[NETCONF-133] "Transaction ... is no longer running" after user sends invalid data. Created: 26/Jan/16  Updated: 15/Mar/19  Resolved: 12/Feb/16

Status: Resolved
Project: netconf
Component/s: netconf
Affects Version/s: None
Fix Version/s: None

Type: Bug
Reporter: Jozef Behran Assignee: Andrej Mak
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Operating System: All
Platform: All


External issue ID: 5096

 Description   

Tested on: stable/beryllium
Caught by: Netconf MDSAL Northbound suite

After user sends invalid data to the "modules" node of ODL's NETCONF configutation and after that request errors out (uncleanly, see NETCONF-132), all other operations in that session fail with

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<rpc-reply message-id="6" xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">
<rpc-error>
<error-type>application</error-type>
<error-tag>operation-failed</error-tag>
<error-severity>error</error-severity>
<error-message>Unexpected error</error-message>
<error-info>
<error>java.lang.IllegalStateException: Transaction DOM-31 is no longer running</error>
</error-info>
</rpc-error>
</rpc-reply>

(the "DOM-31" in the <error> element can vary).



 Comments   
Comment by Ryan Goulding [ 02/Feb/16 ]

Is this still present?

Comment by Jozef Behran [ 03/Feb/16 ]

The "invalid data" are for example the following file in Gerrit:

https://git.opendaylight.org/gerrit/#/c/27672/19/csit/variables/netconf/MDSAL/merge-multiple-replace-request.msg

It contains a complete request which, when sent to ODL, will trigger the bug.

The bug is still there as of today. A Sandbox test run just finished on build 3866 and the bug can still be seen there.

Comment by Ryan Goulding [ 03/Feb/16 ]

Is the workaround just to start a new session? I ask since this sounds pretty bad, but I am not sure if its the type of thing we should raise to a critical/blocker level.

Thanks for your feedback, Jozef.

Comment by Tomas Cere [ 03/Feb/16 ]

New session should fix it, the problem is that the failure of the commit is not correctly handled.

Comment by Andrej Mak [ 04/Feb/16 ]

https://git.opendaylight.org/gerrit/#/c/34089/

Generated at Wed Feb 07 20:14:15 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.