cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Petrov (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CASSANDRA-13152) UPDATE on counter columns with empty list as argument in IN disables cluster
Date Fri, 03 Feb 2017 14:58:51 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-13152?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Alex Petrov updated CASSANDRA-13152:
------------------------------------
    Description: 
On a 3 node cluster
with this table (replication factor of 2):

{code}
CREATE TABLE tracking.item_items_rec_history (
	reference_id bigint,
	country text,
	portal text,
	app_name text,
	recommended_id bigint,
	counter counter,
	PRIMARY KEY (reference_id, country, portal, app_name, recommended_id)
);
{code}
If I execute 

{code}
UPDATE user_items_rec_history 
SET counter = counter + 1 
WHERE reference_id = 1 AND country = '' AND portal = '' AND app_name = '' AND recommended_id
IN ();
{code}

Take note that the IN is empty

The cluster starts to malfunction and responds a lot of timeouts to any query.

After resetting some of the nodes, the cluster starts to function normally again.


  was:
On a 3 node cluster
with this table (replication factor of 2):
CREATE TABLE tracking.item_items_rec_history (
	reference_id bigint,
	country text,
	portal text,
	app_name text,
	recommended_id bigint,
	counter counter,
	PRIMARY KEY (reference_id, country, portal, app_name, recommended_id)
);

If I execute 
UPDATE user_items_rec_history 
SET counter = counter + 1 
WHERE reference_id = 1 AND country = '' AND portal = '' AND app_name = '' AND recommended_id
IN ();

Take note that the IN is empty

The cluster starts to malfunction and responds a lot of timeouts to any query.

After resetting some of the nodes, the cluster starts to function normally again.



> UPDATE on counter columns with empty list as argument in IN disables cluster
> ----------------------------------------------------------------------------
>
>                 Key: CASSANDRA-13152
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13152
>             Project: Cassandra
>          Issue Type: Bug
>         Environment: Linux Ubuntu 16
> 3 Virtual machines
>            Reporter: jorge collinet
>            Assignee: Benjamin Lerer
>             Fix For: 3.0.x, 3.x, 4.x
>
>
> On a 3 node cluster
> with this table (replication factor of 2):
> {code}
> CREATE TABLE tracking.item_items_rec_history (
> 	reference_id bigint,
> 	country text,
> 	portal text,
> 	app_name text,
> 	recommended_id bigint,
> 	counter counter,
> 	PRIMARY KEY (reference_id, country, portal, app_name, recommended_id)
> );
> {code}
> If I execute 
> {code}
> UPDATE user_items_rec_history 
> SET counter = counter + 1 
> WHERE reference_id = 1 AND country = '' AND portal = '' AND app_name = '' AND recommended_id
IN ();
> {code}
> Take note that the IN is empty
> The cluster starts to malfunction and responds a lot of timeouts to any query.
> After resetting some of the nodes, the cluster starts to function normally again.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message