Incident #10909: Regular Failure of Dataminer "Garr" (2 out of 3 execution attempts) - Internal Server Error - D4Science Infrastructure - D4science

Actions

Copy link

Incident #10909

closed

Regular Failure of Dataminer "Garr" (2 out of 3 execution attempts) - Internal Server Error

Added by Emmanuel Blondel over 7 years ago. Updated over 7 years ago.

Status:

Closed

Priority:

High

Assignee:

_InfraScience Systems Engineer

Category:

High-Throughput-Computing

Target version:

UnSprintable

Start date:

Jan 13, 2018

Due date:

% Done:

100%

Estimated time:

Infrastructure:

Production

Description

If I run successively WPS requests, only an average of 1 out of 3 requests are working...This occurred in the past and some cluster instance was not working.

Are all cluster instances up and running? If yes, Isn't there a way to avoid delegating a request to a cluster instance that is down? otherwise we lose the advantage of load balancing...

https://dataminer.garr.d4science.org/wps/WebProcessingService?request=Execute&service=WPS&Version=1.0.0&lang=en-US&Identifier=org.gcube.dataanalysis.wps.statisticalmanager.synchserver.mappedclasses.transducerers.MPA_INTERSECT_V4&gcube-token=yourtoken&DataInputs=Report_Format=json;MPA_Shapefile_Url=https%3A%2F%2Fabsences.zip;Marine_Boundary=EEZ;Region_Id=8417;Selected_Data_Feature=seamounts%2Cguyots%2Ccanyons%2Cridges%2Cplateaus

Related issues

Actions

Copy link

Updated by Andrea Dell'Amico over 7 years ago

Status changed from New to In Progress

Emmanuel Blondel wrote:

If I run successively WPS requests, only an average of 1 out of 3 requests are working...This occurred in the past and some cluster instance was not working.

Are all cluster instances up and running? If yes, Isn't there a way to avoid delegating a request to a cluster instance that is down? otherwise we lose the advantage of load balancing...

We have a global problem that also affects the dataminer in a way that the load balancer cannot detect. We are working on it.

Actions

Copy link