Next: 5 Using the test-box
Up: Internet Delay Measurements using
Previous: 3 First installation
All email regarding the day-to-day operations of test-box should be sent to:
tt-ops@ripe.netThis is a mail-alias for the entire test-traffic group at the RIPE-NCC. Although we will read and respond to email sent to our personal mail-boxes, the response time to email sent to tt-ops@ripe.net will, in general, be much shorter.
We prefer email for all communication regarding the test-boxes. In exceptional cases, we can, however, provide you with a contact phone number or ``24/7'' contact address.
There are several security related issues involved in day to day management. These are discussed elsewhere [2]. We assume that the local contact has read and will follow (the spirit of) this document.
First of all, do not reboot the test-box unless you are asked to do so by the RIPE-NCC. The test-box can be rebooted by pressing the reset button. The reboot process will take a couple of minutes and can optionally be monitored by attaching a console monitor. The test-box is running again when the login prompt appears or when the box can be ping-ed from another machine.
If a test-box has to have its IP or gateway numbers changed, please provide us with the new numbers and the day that they become effective as soon as possible but at least a week in advance. We will then reconfigure the test-box remotely, shut it down and let you know when it can be moved. After the test-box has been moved, follow the procedures for first installation.
If a power outage is planned at your site, please notify us a couple of days in advance. We will then do an orderly shutdown of the test-box shortly before the power outage. After the power is switched back on, the test-box can be booted as usual.
In case of a disk related problem, we will ask the local contact to swap the disks. First switch off the machine. Use the second key to unlock the disks and carefully pull them out of their slots using the black handle. Insert the disks in the other slots and lock them again. Then switch on the machine again.
We will first try to reformat the second disk and start using it as a slave. If this does not work, we will ask the local contact to switch off the machine, unplug the second disk and send the entire unit back to the RIPE-NCC.
If we decide to upgrade the operating system, we will download and install the new kernel on the second hard-disk. After that, we will ask the local contact for a suitable time to swap the disks. At this time, the local contact has to switch off the machine, swap the disks as described in section 4.5, reboot the machine and optionally monitor the boot process. As soon as the machine is up again, we will check if the new kernel works as it should. If not, then the local contact will be asked to swap the disks back and reboot the machine again.
In the first instance, test-boxes with hardware problems not related to the disks have to be dismounted and sent back to the RIPE-NCC for repair. Unless the problem might be related to the GPS receiver, the antenna does not have to be dismounted.
As we get more experience with our test-boxes, we might provide instructions to the local contacts for doing small repairs.
Henk Uijterwaal