Skip Menu |

This queue is for tickets about the File-Map CPAN distribution.

Report information
The Basics
Id: 130984
Status: rejected
Priority: 0/
Queue: File-Map
People
Owner: LEONT [...] cpan.org
Requestors: Jens.Steidl [...] slub-dresden.de
Cc:
AdminCc:
Bug Information
Severity: (no value)
Broken in: 0.66
Fixed in: (no value)

History Show all quoted text
  Wed Nov 13 05:29:07 2019 Jens.Steidl [...] slub-dresden.de - Ticket created
Subject: checksum mismatch on files 8GiB or larger
We encountered mismatching MD5 checksums of memory mapped files of 8GiB or larger. No issues with smaller files. (tested on debian buster 64bit, perl 5.28, File::Map 0.66, compared using Digest::MD5 & mapped file vs md5sum command line tool)
  Wed Nov 13 16:18:02 2019 LEONT [...] cpan.org - Correspondence added
On Wed Nov 13 05:29:07 2019, Jens.Steidl@slub-dresden.de wrote: Show quoted text
> We encountered mismatching MD5 checksums of memory mapped files of > 8GiB or larger. No issues with smaller files. > > (tested on debian buster 64bit, perl 5.28, File::Map 0.66, compared > using Digest::MD5 & mapped file vs md5sum command line tool)
That is most unexpected. If you have a machine at hand with sufficient RAM, could you try to slurp the file and Digest::MD5 that? That should identify if the problem is File::Map or Digest::MD5. Leon
  Wed Nov 13 16:18:03 2019 The RT System itself - Status changed from 'new' to 'open'
  Thu Nov 14 04:10:30 2019 Jens.Steidl [...] slub-dresden.de - Correspondence added
On further testing with slurping I can confirm that Digest::MD5 is the culprit calculating wrong checksums on large scalars. Thank you for the suggestion, sorry for opening the ticket.
  Thu Nov 14 04:12:36 2019 Jens.Steidl [...] slub-dresden.de - Status changed from 'open' to 'rejected'