FAQ
Hi Riccardo,

I assume that you're running on CDH4.6 or newer. As Matt has mentioned
earlier that "you should be aware that Hue does not close queries until you
close the page", please make sure that you close the Hue page!

Thanks,
Alan



On Tue, Jul 22, 2014 at 8:47 AM, Riccardo Carè wrote:

I checked from Cloudera Manager and I found a lot of Impala queries that
are still in "execution" state. I also found that those queries are
launched by HUE. The temporary solution is to cancel the queries from
Cloudera Manager, but what could I do to solve the problem? Maybe I am
missing some configuration parameter

Thank you in advance

Il giorno lunedì 21 luglio 2014 23:44:07 UTC+2, Matthew Jacobs ha scritto:
Hi Riccardo,

This occurs when too many queries are running at the same time,
resulting in queries being queued by Impala's admission controller
[1]. If the queries are waiting in the queue for too long (by default
60sec) they will time out and return an error, which is what you're
seeing. The documentation [1] describes this in much more detail, but
you can do one of several things:
1) Increase the default queue timeout (--queue_wait_timeout_ms impalad
flag)
2) Increase the number of queries that can run concurrently (see the
documentation about configuration)

Are you submitting all of your queries through Hue? If so, you should
be aware that Hue does not close queries until you close the page, so
they count as running queries which may cause you to reach the limit.

Best,
Matt

1: http://www.cloudera.com/content/cloudera-content/
cloudera-docs/CDH5/latest/Impala/Installing-and-Using-
Impala/ciiu_admission.html

On Mon, Jul 21, 2014 at 2:34 PM, Riccardo Carè <riccard...@gmail.com>
wrote:
Hello,

I have a cluster with no evident problems (cloudera Manager does not show
any evident issue), but sometimes it randomly stops working.
More in detail, sometimes when a query is submitted through HUE, I get this
message:

Admission for query id=1d4caa50053cfd3e:d2b29623771bd87 exceeded timeout
60000ms

Usually when this happens, all the following queries are dropped in the same
way and I am not even able to execute the queries through Hive.
Sometimes instead of the previous message I get a message stating that the
database is locked.

Usually, after some minutes everything starts working again.

What can I do to detect what is causing these issues?

Thank you very much in advance

To unsubscribe from this group and stop receiving emails from it, send an
email to impala-user...@cloudera.org.
To unsubscribe from this group and stop receiving emails from it, send an
email to impala-user+unsubscribe@cloudera.org.
To unsubscribe from this group and stop receiving emails from it, send an email to impala-user+unsubscribe@cloudera.org.

Search Discussions

Discussion Posts

Previous

Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 3 of 3 | next ›
Discussion Overview
groupimpala-user @
categorieshadoop
postedJul 21, '14 at 9:44p
activeJul 22, '14 at 4:43p
posts3
users3
websitecloudera.com
irc#hadoop

People

Translate

site design / logo © 2022 Grokbase