[
https://issues.apache.org/jira/browse/ATLAS-2092?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16149210#comment-16149210
]
Ashutosh Mestry edited comment on ATLAS-2092 at 8/31/17 4:45 PM:
-----------------------------------------------------------------
[~grahamwallis] I am reading the PDF now. Analysis is comprehensive and I find
it interesting read. Thanks for putting this together!
One thing we have noticed is that version of _Berkeley DB_ we are using has a
few problems. Mostly around transactions commits. Only yesterday we were
contemplating on effort required to move our tests to use embedded HBASE and
embedded Solr. We have not reached any conclusion yet. Doing this would help us
get around some of the intermittent failures we see in test environments. It
will also make our unit & IT tests run in an environment that is close to
production.
How easy is it for you to duplicate this problem on HBASE/Solr setup? It would
be interesting to see the results.
About 'double wrapping of transactions' (Appendix, Observation 2). I had worked
on trying to resolve it. Please find patch attached.
was (Author: ashutoshm):
[~grahamwallis] I am reading the PDF now. Analysis is comprehensive and I find
it interesting read. Thanks for putting this together!
One thing we have noticed is that version of _Berkeley DB_ we are using has a
few problems. Mostly around transactions commits. Only yesterday we were
contemplating on effort required to move our tests to use embedded HBASE and
embedded Solr. We have not reached any conclusion yet. Doing this would help us
get around some of the intermittent failures we see in test environments. It
will also make our unit & IT tests run in an environment that is close to
production.
How easy is it for you to duplicate this problem on HBASE/Solr setup? It would
be interesting to see the results.
> Failures following concurrent updates
> -------------------------------------
>
> Key: ATLAS-2092
> URL: https://issues.apache.org/jira/browse/ATLAS-2092
> Project: Atlas
> Issue Type: Bug
> Components: atlas-core
> Reporter: Graham Wallis
> Attachments: Investigations and findings relating to concurrent
> updates in Atlas.pdf
>
>
> There is a race condition that causes duplication of schema vertices as a
> result of concurrent graph updates. This in turn leads to failure of queries
> that specify a type such as an edge label used in an attribute that
> references another entity. This problem is known to affect Atlas entity refs
> – which create graph edges that use edge label schema vertices. It is likely
> that it also affects other types in Atlas.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
