cmssw: Fatal Root Error: @SUB=TBasket::ReadBasketBuffers

Hi all,

With one production workflow, we are having the following issue while reading input files:

[a] Fatal Root Error: @SUB=TBasket::ReadBasketBuffers
fNbytes = 15438, fKeylen = 137, fObjlen = 667416, noutot = 0, nout=0, nin=15301, nbuf=667416

Workflow: https://dmytro.web.cern.ch/dmytro/cmsprodmon/workflows.php?prep_id=ReReco-Run2017E-DoubleMuon-UL2017_MiniAODv2-00002 Job log: https://cms-unified.web.cern.ch/cms-unified/joblogs/cmsunified_ACDC1_Run2017E_DoubleMuon_UL2017_MiniAODv2_210604_153724_982/8021/DataProcessing/2a7c2fd2-3cbc-47b8-ad9b-9a3017657f50-0-2-logArchive/

We discussed the issue on JIRA and Matti suggested opening this issue. https://its.cern.ch/jira/browse/CMSCOMPPR-18887

Many thanks, Hasan

About this issue

Original URL
State: open
Created 3 years ago
Comments: 31 (29 by maintainers)

Most upvoted comments

Thanks @haozturk, I was in fact able to reproduce the issue with that file. I’ll run some more tests.

makortel on Apr 1, 2022

Its a problem with an input file, yes. (Likely one of the n)

On Nov 1, 2021, at 2:23 PM, Hasan Öztürk @.***> wrote:

For my education; does this error occur because there is something wrong w/ the input file? If so, what’s exactly wrong?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

davidlange6 on Nov 1, 2021

There is something ‘odd’ with the file. It is an older file that is not well clustered (pre-date the code that strengthen the entry clustering in TTree) and TTreeCache is failing to properly predict which basket to read.

pcanal on Nov 1, 2021