I am trying to install Tensorflow 2 on a Linux virtual machine on Oracle Virtualbox using Python 3.8.2. The machine has the following characteristics:
Operating system: Ubuntu 20.04 LTS (64-bit)
GNOME Version: 3.36.1
Windowing System: X11
Memory: 9.5 GiB
Disk Capacity: 10.7 GB
When doing pip install tensorflow==2.2.0 in the terminal, I get the following error at the very last stage (after the download is completed):
ERROR: Could not install packages due to an EnvironmentError: [Errno 28] No space left on device
Typing du -sh in the terminal yields
12K .
while df yields
Filesystem 1K-blocks Used Available Use% Mounted on
udev 4952816 0 4952816 0% /dev
tmpfs 996192 1324 994868 1% /run
/dev/sda5 9736500 6919960 2302236 76% /
tmpfs 4980940 0 4980940 0% /dev/shm
tmpfs 5120 4 5116 1% /run/lock
tmpfs 4980940 0 4980940 0% /sys/fs/cgroup
/dev/loop0 56320 56320 0 100% /snap/core18/1705
/dev/loop1 56320 56320 0 100% /snap/core18/1754
/dev/loop2 246656 246656 0 100% /snap/gnome-3-34-1804/24
/dev/loop3 261760 261760 0 100% /snap/gnome-3-34-1804/36
/dev/loop4 63616 63616 0 100% /snap/gtk-common-themes/1506
/dev/loop5 51072 51072 0 100% /snap/snap-store/433
/dev/loop6 51072 51072 0 100% /snap/snap-store/454
/dev/loop7 27776 27776 0 100% /snap/snapd/7264
/dev/loop8 31104 31104 0 100% /snap/snapd/7777
/dev/sda1 523248 4 523244 1% /boot/efi
tmpfs 996188 24 996164 1% /run/user/1000
I have tried to follow https://www.maketecheasier.com/fix-linux-no-space-left-on-device-error/ and https://github.com/pypa/pip/issues/5816 , but none of their solutions seems to work for me. Can anyone help me figure out what is wrong?
pip
downloads files to the temporary directory /tmp
during pip install
.
The error implies that /tmp
ran out of space during installation. Looking at the df
output shows that /tmp
isn't mounted as a tmpfs
mount point, indicating that /tmp
is currently mounted under /
.
The available storage for /
is 2,302,236 1K blocks (from your df
output), which is about 2.3 GB (next time use df -h
for more readable values).
But by executing:
sudo mount -t tmpfs tmpfs /tmp
a tmpfs
is mounted to /tmp
. Typically, a tmpfs
partition has its maximum size set to half of the available RAM, which in your case would be 4.75GB, essentially doubling the available space available to pip
.
probably an inode issue. Try a
df -i
or a number of files open too high Try and kill process related:
find /proc/*/fd -ls | grep '(deleted)'
I am also running Ubuntu 20.04 LTS (64-bit) on Oracle VirtualBox and started getting warnings about low disk space.
The culprit in my case was the /var/cache/apt/archives
directory which was taking up 771 MB, it might be the same problem on your system. The cause is obscured by the fact that /var
is a directory under /
so the storage used is aggregated there and not as a separate mount point.
I used the big-hammer approach sudo apt clean
and cleared the cache, trusting that apt
will re-download later any packages it might need. The storage used went down from 771 MB to 36 KB.
The answer is too late to help the OP, hopefully it will help someone else.
$ sudo du -ch /var/cache/apt/archives/
4.0K /var/cache/apt/archives/partial
771M /var/cache/apt/archives/
771M total
$ sudo apt clean
$ sudo du -ch /var/cache/apt/archives/
4.0K /var/cache/apt/archives/partial
36K /var/cache/apt/archives/
36K total
$ sudo du -ch -d 1 /var/cache
40K /var/cache/apt
1.8M /var/cache/man
40K /var/cache/dictionaries-common
56K /var/cache/ldconfig
8.0K /var/cache/PackageKit
2.6M /var/cache/fontconfig
20K /var/cache/cups
15M /var/cache/app-info
5.2M /var/cache/debconf
2.1M /var/cache/fwupd
2.2M /var/cache/snapd
6.1M /var/cache/cracklib
4.0K /var/cache/gdm
484K /var/cache/private
6.7M /var/cache/apparmor
42M /var/cache
42M total
$
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.