MySQL / phpMyadmin: "many connection errors" - mysql

My app no longer connects to its DB and phpmyadmin says #1129 - Host 'xxx.xx.xx.xx' is blocked because of many connection errors; unblock with 'mysqladmin flush-hosts'
mysql_error() prints out the same message. Could this be because of too many connections?
What causes this?

Open a command prompt (or shell in Linux) with administrative privilleges
Flush all hosts in MySQL using mysqladmin:
mysqladmin flush-hosts -u root -p
Open my.cnf (Linux) or my.ini (Windows) and change max_connect_errors variable to a large number. I used:
max_connect_errors= 1000000
Restart MySQL server

From the manual: http://dev.mysql.com/doc/refman/4.1/en/blocked-host.html
The number of interrupted connect
requests allowed is determined by the
value of the max_connect_errors system
variable. After max_connect_errors
failed requests, mysqld assumes that
something is wrong (for example, that
someone is trying to break in), and
blocks the host from further
connections until you execute a
mysqladmin flush-hosts command or
issue a FLUSH HOSTS statement. See
Section 5.1.3, “Server System
Variables”.
You could be having network problems connecting to your database.

We had this problem when using monit to monitor our DB server. Monit will cause connection errors if your config is of the form:
if failed port 3306 with timeout 15 seconds 2 times within 3 cycles then alert
Rather than
if failed port 3306 protocol mysql with timeout 15 seconds 2 times within 3 cycles then alert
The lack of 'protocol mysql' causes the connection error count to notch up.

Related

Error.MysqlData.MysqlClient.MysqlExecption

I am having the following problem since from a winform application I am trying to connect to my DATABASE and it gives me the following error
---------------- ErrorMysql.Data.MySqlClient.MySqlException host: my ip is blocked because of many connection error; unblock whith 'mysqladmin flus-hosts'--------
Please try the following steps:-
Verify the connection. Check the network connection to make sure that there’s no TCP/IP connectivity issue from your host
Increase the value of max_connect_errors. You should find the setting in the MySQL configuration file under the [mysqld] tag (my.ini on Windows, my.cnf on Unix/Linux) and edit the value of max_connect_errors.
For example:
[mysqld]
max_connect_errors=10000;
Or run this query:
SET GLOBAL max_connect_errors=10000;
Flush host cache. If you have shell access to the server, login and execute this command:
mysql -u root -p -e 'flush hosts'
From the SQL console, run this statement:
FLUSH HOSTS;
If all those methods didn’t work, try restarting the server. It should clear the host cache too.

How to simulate MySQL many connection errors

I recently had an issue with my production systems in which a MySQL server was blocking the application server, due to connection errors, and gave the following error:
Host 'xx.xx.xx.xx' is blocked because of many connection errors; unblock with 'mysqladmin flush-hosts'
I'd like to find a way to simulate this condition (or even trigger the actual condition on a non-production server), so I can respond to it properly. It's a pretty rare problem for my systems, but I'd still like to find a good way to respond to it.
You can avoid this problem by increase peer connections in mysql configurations
Logging to console/terminal with admin privileges
Flush all hosts using mysqladmin:
mysqladmin flush-hosts -u root -p
Open my.cnf (Linux) or my.ini (Windows) and change max_connect_error variable
max_connect_errors= 250000
Restart server with changes
To simulate error connection you can connect to MySQL server and then end connection before succesfull authentication. For example by netcat:
nc -i0.1 <hostname> 3306

How to "safely" terminate mysqld?

