Linux administration has a funny rhythm: you’re calm… until a single error message shows up and suddenly you’re speed-scrolling search results like it’s a sport.

This article is a practical guide to the 10 most “googled” Linux admin issue queries—the exact error strings that become evergreen search landing pages—plus safe, copy/paste fixes for each.

Assumptions: Most commands below target Debian/Ubuntu. Where relevant, I include RHEL-family notes.

Table of contents

1. Could not get lock /var/lib/dpkg/lock
2. sudo: unable to resolve host <hostname>
3. SSH: Permission denied (publickey)
4. dpkg was interrupted… dpkg --configure -a
5. df full but du shows less
6. sudo echo "…" > /etc/... permission denied
7. Too many open files / ulimit
8. Temporary failure in name resolution
9. No space left on device but space exists
10. <user> is not in the sudoers file

1) E: Could not get lock /var/lib/dpkg/lock (or lock-frontend)

What’s happening

Another package manager process is running (or crashed and left the system mid-transaction).

Quick fix (safe)

# 1) See what's holding the lock
ps aux | egrep 'apt|apt-get|dpkg' | grep -v egrep

# 2) If nothing meaningful is running, repair the package state
sudo dpkg --configure -a
sudo apt -f install
sudo apt update

If a process is stuck

Try to stop it cleanly before doing anything else:

sudo kill <PID>
sleep 5
sudo kill -9 <PID>   # last resort only

RHEL note: Similar symptoms show up as dnf/yum locks. Use ps aux | egrep 'dnf|yum|rpm' and resolve the active process first.

2) sudo: unable to resolve host <hostname>

What’s happening

Your hostname doesn’t resolve locally—almost always a mismatch between /etc/hostname and /etc/hosts.

Copy/paste fix

# show current values
cat /etc/hostname
hostname
grep -nE '^(127\.0\.0\.1|127\.0\.1\.1)\s' /etc/hosts

Edit the files:

/etc/hostname

myhostname

/etc/hosts

127.0.0.1   localhost
127.0.1.1   myhostname

Then apply:

sudo hostnamectl set-hostname myhostname

3) SSH: Permission denied (publickey)

What’s happening

The server rejected the key you offered. This is usually one of:

• wrong username or wrong key
• authorized_keys missing
• permissions/ownership too open (sshd refuses unsafe .ssh files)
• server policy disables your method
• crypto/algorithm mismatch (older servers, strict policies)

Diagnose fast

Client (shows which key is offered and why it fails):

ssh -vvv user@host

Server logs (choose what exists on your distro):

sudo journalctl -u ssh -n 200 --no-pager || sudo tail -n 200 /var/log/auth.log

Fix #1: permissions/ownership (the repeat offender)

On the server:

sudo chown -R user:user /home/user/.ssh
sudo chmod 700 /home/user
sudo chmod 700 /home/user/.ssh
sudo chmod 600 /home/user/.ssh/authorized_keys

Fix #2: wrong key / missing key

If password login works:

ssh-copy-id -i ~/.ssh/id_ed25519.pub user@host

Fix #3: modern key type (helps with policy mismatch)

ssh-keygen -t ed25519 -a 64

4) E: dpkg was interrupted, you must manually run 'sudo dpkg --configure -a'

What’s happening

A package install/upgrade was interrupted and dpkg needs to finish configuring packages.

Copy/paste fix

sudo dpkg --configure -a
sudo apt -f install
sudo apt update
sudo apt upgrade

If it “hangs”

It’s usually waiting for a prompt (config file prompt, services restart, etc.). Run it in an interactive terminal and answer prompts.

5) df shows full but du shows less

What’s happening

You’re missing disk usage that du can’t “see” as files, most commonly:

• deleted-but-still-open files
• mount points/overlays confusing what you’re counting
• filesystem reserved blocks/overhead

Fix path A: find deleted-but-open files

sudo lsof +L1

If you see huge deleted files, restart the offending service/process (or fix log rotation).

Fix path B: ensure you’re measuring the same filesystem

df -hT
mount | column -t
sudo du -xhd1 /

Fix path C: ext filesystem reserved space

(Verify before changing anything.)

sudo tune2fs -l /dev/<device> | egrep 'Reserved block|Reserved block percentage|Block count'

6) sudo echo "…" > /etc/... permission denied

What’s happening

Shell redirection (>) happens before sudo, so the write is attempted as your non-root user.

Correct patterns

Best default: tee

echo 'myline' | sudo tee /etc/somefile >/dev/null

Or: root shell for the whole command

sudo bash -c "echo 'myline' > /etc/somefile"

7) Too many open files / ulimit issues

What’s happening

The process hit its file descriptor limit (common for web servers, proxies, databases, and high-connection workloads).

See the current limit

ulimit -n
cat /proc/$(pidof <service> | awk '{print $1}')/limits | sed -n '1,60p'

Fix for systemd services (recommended)

sudo systemctl edit <service>

Add:

[Service]
LimitNOFILE=65535

Apply:

sudo systemctl daemon-reload
sudo systemctl restart <service>

Tip: Also check the app config (e.g., Nginx worker_connections) so it matches reality—raising one knob while the other stays low is how you get the same error again.

8) Temporary failure in name resolution

What’s happening

DNS resolution is broken (not necessarily networking).

Diagnose: IP vs name

ping -c2 1.1.1.1
ping -c2 google.com
cat /etc/resolv.conf

Common fixes

Restart the resolver:

sudo systemctl restart systemd-resolved 2>/dev/null || true

If you’re on a server with static config, verify your network stack (netplan/NetworkManager) is actually applying DNS servers and not being overwritten by cloud-init, VPN clients, or stale symlinks.

9) No space left on device even though space exists

What’s happening

Often: you ran out of inodes, not bytes.

Confirm inodes

df -h
df -i

Find inode-heavy directories

sudo du -x --inodes -d1 / 2>/dev/null | sort -n | tail -20

Other culprits:

• tmpfs full (/run, /tmp)
• runaway app creating millions of tiny files
• container overlay layers/log growth
• quotas

10) <user> is not in the sudoers file. This incident will be reported.

What’s happening

Your user doesn’t have sudo rights (or sudoers config is broken).

Confirm group membership

id -nG

Fix (Debian/Ubuntu)

From root (console/recovery/another admin):

su -
usermod -aG sudo <username>

Then log out/in (or reboot).

If sudoers is corrupted

Use:

visudo

If sudo is completely unusable, fix from recovery/rescue mode.

Bonus: “First response” triage bundle (paste this first)

echo "== pkg state =="
ps aux | egrep 'apt|apt-get|dpkg|dnf|yum|rpm' | grep -v egrep || true

echo "== net vs dns =="
ping -c1 1.1.1.1 || true
ping -c1 google.com || true
cat /etc/resolv.conf || true

echo "== disk bytes + inodes =="
df -h
df -i

echo "== deleted-but-open files (top) =="
sudo lsof +L1 2>/dev/null | head -50 || true