Home | News | Documentation | Download

Follow up on RAM- vs disk-resident TTree compression bug

Hi ROOTers,

I would like to follow up on the discussion in TTree compression factor 1.00 that ended without conclusion.

As of the ROOT version I tried below the bug is still present. Has this bug been fixed, and if so, in which version can I start benefiting from it?

If not, are there plans and ETA for the fix?


ROOT Version: /cvmfs/cms.cern.ch/slc7_amd64_gcc820/lcg/root/6.18.04-nmpfii2/
Platform: slc7_amd64_gcc820
Compiler: GCC 8.3.1

the original thread that you link to contains a workaround, and it does not seem like a bug report was ever filed, so I don’t think there was any further action on the topic.

@pcanal is currently off, but he might comment further when he’s back.
Please ping us if again if you don’t hear anything in 2-3 weeks, or feel free to open a jira ticket.


Hi Enrico,

Ok, that sounds fine. Let’s see what @pcanal has to say on the matter.

While I’ve verified that the workaround is indeed effective, I have some preference towards RAM-residence, which led me to realize the compression settings had no effect whatsoever and then that this is an old issue.