Hi,
If possible, kindly share one output file to inspect, in the meanwhile you
could also give a try with "org.apache.hadoop.io.compress.GzipCodec"
Regards,
Ravi
On Tue, Oct 22, 2019 at 7:25 PM amran dean wrote:
>
> Hello,
> These changes result in the following error:
> $ lzop -d part-1-0
> l
Hi Samya,
Were you able to get this resolved? Seeing the same issue here after upgrading
to Flink 1.9 from 1.6.
Thanks,
Thad
From: Andrey Zagrebin
Sent: Wednesday, July 3, 2019 9:09 AM
To: Haibo Sun
Cc: Patro, Samya ; user@flink.apache.org; Bari, Swapnil
Subject: Re: Could not load the nat
Hello,
These changes result in the following error:
$ lzop -d part-1-0
lzop: part-1-0: not a lzop file
public class BulkRecordLZOSerializer implements BulkWriter {
private final CompressionOutputStream compressedStream;
public BulkRecordLZOSerializer(OutputStream stream) {
Compr
Thanks Vino, this is very helpful.
At the end of the day we are looking for something like:
A job that requires N nodes in Storm, only requires K nodes when
implemented in Flink, where hopefully K < N :)
Cheers,
Gyula
On Tue, Oct 22, 2019 at 2:31 PM vino yang wrote:
> Hi Gyula,
>
> Based on ou
Hi Gyula,
Based on our previous experience switching from Storm to Flink. For the
same business, resources of the same size are completely sufficient, and
the performance indicators are slightly better than Storm. As you said,
this may be related to using some of Flink's special features like stat
Thank you for your help Pritam. I got this working now. The missing pool
piece for me was that I had to set the "web.upload.dir" which my default is
some auto generated directory under tmp.
One problem I had was that the jar actually needs to be placed under
$web.upload.dir/flink-web-upload, i.
Just to give you more context,
We are using `com.google.cloud.bigtable` as well in our job dep. Could it
be due to shaded plugin issue with `bigtable-hbase-2.x` ?
-
Dhanesh Arole ( Sent from mobile device. Pardon me for typos )
On Tue, Oct 22, 2019 at 2:06 PM dhanesh arole
wrote:
> Hello al
Hi to all,
I was looking into the Flink example of the Flink training trying to
understand why in the ClickEventCount[1] one task manager was reading
twice the speed of the other.
I had to debug a lot of internal code of Flink to understand that it
depends on the adopted hash function (used by Fl
