Mate 22.04 Losing Disk Space on ROOT

Support & Help Requests
#1

This problem just started a week ago...
When I first boot up and log in, ROOT (system files) is 51% free space. I will leave it sit over night, not doing nothing to it. And in the morning I will have a WARNING window saying LOW disk space on ROOT. 1.2Gb or lower. The partition was formatted 30Gb.
I clean up my TEMP file but not that much about 150Mb. I clean up my JOURNAL files down to 200Mb. Daily. I clean out APT cache and clean up SNAP files. Daily.
This keeps happening on a daily basis. What more can I do, and why is this happening every day? I have left the system on for weeks on end and never had this problem.
Any and all HELP is appreciated.
THANKS.

P.S. HOME is on it's own partition. And SWAP also.

1 Like
#2

Well, recollection of what has been done to your system a week ago could give some clues regarding the problem's root cause.

Next, Applications > System Tools > Mate Disk Usage Analyser can help to find out what files or directories actually grow that much.

Since that lsof command can be used to identify a process which writes into these files.

Good luck!

3 Likes
#3

Here are screenshots of last night and then this morning. PC just sitting idle over night, NO programs running.

Screenshot at 2023-09-15 20-32-54
Screenshot at 2023-09-15 20-32-54952×652 132 KB

Screenshot at 2023-09-16 05-42-46
Screenshot at 2023-09-16 05-42-46952×652 112 KB

Something is eating up disk space. Free space will fall to less than 1Gb. in 24-48 hours.

#4

I ran your command suggestion, I don't understand the output. What am I looking for?
OUTPUT:
administrator@MATE:/$ ls -of
lib64 etc run snap .. . dev lost+found usr proc sbin libx32 sys home opt tmp boot mnt root lib32 bin var media lib srv cdrom

#5

No, I meant lsof command like that:

ugn@evm:~$ lsof | grep -i django
soffice.b 75651                        ugn   20u      REG                8,6     420405   20321877 /home/ugn/Desktop/Django Web Framework for DB-driven apps.odt
...

it shows which command and which process has opened which file.
Ok, probably, iotop is better suited for continuous HD monitoring.

Hopefully, this can be of some help: ubuntu which process is using disk - Google Search

There are a lot of system and/or background processes which can write to disk. For example, file system indexing service, etc, etc, etc...

1 Like
#6

These outputs were at the top of the list, and the highest write figures:
270 be/3 root 0.00 B 2.80 M ?unavailable? [jbd2/sda7-8]
317 be/3 root 8.00 K 204.64 M ?unavailable? systemd-journald
638 be/3 root 0.00 B 1700.00 K ?unavailable? [jbd2/sda8-8]
945 be/4 syslog 0.00 B 13.37 M ?unavailable? rsyslogd -n -iNONE [rs:main Q:Reg]
See anything out of place or strange? This was for a 10 min. period.
I think this explains why my JOURNAL files are so big.

#7

Do I get it right that you have already found out that disk space is consumed by journal files?

If yes, I'd try to find out what are messages which overflow journals and eliminate the reason. You see, I've stumbled upon cases when log files grow endlessly due to a lot of errors reported by background process.

2 Likes
#8

NO, I'm still looking for answers. But the journal files are a good starting point. I have my system log files set to rotate and delete every 7 days.

#9
1 Like