MSI Server Bug - Freezes on Large File(s) Copy
The MSI Server has been experiencing an issue where it freezes and becomes completely unresponsive after a few minutes of copying a large amount of files.
Reproduce the Bug
To reproduce, you can use the following steps:
# ###############################################################
# Check drive mount points, specifically the T7
lsblk
# Modify sdb1 if necessary then mount the T7
sudo mount /dev/sdb1 /media/gqc/T7 -o umask=0022,gid=1000,uid=1000
# CD into the working directory
cd CCTV/SD1/
# Remove the previously copied files
rm -rf SD1_blurred_frames_unzipped
# Start the copy process
cp -r /media/gqc/T7/cctv/SD1/SD1_blurred_frames_unzipped .
# This should run for about 5 minutes before freezing up...
# ###############################################################
Attempted Solutions
These posts have some suggestions, most of which have been tested and did not help:
- system freezes unresponsive unusable when copying large file to usb
- why is my pc freezing while im copying a file to a pendrive
- https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1208993
The following attempts have been made to avoid the errors:
# ----------------------------------------------------------------------
# 1) Clear the RAM cache periodically as the cp command is running:
sudo sysctl -w vm.drop_caches=3
# ----------------------------------------------------------------------
# 2) Lower swappiness and raise min free kbytes:
sudo sysctl -w vm.min_free_kbytes=365536
sudo sysctl -w vm.swappiness=10
# ----------------------------------------------------------------------
# 3) Modify dirty bytes:
sudo su
echo $((16*1024*1024)) > /proc/sys/vm/dirty_background_bytes
echo $((48*1024*1024)) > /proc/sys/vm/dirty_bytes
exit
# ----------------------------------------------------------------------
# 4) Completely disable swap:
sudo swapoff -a
Next Steps
- Test copying from a different USB drive (not T7)
- Look into the file systems (ext4 should be most stable)
- Look into disk corruption
- Look into other possible hardware faults?