Linux Bug 5% to 97% storage growth

Some problem with Linux took what was 5% of the SSD to 97% and rendered my server inoperable for obvious reasons. I was able to recover my files but there is a problem that I have not been able to put my finger on.

Looking at df -h showed the database was huge but in MySQL nothing seems amis. I am using MySQL 8 which is the newer version.

I will have to install certbot fresh once the server is reinstalled

Hi @HardcoreGames,

This is probably not a Let's Encrypt question. :slight_smile:

Consider a command like

du -a | sort -n

to find the individually largest files and directories on your filesystem.

3 Likes

I agree with this.

However, this might be a good opportunity to warn @HardcoreGames to also backup the /etc/letsencrypt/ directory and restore it after the reinstall.

2 Likes

the file system was damaged so I was not able to recover my certificates

maybe I need to make a bash script too gzip everything to make recovery faster

that and one to install linux fresh with my software stack

certbot.eff.org is the bookmark I have

I don't understand, you said this:

Or aren't certificates counted as files? :roll_eyes:

I was not able to gzip anything as the file system was unwritable

I have daily backups of the /var/www and my personal file but that is all that I have mechanized

like I said the system storage was overwhelmed, not sure what happened

That's where USB sticks, DVD drives and/or remote shares come in :slight_smile:

1 Like

I have vsftpd installed which is how i backup my files ordinarily

Should work fine for /etc/letsencrypt/

I will add that to my backup script but I believe that certbot can figure out a certificate for the server

my worry now is to discover what is gobbling up all of the storage

What does that mean? You mean issue a new one? If you got the certificates and private key lying around, it's a shame to issue a new one and add unnecessary load to the Let's Encrypt infrastructure.

I have wondered how to mechanize the update process as chron is a daily event but certificates are 90 day elements which means chron is in need of some modification

no scheduled events that I am aware of

certbot is usually run through cron twice a day. Certbot consequently will determine if a cert needs to be renewed, not cron.

checking a server twice daily is a lot of workload

Reading a handful of files twice daily is "a lot of workload"? Heck, an Arduino could manage that without noticing the difference.

next up is a vhost file for the website

<VirtualHost *:80>
    ServerName hardcoregames.ca
    ServerAlias www.hardcoregames.ca
    DocumentRoot /var/www/hardcoregames.ca/index.php
    <Directory var/www/hardcoregames.ca/
        Options -Indexes +FollowSymLinks
        AllowOverride All
    </Directory>
    #ErrorLog ${APACHE_LOG_DIR}/domain1.com-error.log
    #CustomLog ${APACHE_LOG_DIR}/domain1.com-access.log combined
</VirtualHost>

Next up? What's the purpose or goal of this thread?

Without any I'm inclined to close it..

seem that apache2 is not bring up the site as usual

so now the fun is figuring out what us wrong

looking in apache2/sites-enabled I see the ln to my conf file as above


 hardcoregames.ca.conf -> ../sites-available/hardcoregames.ca.conf

As it seems this thread hasn't got a specific link to Let's Encrypt and as this is the #help section of the Let's Encrypt Community (and not a general forum), I'm closing this thread.

2 Likes