You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Alexander Lapin (Jira)" <ji...@apache.org> on 2021/05/21 14:26:00 UTC
[jira] [Updated] (IGNITE-14764) Consolidation of managers and
metastorage server implementation.
[ https://issues.apache.org/jira/browse/IGNITE-14764?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Alexander Lapin updated IGNITE-14764:
-------------------------------------
Description:
After merging meta storage server logic few bugs of ignition start, table creation and atomic operations were revealed. Some of them will be specified in greater detail and fixed within corresponding tickets:
* https://issues.apache.org/jira/browse/IGNITE-14756 - NPE on reading tableConfiguration from ConfigurationManager within SchemaManager
* https://issues.apache.org/jira/browse/IGNITE-14762 - Leader refresh sometimes blocks indefinitely
* https://issues.apache.org/jira/browse/IGNITE-14290 Includes INT32, INT64, etc support
Other issues will be fixed within given ticket:
* Incoorect usage of appliedRevision as watch revision. appliedRevision + 1 should be used instead. That produced Mutilple AssertionErros.
* Incorrect masterKey invoke condition in DistributedConfigurationStorage
{code:java}
Conditions.revision(MASTER_KEY).eq(ver.get())
{code}
That won't work if there's no MASTER_KEY, which is always true on initial node startup without PDS
was:
After merging meta storage server logic few bugs of ignition start, table creation and atomic operations were revealed. Some of them will be specified in greater detail and fixed within corresponding tickets:
* https://issues.apache.org/jira/browse/IGNITE-14756 - NPE on reading tableConfiguration from ConfigurationManager within SchemaManager
* https://issues.apache.org/jira/browse/IGNITE-14762 - Leader refresh sometimes blocks indefinitely
* https://issues.apache.org/jira/browse/IGNITE-14290 Includes INT32, INT64, etc support
Other issues will be fixed within given ticket:
* Incoorect usage of appliedRevision as watch revision. appliedRevision + 1 should be used instead. That produced Mutilple AssertionErros.
* Incorrect masterKey invoke condition in DistributedConfigurationStorage
{code:java}
Conditions.revision(MASTER_KEY).eq(ver.get())
{code}
That won't work if there's no MASTER_KEY, which is always true on initial node startup without PDS
> Consolidation of managers and metastorage server implementation.
> ----------------------------------------------------------------
>
> Key: IGNITE-14764
> URL: https://issues.apache.org/jira/browse/IGNITE-14764
> Project: Ignite
> Issue Type: Bug
> Reporter: Alexander Lapin
> Assignee: Vladislav Pyatkov
> Priority: Blocker
> Labels: 3.0.0-alpha2, ignite-3
>
> After merging meta storage server logic few bugs of ignition start, table creation and atomic operations were revealed. Some of them will be specified in greater detail and fixed within corresponding tickets:
> * https://issues.apache.org/jira/browse/IGNITE-14756 - NPE on reading tableConfiguration from ConfigurationManager within SchemaManager
> * https://issues.apache.org/jira/browse/IGNITE-14762 - Leader refresh sometimes blocks indefinitely
> * https://issues.apache.org/jira/browse/IGNITE-14290 Includes INT32, INT64, etc support
> Other issues will be fixed within given ticket:
> * Incoorect usage of appliedRevision as watch revision. appliedRevision + 1 should be used instead. That produced Mutilple AssertionErros.
> * Incorrect masterKey invoke condition in DistributedConfigurationStorage
> {code:java}
> Conditions.revision(MASTER_KEY).eq(ver.get())
> {code}
> That won't work if there's no MASTER_KEY, which is always true on initial node startup without PDS
--
This message was sent by Atlassian Jira
(v8.3.4#803005)