[
https://issues.apache.org/jira/browse/ARTEMIS-4801?focusedWorklogId=922426=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-922426
]
ASF GitHub Bot logged work on ARTEMIS-4801:
---
Author: ASF GitHub Bot
Created on: 06/Jun/24 20:54
Start Date: 06/Jun/24 20:54
Worklog Time Spent: 10m
Work Description: tabish121 opened a new pull request, #4965:
URL: https://github.com/apache/activemq-artemis/pull/4965
When caching address query results the remote session can be blocked forever
from creating links on an address if the "does not exist" value is cached since
it is never updated again and will always report "does not exist" even if the
address is added manually via management later. The cache state can cause other
issues for long running sessions as well and should be removed to avoid attach
failures for cases where the current broker state could allow the attach to
succeed but the cached entry won't allow it.
Issue Time Tracking
---
Worklog Id: (was: 922426)
Remaining Estimate: 0h
Time Spent: 10m
> AMQP Session address query cache can have invalid state for long lived
> sessions
> ---
>
> Key: ARTEMIS-4801
> URL: https://issues.apache.org/jira/browse/ARTEMIS-4801
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Components: AMQP
>Affects Versions: 2.34.0
>Reporter: Timothy A. Bish
>Assignee: Timothy A. Bish
>Priority: Major
> Fix For: 2.35.0
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> The AMQPSessionCallback retains a map of AddressQueryResult instances for
> previous address checks. This map is not updated if the state of broker
> addresses changes, neither addresses added, removed or updated states are
> reflected in the cache. This leads to issues for long running sessions where
> a link attach may fail for a non-existent address and on a later attempt
> should succeed if the address was added but can't because the cache will
> still hold the non-exists query result. Other scenarios are possible such as
> an address removed and re-added with different routing type but the former
> case is more serious.
> The cache should be removed and if a similar optimization is actually found
> to be needed a better mechanism should be chosen to avoid the issues found.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
-
To unsubscribe, e-mail: issues-unsubscr...@activemq.apache.org
For additional commands, e-mail: issues-h...@activemq.apache.org
For further information, visit: https://activemq.apache.org/contact
