Description of problem: We overestimate the size of the object for reads with trimtrunc, which results in the hitting assert((offset + length) <= (range.first.get_off() + range.first.get_len())) during reads. Actual results: The above failure seen in the multimds suite with EC pools. Expected results: No failure during EC pool reads.
Verified in multimds run http://pulpito.ceph.redhat.com/vasu-2019-02-17_21:10:34-multimds-luminous-distro-basic-bruuni/
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:0475