After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 773068 - Endless loop when evaluating folder size
Endless loop when evaluating folder size
Status: RESOLVED OBSOLETE
Product: nautilus
Classification: Core
Component: File Properties Dialog
3.26.x
Other Linux
: Normal normal
: future
Assigned To: Nautilus Maintainers
Nautilus Maintainers
Depends on:
Blocks:
 
 
Reported: 2016-10-17 06:56 UTC by Andrea Vai
Modified: 2018-05-15 14:12 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Andrea Vai 2016-10-17 06:56:08 UTC
If I select many folders and get properties for them, the count never ends: it increases until reaching some (random) value and begins again.

The problem seems to be related to the number of foders selected (and/or the amount of data contained): if I select only a few folders, containing just some GB, the bug doesn't happen.

Running Nautilus 3.14.3 on Ubuntu 16.04
Comment 1 Ernestas Kulik 2016-10-17 07:22:09 UTC

*** This bug has been marked as a duplicate of bug 755483 ***
Comment 2 Andrea Vai 2016-10-17 07:26:11 UTC
I'm not sure that this is a duplicate of bug 755483 because I experience the bug in a folder structure without symlinks
Comment 3 Ernestas Kulik 2016-10-17 07:49:33 UTC
(In reply to Andrea Vai from comment #2)
> I'm not sure that this is a duplicate of bug 755483 because I experience the
> bug in a folder structure without symlinks

Could you try and come up with a reproducer for this?
Comment 4 Andrea Vai 2016-10-17 08:57:44 UTC
I would really love to, but I am experiencing the bug in a set of personal file folders (containing primarily photos and videos, without symlinks (*)). I can reproduce the bug also with a set of folders inside the /etc/ directory (note that here inside there are symlinks (*)):

- selecting all folders from A to S --> bug
  - select folders from A to I --> no bug
  - select folders from J to S --> no bug

Let me know if you have any other idea to suggest me to try.

(*) Please note that I use the command 

find . -type l

to query if a directory tree has some symlinks inside.
Comment 5 Ernestas Kulik 2016-10-17 13:12:52 UTC
(In reply to Andrea Vai from comment #4)
> I would really love to, but I am experiencing the bug in a set of personal
> file folders

A file count and total size would be a step in the right direction. Anything concrete to go by, really.
Comment 6 Andrea Vai 2016-10-19 10:39:57 UTC
I can reproduce the problem with some 25 folders, summing up to 500 files and about 2.5GB.
Comment 7 Andrea Vai 2016-10-27 11:26:09 UTC
Some further investigation and some scripts ([1], [2], [3]) did not help me in finding a way to determinedly reproduce the bug, but seem to point out that

- the problem is not (only) related to the number of items involved, because it does not happen with ~1M items (for example created by [1]) , but it happens with some 500 total files;

- the problem is not (only) related to the depth level of the nested folders, because it does not happen with 3 levels (and ~1M items total) (for example created by the script [2]), but it happens with two level depth (i.e. dir foo and foo/bar1, foo/bar2, foo/bar3, ... --> the dir "foo" gets the bug);

- the problem is not (only) related to the size of the file(s), because it does not happen with (~10k) files for example created by script [3] (which makes multiple copies of a 1.4M image, summing up the folder size to 14G), but it happens with a 2.5GB folder;

- I can reproduce the problem in NTFS or ext4 filesystems, and both using a usb external drive and a local drive.

Let me know if you have any other idea for any more testing scenarios.

---
[1] for i in {1..1000}; do mkdir $i; for j in {1..1000}; do echo $i-$j > 
$i/$j; done; done

[2] for i in {1..100}; do mkdir $i; for j in {1..100}; do mkdir $i/$j; for k 
in {1..100}; do echo $i-$j-$k > $i/$j/$k; done; done; done

[3] for i in {1..100}; do mkdir $i; for j in {1..100}; do cp img.jpg 
$i/img-$i-$j.jpg ; done; done
Comment 8 Alexandre Franke 2016-11-24 19:06:10 UTC
(In reply to Andrea Vai from comment #0)
> Running Nautilus 3.14.3 on Ubuntu 16.04

Can you check that again? Ubuntu 16.04 should have 3.18.x.
Comment 9 Andrea Vai 2016-11-25 11:34:21 UTC
yes, here is it:

$ nautilus --version
GNOME nautilus 3.14.3

$ lsb_release -a
No LSB modules are available.
Distributor ID:	Ubuntu
Description:	Ubuntu 16.04.1 LTS
Release:	16.04
Codename:	xenial

may I have to repair anything, and which way?
Comment 10 Alexandre Franke 2016-11-25 11:54:05 UTC
Can you update your system?
Comment 11 Andrea Vai 2016-11-25 11:59:20 UTC
I would like to, but 

$ sudo apt-get update && sudo apt-get upgrade

tells me there's nothing to do, so I'm afraid there is something broken, but don't know how to find and fix it... I don't know if this is the right place to ask, though
Comment 12 Alexandre Franke 2016-11-25 16:16:04 UTC
(In reply to Andrea Vai from comment #11)
> I would like to, but 
> 
> $ sudo apt-get update && sudo apt-get upgrade
> 
> tells me there's nothing to do, so I'm afraid there is something broken, but
> don't know how to find and fix it... I don't know if this is the right place
> to ask, though

Nevermind, I misread http://packages.ubuntu.com/xenial/nautilus and you're right, Ubuntu 16.04 has Nautilus 3.14.

That version is really old though. We released 3.16, 3.18. 3.20 and our current stable release 3.22. Note that Ubuntu 16.10 has 3.20, so maybe you could upgrade to this and try again? Of course I understand that you may want or have to stay with the LTS.

Given the old version and the fact that we are unable to come with a way to reproduce the issue on our end, the only reasonable course of action I see for now is to close this report as OBSOLETE. You can then reopen it if you manage to reproduce the issue with a recent version. Maybe try with a Live Media of Ubuntu 16.10 or Fedora 25, since you have access to the hardware configuration where the problem occurs. If you do that and the issue is gone, confirmation would be welcome too.
Comment 13 Andrea Vai 2018-05-15 14:10:24 UTC
Reopening because still exists in Nautilus 3.26.3.1 (tested on Fedora 27).
Comment 14 Ernestas Kulik 2018-05-15 14:12:53 UTC
Since we’ve moved to GitLab, I’m going to close this in favor of https://gitlab.gnome.org/GNOME/nautilus/issues/363.