[NAS-2474] Postgresql breaks connections to client prematurely Created: 26/Nov/15  Updated: 15/Jan/16  Resolved: 15/Jan/16

Status: Resolved
Project: NetarchiveSuite
Component/s: Common
Affects Version/s: 4.4, 5.0
Fix Version/s: 5.1

Type: Bug Priority: Major
Reporter: Søren Vejrup Carlsen (Inactive) Assignee: Søren Vejrup Carlsen (Inactive)
Resolution: Cannot Reproduce  
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

External reference:

https://sbprojects.statsbiblioteket.dk/jira/browse/NARK-888


 Description   

We have seen this exception some times:

Exception while schedulingharvestdefinition #241(2015-4-10GB). The harvestdefinition has been deactivated!
dk.netarkivet.common.exceptions.IOFailure: SQL Error while reading domain oeis.dk
SQLException trace:
SQL State:57P01
Error Code:0
org.postgresql.util.PSQLException: FATAL: terminating connection due to administrator command

And there is nothing to seen in the database log except for lines like these

2015-11-19 11:33:01.520 CET # LOG:  unexpected EOF on client connection
2015-11-19 11:33:01.520 CET # LOG:  unexpected EOF on client connection
2015-11-19 11:33:01.520 CET # LOG:  unexpected EOF on client connection
2015-11-19 11:33:01.521 CET # LOG:  unexpected EOF on client connection


 Comments   
Comment by Søren Vejrup Carlsen (Inactive) [ 26/Nov/15 ]

An theory of mine is that the client keeps the connection open for too long a time.

I will try and make the maxConnetionAge configurable in the NAS settings, and the default to 1 hour

Generated at Wed Apr 24 15:24:58 CEST 2024 using Jira 9.4.15#940015-sha1:bdaa9cbecfb6791ea579749728cab771f0dfe90b.