ConvertAvroSchema.onTrigger(
>> I have a number of flows that use convertAvroSchema processor but I just have the
default locale of ‘default’ in there which would appear to be the problem. I’m running
on Java 8 if that make a difference.
>> Thanks
>> Conrad
>> On 06/05/2016, 12:22, "Joe Witt" <> wrote:
>>>Inside the conf/ file there is a property called
>>>  nifi.version=
>>>You will need to change that to the version you've upgrade to if you
>>>just carried forward the old config file.  That value is what ends up
>>>being displayed in the web-ui.
>>>Now, having said this if you are running and see old processors and
>>>such listed then it means the NiFi instance you're connecting to is
>>>likely still an old one.  Are you sure you stopped the old one before
>>>starting the new one?  Could you show a listing of the nifi lib
>>>directory as found in the ps listing you reference?
>>>On Fri, May 6, 2016 at 5:36 AM, Conrad Crampton
>>><> wrote:
>>>> Hi,
>>>> I have followed the advice on upgrade planning [1] and [2] and had success
>>>> since upgrading from 0.4.1 to 0.6.1 (via .5x) or so I thought!
>>>> I am using ansible to perform the upgrades to 6 node cluster but then go
>>>> into each server to do a ./ install to install new version as
>>>> service. Running service nifi start shows correct bootstrap messages in that
>>>> 0.6.1 is running with correct home (/opt/nifi-0.6.1) and all seems fine.
>>>> There is a /work directory expanded with all correct versions of nars etc.
>>>> and ps aux | grep nifi shows that nifi is in fact running from
>>>> /opt/nifi-0.6.1.
>>>> So what’s the problem?
>>>> In the UI, I was trying to use the new ListenTCP processor and it isn’t
>>>> listed when I drag the processors onto the workspace (I get 125 processors
>>>> listed), and when I do help…about, I get 0.5.1 as version.
>>>> Can anyone suggest how to force UI to update (if indeed this is the
>>>> problem)??
>>>> Thanks
>>>> Conrad
>>>> [1]
>>>> [2]
>>>> SecureData, combating cyber threats
>>>> ________________________________
>>>> The information contained in this message or any of its attachments may be
>>>> privileged and confidential and intended for the exclusive use of the
>>>> intended recipient. If you are not the intended recipient any disclosure,
>>>> reproduction, distribution or other dissemination or use of this
>>>> communications is strictly prohibited. The views expressed in this email
>>>> those of the individual and not necessarily of SecureData Europe Ltd. Any
>>>> prices quoted are only valid if followed up by a formal written quote.
>>>> SecureData Europe Limited. Registered in England & Wales 04365896.
>>>> Registered Address: SecureData House, Hermitage Court, Hermitage Lane,
>>>> Maidstone, Kent, ME16 9NT
>>> ***This email originated outside SecureData***
 to report this email as spam.
View raw message