qpid-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ted Ross (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DISPATCH-90) Routing table is corrupt after a router quickly restarts
Date Mon, 22 Dec 2014 14:09:16 GMT

    [ https://issues.apache.org/jira/browse/DISPATCH-90?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14255764#comment-14255764

Ted Ross commented on DISPATCH-90:

Rather than attempt to fix this specific symptom, I plan to re-design the node tracking component
of the router.  Currently, the router tracks neighbor and remote peers as separate entities.
 When a remote peer becomes a neighbor (or vice versa), the entity is deleted and then re-created.
 This is inefficient and error-prone.
I propose to change the node tracking component to treat all routers equally and simply manage
their peer links (for neighbors) and next-hops (for remotes).

> Routing table is corrupt after a router quickly restarts
> --------------------------------------------------------
>                 Key: DISPATCH-90
>                 URL: https://issues.apache.org/jira/browse/DISPATCH-90
>             Project: Qpid Dispatch
>          Issue Type: Bug
>          Components: Router Node
>            Reporter: Ted Ross
>            Assignee: Ted Ross
>            Priority: Critical
>             Fix For: 0.3
> While testing a six-node topology, I observed that when a router was killed and quickly
restarted (before it could time out on other nodes), the routing tables of other nodes sometimes
became corrupt and did not recover.
> I believe this occurs when a router transitions between being a neighbor and being remote
(reachable by two or more hops).

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: dev-unsubscribe@qpid.apache.org
For additional commands, e-mail: dev-help@qpid.apache.org

View raw message