Joey,
I was able to reproduce the issue with a 0 byte file.
Reverting the commit you were referring to ([1]) fixes the problem.
Denes
[1]
https://github.com/apache/nifi/commit/0ed35345245da057400a7d875e47f57bc2d3794d
On Tue, Aug 17, 2021 at 8:12 AM Joey Frazee wrote:
> Vibhath, there was a
Vibhath, there was a change to allow reading by ranges that I’m wondering if is
causing this.
When you say small, exactly how small are the files?
-joey
> On Aug 16, 2021, at 10:53 PM, Vibhath Ileperuma
> wrote:
>
>
> Hi All,
>
> I'm using Nifi 1.14.0 to fetch a set of files from a S3
We've rolled back to 1.11.4 in a couple scenarios. We have also setup a
cron job to restart 1.13.2 nodes once a day.
Ryan
On Tue, Aug 17, 2021 at 1:25 AM Axel Schwarz wrote:
> Hey Ryan,
>
> that sounds awefully familiar. What we successfully battled so far is the
> load balancing problem.
>
If restarting is helping you then generally it should be easily found. Is
mem usage spiking? Are thread dumps revealing?
On Tue, Aug 17, 2021 at 6:12 AM Ryan Hendrickson <
ryan.andrew.hendrick...@gmail.com> wrote:
> We've rolled back to 1.11.4 in a couple scenarios. We have also setup a
>
Ryan,
Have you filed Jiras for these? If so, can you share the links?
Thanks
-Mark
On Aug 16, 2021, at 9:32 PM, Ryan Hendrickson
mailto:ryan.andrew.hendrick...@gmail.com>>
wrote:
Axel,
We've had significant issues with 1.13.2 in a Cluster as well. We're
working on a test config...
Hi Joey and Denes,
I'm also getting this error for 0 byte files.
Thank You.
On Tue, Aug 17, 2021 at 12:25 PM Denes Arvay wrote:
> Joey,
>
> I was able to reproduce the issue with a 0 byte file.
> Reverting the commit you were referring to ([1]) fixes the problem.
>
> Denes
>
> [1]
>
final Long rangeLength = (context.getProperty(RANGE_LENGTH).isSet() ?
context.getProperty(RANGE_LENGTH).evaluateAttributeExpressions(flowFile).asDataSize(DataUnit.B).longValue()
: null);
if (rangeLength != null) {
request.setRange(rangeStart, rangeStart + rangeLength - 1);
}
I think the
Arpad, the validator on range length should prevent that from happening if
there’s no EL. Not set or >= 1 should be the only allowed values.
Vibhath, does this happen with small, non-zero length files?
FYI, opened https://issues.apache.org/jira/browse/NIFI-9055
-joey
> On Aug 17, 2021, at
Hi Joey,
It happens only for 0B files.
Sorry for the confusion caused.
Vibhath.
Axel,
So a few things to consider here.
The NullPointer that you’re showing there in the logs was fixed in 1.14.0 [1].
Secondly, is a question of why the ZooKeeper client is disconnecting. Given
that you’re running an embedded zookeeper server, it could be for any number of
reasons - low java
Hi Mark,
We just submitted this one:
https://issues.apache.org/jira/browse/NIFI-9056
Another one will be on the way today/tomorrow for Relationship Load
Balancing issues.
Ryan
On Tue, Aug 17, 2021 at 10:17 AM Mark Payne wrote:
> Ryan,
>
> Have you filed Jiras for these? If so, can you
I was able to replicate this on Minio too and have a fix for it. The change is
trivial but it needs more testing.
-joey
> On Aug 17, 2021, at 8:22 AM, Vibhath Ileperuma
> wrote:
>
>
> Hi Joey,
>
> It happens only for 0B files.
> Sorry for the confusion caused.
>
> Vibhath.
Sorry, setting the property "Ignore response's content" to true, as described
in the commit, is test OK.
Works as expected now!
From: Hendrik Ruijter
Sent: den 16 augusti 2021 14:56
To: users@nifi.apache.org
Subject: Power BI Real Time REST API failure
Hello, I use docker apache/nifi 1.13.2
Hello ,
I am using DistributedMapCacheServer with setting of Persistent Directory. I
can see contents of cache in the directory, but the contents do not survive
nifi restart .
Older values from cache should be available after restart, right ?
I have attached template where I have configured
14 matches
Mail list logo
