ls -alR "cannot allocate memory"

[tucansam]

6.5.3 system, my backup sever.  Last month I was running an rsync-based backup when it threw thousands of CRC errors across all disks.  I haven't messed with it since.

 

Tonight I pulled it apart and reseated everything. 

 

Fired it back up and it came up fine.  Before I start a rsync backup from my main sever, I wanted to generate a bunch of traffic across all disks to see what happens.  From the command line I executed a simple "ls -alR" from /mnt/user and after a few million screens of files, everything errored out with a "cannot allocate memory" error. 

 

I have 2GB RAM installed, no dockers, only a handful of plugins.  htop shows "1.58G" of installed memory.  When I first logged in, there was 1.27G in use.  Then it went down to 1.08G after a few minutes.  As I type this, its at 845M in use.

 

I just now typed the command again, and I am watching RAM in use climb.  It is slowly climbing, 1.02G in use, 1.03, 1.06, 1.07....

 

I didn't figure 'ls' would consume a bunch of memory.  Maybe its the recursive switch and I just don't understand what its doing.

 

I think I have some spare RAM laying around, maybe I'll add more.  Weird though.  I've always thought of unraid, without dockers and VMs, as a super lightweight system.  Never figured 2GB RAM wouldn't be enough.