Closed james-cook closed 2 years ago
A backtrace would be helpful to see what step it's actually on. Otherwise I'd just be making guesses.
I'm very happy to report that there is NO problem with rmlint!
The run just took 14 hours which was longer than I was expecting.
By using -vvv
I could be sure that something was happening.
I took a copy of rmlint.sh to check for loops at some point.
The resulting rmlint.sh is ca. 650MB and rmlint.json is 1.3GB.
The run completed:
DEBUG: Freeing device 2128 (pointer 0x456ec0)
Waiting for progress counters to catch up...Done
DEBUG: Remaining 0 bytes in 0 files
DEBUG: Dupe search finished at time 51109.383
Dragan Aleksic _ Matthes & Seitz Berlin-00-Döbler,Katharina.mp3'
==> Note: Please use the saved script below for removal, not the above output.
==> In total 2434399 files, whereof 1157270 are duplicates in 454040 groups.
==> This equals 383.98 GB of duplicates which could be removed.
==> 278702 other suspicious item(s) found, which may vary in size.
==> Scanning took in total 14h 11m 49.387s.
Sorry about the false alarm
In my current run preprocessng seems to get stuck.
The HDD is working away. No spin down. But the numbers in the preprocessing line do not change, even after many hours. This happens with master and with develop + some patches.
The original command:
rmlint --progress --xattr --keep-all-tagged --must-match-tagged '/srv/dev-disk-by-label-OMV2/shd2/from.ext.hdd.M-family.TOSH_SOURCE-HDD_ORPHANS/from.ext.hdd.F' // '/srv/dev-disk-by-label-OMV2/shd2/from.ext.hdd.F'
The first dir is "new" and contains no hardlinks of any kind (internal to the directory or outside the directory). I more or less assume that all files in dir1 are inside dir2 - I am using rmlint to check this for sure. The second dir is the product of many
rmlint -c sh:hardlink
runs and contains many repeated inodes.Preprocess output gets "stuck" here:
Thinking there might be a very large file on which rmlint is taking so much time I limited the max file size to 2GB with
-s -2GB
in the command line. This doesn't change much, rmlint is stuck preprocessing at:I am now running the same job overnight with -vvv instead of --progress to see what emerges.
Any tips on how to find if and why preprocessing may be stuck in a loop e.g. Debug line in preprocess.c?