page content

What to do if my dedicated server is unreachable?

We are monitoring all physical servers running on our network via a Nagios monitoring infrastructure. If a server is down the system will send a notification to our support team. If the system is managed by all2all, we will try to restart the server and analyse the problem. In certain cases like hardware failure it might be necessary to send a support team to our datacenter and to physically assist the dysfunctionning boot process. We can then determine if new parts have to be installed or if a rescue boot has to be done. This is also the moment where we will start the emergency procedure as included in our hosting contrat, keep contact with you for all further decisions and finally make a report.

If you are just using the all2all facilities for colocation of your server hardware we can give you access to our remote switchable PDU. In most cases you can then immediately reboot your system without having to contact the all2all support team. Further there is also a console port management available which we can connect to your server's serial port and which enables you to log into your server even if you are for instance locked out by a wrong firewall rule.

Prior to taking as drastic mesures as remotely rebooting your system you should first determine if your server is still available and check if just a certain service is down or the system is just heavyly loaded. It can happen that the server is still running but the load average is that high that the server doesn't respond anymore to any service requests, also admin logins can then be impossible. In such cases one has either to wait until the system goes back to a normal state and responds again or hard reset the system. If this happens often, you should maybe evaluate if the system still fits your production environment (maybe the system is under-dimensioned?).

To monitor services running on your host you can use for instance the Webmin System and Server Status module which is included in the standard installation. This module can restart crashed services for you or send you a message if one of your services is down or if your server is in a critical state.

If you have traced the problem and fully determined that your server is down contact us asap when you cannot deal with the situation on your own. If you just have your server colocated at all2all we can serve you as remote hands. In certain cases you would need to send us the admin password of your server, please do this exclusively by PGP-protected mail. In difficult situations, colocated servers can also been handled out to you so that you can repair them and bring them back when they are repaired. You can find all our public PGP keys on our team page. Don't forget to send us the output of the traceroute you just made. You may as well call us or send a SMS on number +32 486 57 57 97 if you don't get a reaction by email.

You can also subscribe to our new Twitter acount on which we annonce any news regarding the all2all network.