You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@kudu.apache.org by Saeid Sattari <sa...@gmail.com> on 2018/05/17 09:36:55 UTC

Kudu query error

Hi all,

Lately, I have a critical problem with Kudu cluster (V1.6). For most of the
queries submitted by Impala (even very simple and light one)
regularly encountered with the following error. There are 120 datanode in
my cluster and changed following parameters on datanodes:

unlock_experimental_flags=true
rpc_service_queue_length=15000
rpc_acceptor_listen_backlog=1024

The error:

> WARNINGS: Unable to advance iterator: Timed out: Scan RPC to
> 172.18.77.131:7050 timed out after 0.000s (SENT): Not found: Scanner not
> found


Any hints and tips are appreciated. Thank you in advance.

Best,
Saeid

Re: Kudu query error

Posted by Saeid Sattari <sa...@gmail.com>.

Hi Yao,

Thanks. I will try it and let you know about the result.

Regards,
Saeid

On Thu, May 17, 2018 at 4:09 PM, 徐瑶 <oc...@gmail.com> wrote:

> Hi Saeid,
>
> I think it may be had a long time interval between ScanRequest.
> Setting the parameter --scanner_ttl_ms(default 60s) larger may solve the
> problem.
>
> Yao
>
> 2018-05-17 17:36 GMT+08:00 Saeid Sattari <sa...@gmail.com>:
>
>> Hi all,
>>
>> Lately, I have a critical problem with Kudu cluster (V1.6). For most of
>> the queries submitted by Impala (even very simple and light one)
>> regularly encountered with the following error. There are 120 datanode in
>> my cluster and changed following parameters on datanodes:
>>
>> unlock_experimental_flags=true
>> rpc_service_queue_length=15000
>> rpc_acceptor_listen_backlog=1024
>>
>> The error:
>>
>>> WARNINGS: Unable to advance iterator: Timed out: Scan RPC to
>>> 172.18.77.131:7050 timed out after 0.000s (SENT): Not found: Scanner
>>> not found
>>
>>
>> Any hints and tips are appreciated. Thank you in advance.
>>
>> Best,
>> Saeid
>>
>
>

Re: Kudu query error

Posted by 徐瑶 <oc...@gmail.com>.

Hi Saeid,

I think it may be had a long time interval between ScanRequest.
Setting the parameter --scanner_ttl_ms(default 60s) larger may solve the
problem.

Yao

2018-05-17 17:36 GMT+08:00 Saeid Sattari <sa...@gmail.com>:

> Hi all,
>
> Lately, I have a critical problem with Kudu cluster (V1.6). For most of
> the queries submitted by Impala (even very simple and light one)
> regularly encountered with the following error. There are 120 datanode in
> my cluster and changed following parameters on datanodes:
>
> unlock_experimental_flags=true
> rpc_service_queue_length=15000
> rpc_acceptor_listen_backlog=1024
>
> The error:
>
>> WARNINGS: Unable to advance iterator: Timed out: Scan RPC to
>> 172.18.77.131:7050 timed out after 0.000s (SENT): Not found: Scanner not
>> found
>
>
> Any hints and tips are appreciated. Thank you in advance.
>
> Best,
> Saeid
>