I am following mysql on Cygwin.
# To begin MySQL setup run the following:
mysql_install_db
# Run mysql - you'll get a firewall alert from windows if you have it active.
mysqld_safe &
# Immediately following that, it would be wise to run the following:
mysql_secure_installation
During the mysql_install_db, I was asked who the database (DB) would be for. I specified a non-administrator account nonadmin.
For the mysql_secure_installation process, I followed the similarly named MariaDB page. In contrast to the description, I was told that I had no root password, so I set it to the same as my cygwin and windows administrator account.
I lost track of whether which account was used for each of the three commands above, but here is what I remember:
mysql_install_db: Ran as nonadmin, but prompted for admin password
mysqld_safe: Run as nonadmin
mysql_secure_installation: Attempted as nonadmin, abandoned, and
completed as administrator account admin
It seems that I should restart the server after the final step. Using ps -ef, I noted that process mysqld belongs to nonadmin. Using nonadmin, I attempted the mysqladmin shutdown command at mysql on Cygwin. I get:
mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'nonadmin'#'localhost' (using password: NO)'
I tried the variation of the command that prompts the user for a password: mysqladmin -p shutdown. I did this twice, entering the cygwin/windows passwords for nonadmin and admin. Both times, I get:
mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'nonadmin'#'localhost' (using password: YES)'
This totalled 3 attempts. I then switched to the admin account and retried all 3 attempts. I got the above messages, but with the username admin:
mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'admin'#'localhost' (using password: NO)'
mysqladmin: connect to server at 'localhost' failed
error: 'Access denied for user 'admin'#'localhost' (using password: YES)'
So no joy connecting to the daemon to shut it down. As an alternative, I want to avoid clobbering mysqld with kill -9 because I like things to terminate cleanly (though I will probably do so soon).
As another alternative, I've come across references to /etc/init.d, but I have no such file.
Can anyone please explain why I can't connect to the daemon to shut it down?
Also, I'm new to MySQL, so if possible, it'd be nice to find a way back to the process at mysql on Cygwin rather than deviating from it too much.
Thanks.
What I ended up doing was kill -9 on all mysql looking processes shown by ps -ef. I then deleted /var/lib/mysql -- and accidentally deleted all of /var/lib/*. Subsequently, therefore, I reinstalled all of my Cygwin packages, thus starting from a clean slate.
I also found the following more detailed process at this alaya page for installing a MySQL/MariaDB database on Cygwin, after which I was able to shut down the MySQL server.
from the Administrator terminal run:
mysql_install_db
Start the mysql server:
mysqld_safe &
then run:
mysql_secure_installation
You should then be able to log in to MariaDB from a non-administrative shell with:
mysql -u root -p
The initial mysql_install_db step generates the following useful looking information, which I'm capturing here because I'm not sure if there is a way to get it again.
Installing MariaDB/MySQL system tables in '/var/lib/mysql' ...
2019-05-01 23:00:06 0 [Warning] InnoDB: innodb_open_files 300 should not be greaterthan the open_files_limit 256
OK
To start mysqld at boot time you have to copy
support-files/mysql.server to the right place for your system
PLEASE REMEMBER TO SET A PASSWORD FOR THE MariaDB root USER !
To do so, start the server, then issue the following commands:
'/usr/bin/mysqladmin' -u root password 'new-password'
'/usr/bin/mysqladmin' -u root -h fmaPC password 'new-password'
Alternatively you can run:
'/usr/bin/mysql_secure_installation'
which will also give you the option of removing the test
databases and anonymous user created by default. This is
strongly recommended for production servers.
See the MariaDB Knowledgebase at http://mariadb.com/kb or the
MySQL manual for more instructions.
You can start the MariaDB daemon with:
cd '/usr' ; /usr/bin/mysqld_safe --datadir='/var/lib/mysql'
You can test the MariaDB daemon with mysql-test-run.pl
cd '/usr/mysql-test' ; perl mysql-test-run.pl
Please report any problems at http://mariadb.org/jira
The latest information about MariaDB is available at http://mariadb.org/.
You can find additional information about the MySQL part at:
http://dev.mysql.com
Consider joining MariaDB's strong and vibrant community:
https://mariadb.org/get-involved/
During the mysql_secure_installation step, I followed the recommendation to set a root password. Instead of the mysql -u root -p step, then, I could instead issue the following to stop the MySQL server:
mysqladmin -u root -p shutdown
I made an important discovery made about this process from the point of view of a user without administrator rights. Contrary to what is suggested on the alaya page cited above, the above process does not have to be done from an administrator account. Importantly, the user accounts with which one can connect to the MySQL server has nothing to do with the user accounts on the host system (Windows or *nix) (a fact that I got from perusing MariaDB knowledgebase). I suspect that any user on the host system can connect to any user account on the server via the socket that it presents [1], with the right password.
I also found that the above process works on a machine that has very restrictive Windows firewall policies. Even if the server (the mysqld_safe & step) and client (the mysql -u root -p step) are started from the same non-administrative account, the Windows firewall will issue a warning about curtailing some features. I suspect that the trigger is the creation of a socket for connecting to the server. Despite the warning, however, ps -ef shows the server to be running. Therefore, I suspect that the firewall restriction consists of allowing only loop-back connections. The mysql_secure_installation step apparently disallows external connections, and I found that after this step, stopping and relaunching the server no longer triggers the firewall warning. So the effect of mysql_secure_installation seems to persist into subsequent sqld launches, at least when they are started via mysqld_safe.
[1] "Present" might not be the right terminology here, as I'm more of an analyst rather than a developer or administrator.

