From my monitoring server I received several SSH alerts from another Linux server, running CentOS 6.3.
The alerts were meanwhile flapping between CRITICAL and OK:
[08:48:26] SERVICE ALERT: target-server;SSH;CRITICAL;HARD;3;Server answer:
[08:47:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;2;Server answer:
[08:46:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;1;Server answer:
[08:43:26] SERVICE ALERT: target-server;SSH;OK;HARD;3;SSH OK - OpenSSH_5.3 (protocol 2.0)
[08:37:26] SERVICE ALERT: target-server;SSH;CRITICAL;HARD;3;Server answer:
[08:36:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;2;Server answer:
[08:35:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;1;Server answer:
[08:26:26] SERVICE ALERT: target-server;SSH;OK;HARD;3;SSH OK - OpenSSH_5.3 (protocol 2.0)
[08:05:26] SERVICE ALERT: target-server;SSH;CRITICAL;HARD;3;Server answer:
[08:04:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;2;Server answer:
[08:03:26] SERVICE ALERT: target-server;SSH;CRITICAL;SOFT;1;Server answer:
[08:00:26] SERVICE ALERT: target-server;SSH;OK;SOFT;3;SSH OK - OpenSSH_5.3 (protocol 2.0)
This went on and on...
A manual ssh connection confirmed the alert:
ssh target-server
ssh_exchange_identification: Connection closed by remote host
I ran a tcpdump on the target server and whenever the connection was closed, the output was the following:
tcpdump -n "src host my.own.ip.address and dst port 22"
09:02:05.192513 IP my.own.ip.address.38586 > target-server.ssh: Flags [S], seq 2270934900, win 14600, options [mss 1380,sackOK,TS val 516633443 ecr 0,nop,wscale 7], length 0
09:02:05.193403 IP my.own.ip.address.38586 > target-server.ssh: Flags [.], ack 1633368629, win 115, options [nop,nop,TS val 516633443 ecr 1699421739], length 0
09:02:05.194575 IP my.own.ip.address.38586 > target-server.ssh: Flags [F.], seq 0, ack 2, win 115, options [nop,nop,TS val 516633443 ecr 1699421740], length 0
Interesting is the last entry with the flag F (FIN). So the server definitely sent the closing signal.
To debug SSH, I activated verbose logging in /etc/ssh/sshd_config and restarted /etc/init.d/sshd. I then tailed /var/log/secure and to my big surprise the following entries appeared:
Failed password for root from 144.0.0.34 port 39899 ssh2
Failed password for root from 144.0.0.34 port 39534 ssh2
Failed password for root from 144.0.0.34 port 39439 ssh2
Failed password for root from 144.0.0.34 port 39329 ssh2
Disconnecting: Too many authentication failures for root
[...]
netstat confirmed that there were at least 36 opened ssh connections. So whenever all available tcp slots for ssh were occupied, the server of course immediately closed the connection.
Within four hours, this IP tried 42929 times to log in as root. A typical brute force attack it seems.
After blocking the IP the monitoring (and therefore having all necessary ssh tcp slots available again), checks were OK again.
No comments yet.
AWS Android Ansible Apache Apple Atlassian BSD Backup Bash Bluecoat CMS Chef Cloud Coding Consul Containers CouchDB DB DNS Database Databases Docker ELK Elasticsearch Filebeat FreeBSD Galera Git GlusterFS Grafana Graphics HAProxy HTML Hacks Hardware Icinga Influx Internet Java KVM Kibana Kodi Kubernetes LVM LXC Linux Logstash Mac Macintosh Mail MariaDB Minio MongoDB Monitoring Multimedia MySQL NFS Nagios Network Nginx OSSEC OTRS Office PGSQL PHP Perl Personal PostgreSQL Postgres PowerDNS Proxmox Proxy Python Rancher Rant Redis Roundcube SSL Samba Seafile Security Shell SmartOS Solaris Surveillance Systemd TLS Tomcat Ubuntu Unix VMWare VMware Varnish Virtualization Windows Wireless Wordpress Wyse ZFS Zoneminder