[NAS-2005] Mail sent whenever scheduling is skipped because the previous scheduling is still running Created: 05/Dec/11  Updated: 30/Jul/12  Resolved: 29/Mar/12

Status: Resolved
Project: NetarchiveSuite
Component/s: None
Affects Version/s: 3.17.0
Fix Version/s: I50, 3.18.3, 3.19.0

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

Inspector: Søren Vejrup Carlsen Søren Vejrup Carlsen (Inactive)
External reference:

NARK-192

Verification:

Easily verified by generating jobs for a broadcrawl using production data (TEST7).
Alternately, ingest 30000 domains into an empty database, and then schedule a broadcrawl.
All harvesters should be shutdown in this test.


 Description   

In TEST7 I am trying to create jobs for a snapshot harvest but the HarvestJobManager is giving me errors of the form

WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running

and no jobs are being created.



 Comments   
Comment by Søren Vejrup Carlsen (Inactive) [ 30/Jul/12 ]

Problem finally fixed in NAS-2088

Comment by Søren Vejrup Carlsen (Inactive) [ 24/Feb/12 ]

The current solution is broken, because it also creates a notification for snapshot harvests, which are only run once.

Comment by Søren Vejrup Carlsen (Inactive) [ 14/Dec/11 ]

As it negates the implementation of NAS-1968

Comment by Søren Vejrup Carlsen (Inactive) [ 13/Dec/11 ]

This actually negates issue NAS-1968!

Comment by Søren Vejrup Carlsen (Inactive) [ 05/Dec/11 ]

How often are we sending these messages?
I think, TLR actually wants a message to be sent, otherwise the problem can be ignored.

Comment by Colin Rosenthal [ 05/Dec/11 ]

I think it's a minor issue that the harvest job name doesn't appear in the log message when a job is generated so you can't grep for it:

WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
FINE: Creating Job 136664 (state = NEW, HD = 155, priority = LOWPRIORITY, forcemaxcount = -1, forcemaxbytes = 100000, forcemaxrunningtime = 0, orderxml = default_orderxml, numconfigs = 10000)
FINE: Generated job: 'Job 136664 (state = NEW, HD = 155, priority = LOWPRIORITY, forcemaxcount = -1, forcemaxbytes = 100000, forcemaxrunningtime = 0, orderxml = default_orderxml, numconfigs = 10000)'
 25217155.dk:defaultconfig
 155.dk:defaultconfig
 36700155.dk:defaultconfig
 155mbit.dk:defaultconfig
 31555555.dk:defaultconfig
 22215545.dk:defaultconfig
 1553kbhv.dk:defaultconfig
 1553k�benhavnv.dk:defaultconfig
 1553københavnv.dk:defaultconfig
WARNING: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running
SEVERE: Mailing netarkivet error: Not creating jobs for harvestdefinition #155 (test7) as the previous scheduling is still running

So grepping on the job-number shows jobs being created but grepping on the harvest name just gives a list of "SEVERE" log messages.

Comment by Colin Rosenthal [ 05/Dec/11 ]

After restarting HarvestJobManager the harvest is now scheduling jobs. This issue may be hard to reproduce or may not have been a misunderstanding caused by the unnecessarily high severity of the log messages ???

Generated at Thu Apr 25 07:33:44 CEST 2024 using Jira 9.4.15#940015-sha1:bdaa9cbecfb6791ea579749728cab771f0dfe90b.