Website connection error related with MySQL 4.1.24

I get an error on my website (MySQL with TCP/IP connection)
Can't connect to local MySQL server
It sounds easy, perhaps, but not so easy to solve as would seem. At first i verified that mysqld process is running. Then connected to SSH server and started /etc/init.d/mysqld restart, but nothing has changed. Also i checked is the server running, using telnet your-host-name tcp-ip-port-number, just in case, but MySQL does not work on this port and shows another error: telnet: Unable to connect to remote host: Connection refused
The last i have tried is to run mysql with the skip-networking option, and it did not help as well as other attempts above. Any suggestions would be very appreciated.
This error normally means that MySQL is not running on the system or that you are using a wrong TCP/IP port number while trying to connect to the mysqld server.
Try to connect to the mysqld daemon on the local machine and check by mysqladmin variables, which TCP/IP port is configured to use mysqld (variable port).
Perhaps you are running MySQL-server with no corresponding privileges for the directory holding the socket file. In this case, either change the privilege for the directory or restart mysqld
Also this discussion might be interesting for you: http://community.office365.com/en-us/f/172/p/266451/815406.aspx
Unfortunately, i am not a pro and my advices can help not much, but that all what i found about your issue
This sounds like you are being blocked by the firewall on the server. You can disable the firewall for a quick test:
service iptables stop
service iptables start
This article will show you how to apply rules to the firewall to allow mysql access.
http://www.cyberciti.biz/tips/linux-iptables-18-allow-mysql-server-incoming-request.html
If you have SELinux, you have another set of problems which I can't help you with.
Sounds to me permission issue.
Does it run on the local machine?
If yes, then it might me possible it isn't allowed to listen any port(security purposes: mostly on a linux based server). Otherwise, Network user might not have permission granted to access mysql over internet.
As you say, if it's a website, I would never allow user access mysql directly but only the server shall: possible security breach otherwise.(you don't want your database to be dropped, do you?)
Well still, proper answer to you question resides here: http://dev.mysql.com/doc/refman/5.1/en/grant.html
and this might help:
Host 'xxx.xx.xxx.xxx' is not allowed to connect to this MySQL server
Firstly, you should verify that MySQL is actually running by checking your processlist. On linux you could do that like this, note that you should see both mysqld_safe and then mysqld as two separate processes.
sudo ps auxwww|grep -i mysqld
If it is not running, I would check the MySQL error log for clues as to why it is not starting.
If you then verify that it is running, we can check to see what ports or unix sockets it is listening on like so. If this doesn't work, get the process ID of mysqld (not mysqld_safe) and try search for that with grep instead of 'mysql'
sudo netstat -anp|grep -i mysql
You'll obviously want to restart without skip-networking to see a TCP socket appear.
Based on the output of that, you should see both a unix socket and a tcp socket. Also check the address in the fourth column for the TCP socket, it will likely say either 127.0.0.1:3306 or 0.0.0.0:3306. The former means that you can only connect via localhost (127.0.0.1) and the latter means the connection will work on any IP address.
If you've gone through all of that and are still not sure why it is working, you could post the processlist and netstat outputs for further review, along with the exact settings you are using to try and connect to MySQL and indicate whether you are connecting from the same server (i.e. locally) or from another server over the network.
See also:
http://dev.mysql.com/doc/refman/5.6/en/can-not-connect-to-server.html

Will a mysql restart flush the hosts cache (like FLUSH HOSTS)?

I'm getting "'host_name' is blocked because of many connection errors". I know the reason for the failed connections, and the cause is resolved.
To reset the condition, I assume that FLUSH HOSTS needs to be run at the remote host, but I don't have sufficient (RELOAD) privilege there.
If mysql is (ever) restarted on the remote host, will the hosts cache be cleared? (as if a FLUSH HOSTS had been run)
Otherwise, will the condition persist for some period of time? Or for ever?
I can not find documentation that the information in the host cache is cleared on MySQL restart.
I have been able to get an answer in practice with MySQL 5.5...
force a host to reach max_connect_errors (telnet hostname 3306 repeatedly until it fails)
verify MySQL client connections would fail with ERROR 1129 (HY000): Host '$hostname' is blocked because of many connection errors; unblock with 'mysqladmin flush-hosts'
restart MySQL
successfully connect with MySQL client after the restart.