[
https://issues.apache.org/jira/browse/CASSANDRA-15883?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yazal Ulloa updated CASSANDRA-15883:
------------------------------------
Description:
I have a table like this: key1, key2, key3, column1, column2.
When I delete a row with keys 1, 2, 3. and then insert a new row almost
immediately with the same keys, cassandra does not store the new row even
though the query executes successfully.
How can I force Cassandra to store the new row, or do I have to change my data
model?
EDIT:
I'm running Cassandra 3.11.6 in Docker, a single instance for development
purposes.
This has happen already with 3 tables.
{code:java}
// CREATE TABLE my_keyspace.my_chat ( country text, user_id timeuuid, chat_id
bigint, email text, first_name text, last_name text, username text, PRIMARY KEY
(country, user_id, chat_id));
{code}
{code:java}
// CREATE TABLE my_keyspace.my_profile (CREATE TABLE my_keyspace.my_profile (
country text, user_id timeuuid, profile_code text, id_doc text, id_doc_type
text, user_email text, user_name text, PRIMARY KEY (country, user_id)) WITH
CLUSTERING ORDER BY ( user_id DESC );
{code}
{code:java}
// CREATE TABLE my_keyspace.role ( realm text, business_id timeuuid, user_id
timeuuid, owner_id timeuuid, name text, status text, icon text, url text,
description text, owner_email text, owner_name text, scope_id timeuuid,
scope_name text, scopes SET<text>, authorization boolean, user_email text,
user_name text, views SET<FROZEN<my_dt>>, PRIMARY KEY (realm, business_id,
user_id, owner_id, name, status)) WITH CLUSTERING ORDER BY ( business_id DESC,
user_id DESC, owner_id DESC, name DESC, status DESC);
{code}
The value of the keys for this tables come from other tables or outside
immutable data.
I'm also using the Datastax OSS Java Driver v4.7.0 for comunications with
Cassandra, but the problem have also presented using cqlsh.
The queries I use are standard read by partition key, insert the whole row and
delete by full partiion key.
was:
I have a table like this: key1, key2, key3, column1, column2.
When I delete a row with keys 1, 2, 3. and then insert a new row almost
immediately with the same keys, cassandra does not store the new row even
though the query executes successfully.
How can I force Cassandra to store the new row, or do I have to change my data
model?
EDIT:
I'm running Cassandra 3.11.6 in Docker, a single instance for development
purposes.
This has happen already with 3 tables.
{code:java}
// CREATE TABLE my_keyspace.my_chat ( country text, user_id timeuuid, chat_id
bigint, email text, first_name text, last_name text, username text, PRIMARY KEY
(country, user_id, chat_id));
{code}
{code:java}
// CREATE TABLE my_keyspace.my_profile (CREATE TABLE my_keyspace.my_profile (
country text, user_id timeuuid, profile_code text, id_doc text, id_doc_type
text, user_email text, user_name text, PRIMARY KEY (country, user_id)) WITH
CLUSTERING ORDER BY ( user_id DESC );
{code}
{code:java}
// CREATE TABLE my_keyspace.role ( realm text, business_id timeuuid, user_id
timeuuid, owner_id timeuuid, name text, status text, icon text, url text,
description text, owner_email text, owner_name text, scope_id timeuuid,
scope_name text, scopes SET<text>, authorization boolean, user_email text,
user_name text, views SET<FROZEN<my_dt>>, PRIMARY KEY (realm, business_id,
user_id, owner_id, name, status)) WITH CLUSTERING ORDER BY ( business_id DESC,
user_id DESC, owner_id DESC, name DESC, status DESC);
{code}
The value of the keys for this tables come from other tables or outside
immutable data.
The queries I use are standard read by partition key, insert the whole row and
delete by full partiion key.
> Delete and save with same keys
> ------------------------------
>
> Key: CASSANDRA-15883
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15883
> Project: Cassandra
> Issue Type: Bug
> Reporter: Yazal Ulloa
> Priority: Normal
>
> I have a table like this: key1, key2, key3, column1, column2.
>
> When I delete a row with keys 1, 2, 3. and then insert a new row almost
> immediately with the same keys, cassandra does not store the new row even
> though the query executes successfully.
> How can I force Cassandra to store the new row, or do I have to change my
> data model?
> EDIT:
> I'm running Cassandra 3.11.6 in Docker, a single instance for development
> purposes.
> This has happen already with 3 tables.
> {code:java}
> // CREATE TABLE my_keyspace.my_chat ( country text, user_id timeuuid, chat_id
> bigint, email text, first_name text, last_name text, username text, PRIMARY
> KEY (country, user_id, chat_id));
> {code}
> {code:java}
> // CREATE TABLE my_keyspace.my_profile (CREATE TABLE my_keyspace.my_profile (
> country text, user_id timeuuid, profile_code text, id_doc text, id_doc_type
> text, user_email text, user_name text, PRIMARY KEY (country, user_id)) WITH
> CLUSTERING ORDER BY ( user_id DESC );
> {code}
> {code:java}
> // CREATE TABLE my_keyspace.role ( realm text, business_id timeuuid, user_id
> timeuuid, owner_id timeuuid, name text, status text, icon text, url text,
> description text, owner_email text, owner_name text, scope_id timeuuid,
> scope_name text, scopes SET<text>, authorization boolean, user_email text,
> user_name text, views SET<FROZEN<my_dt>>, PRIMARY KEY (realm, business_id,
> user_id, owner_id, name, status)) WITH CLUSTERING ORDER BY ( business_id
> DESC, user_id DESC, owner_id DESC, name DESC, status DESC);
> {code}
> The value of the keys for this tables come from other tables or outside
> immutable data.
>
> I'm also using the Datastax OSS Java Driver v4.7.0 for comunications with
> Cassandra, but the problem have also presented using cqlsh.
>
> The queries I use are standard read by partition key, insert the whole row
> and delete by full partiion key.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
