RDB Progress-o-matic

• Firewall transported from R.I. to West Comms;
• DMZ switch transported from R.I. to West Comms;
• West Green switch transported from R.I. to West Comms;
• East Green switch transported from R.I. to East Comms;
• Gigabit blades for Green switches transported from R.I. to West Comms;
• Power cables, keyboard, TFT screen transported from R.I. to West Comms;

• Email server transported from R.I. to West Comms;
• Web server transported from R.I. to West Comms;

• DNS/DHCP server transported from R.I. to East Comms;
• Patch cables, fibre patch cables taken from R.I. to East Comms;
• Firewall temporarily relocated to East Comms to be setup for testing, together with TFT screen, keyboard and appropriate cables; fibre interface configured, connected to fibre presentation panel in East Comms;
• Incoming Oxford Univ fibre patched to CEU East Comms in Ground Floor comms;
• Confirm that power socket underneath main rack in East Comms is not a standard three-pin 13A socket, but a larger 16A;
• Arrange for Gavin Litchfield (OUCS) to visit RDB on Tuesday 24 May at 10:00, to make the final router connection live;

• Acquired 16A plug from Wickes on Botley Road;

• Made East Green switch live (need to check NVRAM/boot errors);
• Made West Green switch live;
• Installed gigabit blades in each switch;
• Connected both switches to the fibre panel: still need to read docs on trunking;

• Successfully used USB-to-Serial converter to connect to the consoles on the switches from the new Sony laptop;
• Gavin Litchfield from OUCS came to connect up external network connection - presumably all OK, but the Firewall has the wrong NIC in it for the Oxford Univ incoming fibre, needs to be 100-Base-FX;
• Trunking between the West Green switch and the East Green switch is not yet working: trunking works when switches are connected directly to each other, but not across the main fibre run. Need to investigate fibre compatability and do further testing;

• On assumption that comms-to-comms trunking doesn't work, moved DMZ switch and email server to East Comms in an attempt to patch them into a hopefully-live external network connection once a new 100-Base-FX fibre NIC arrives;
• Fibre NIC arrives, but cannot be made to work. Not sure why this is;
• Again fail to make West Switch and East Switch communicate across the comms-to-comms fibre, so bring West Switch to East Comms and set up in parallel;
• Call Gavin Litchfield at OUCS and arrange to collect a fibre testing unit; collected and returned to RDB by taxi;
• By using fibre tester, it seems that external incoming fibre is live, all the way to our fibre panel; also, the new 100-Base-FX fibre is live at the end of its patch cable which gets plugged into the fibre panel, but as indicated above the interface doesn't go live;
• DMZ switch needs some configuration before being used by the Firewall and the email server for the DMZ network - it runs IOS rather than the more familiar CatOS, for which I have no documentation: problem is probably that it's trying to do layer 3 routing, rather than just basic switching;
• May approach OUCS to see if they can solve our external networking issue by supplying and fitting (say) a switch, which could then be used by one of our firewall's standard 10/100/1000-over-copper interfaces;

• A very efficient OUCS brought us a fibre-to-copper converter and we now have external connectivity ... Hurrah;
• Configured firewall sufficiently for routing/NAT to work for connections out from the GREEN network;
• Full testing of the inter-comms multi-mode fibre indicates that it is indeed dead; our options are to source some new GBICs and fibre cables to allow us to make use of the (undamaged) inter-comms single-mode fibre run that we were not intending to make use of at this stage. An alternative is to run a very long length of Cat5e from one comms room to the other, although this is far from ideal;
• 16A plug now attached to East comms cabinet, giving it full power;
• Email server appears to be doing odd things, will try a re-install tomorrow;

• Re-installed Debian Sarge on to the email server, sorted out appropriate firewall routing rules for it to be accessible to the world;
• Created accounts for all CEU staff and added the 'firstname.lastname' aliases;
• Email server is fully up and running!
• Since the multi-mode fibre run is confirmed as dead, have ordered GBICs and fibre leads to make use of the single-mode fibre; also, contractors are making sounds which suggest they will replace the multi-mode fibre: timescale on both these options are unclear, however; the alternative of running Cat5e, while probably workable, doesn't sound like the sort of solution which our Director would approve of and as such has been dismissed as an option for now

• My birthday - no work;

• Final clearing and tidying of 'old' office';
• Shutdown of development server Voyager;

• ...

• Email server not accessible from outside - turns out to be due to dead DMZ switch. Have put a cross-over cable between the email server and the firewall's DMZ interface for now;
• Telephones now alive in both comms rooms;
• East comms tidied to make space for arrival of existing racks of kit on Thursday;
• 1000-Base-LX GBICs arrived and installed in West Green switch and East Green switch: awaiting FC-SC fibre patch cables for them;
• West Green Switch returned to West Comms;

• FC-SC fibre patch cables arrived, also hundreds of Cat5e patch cables for main patch panels and switches;
• Finished packing of DE's office;
• Unracked and uncabled kit for returning to LIF;
• Labelled all equipment ready for removals tomorrow and Friday;

• All IT server-room equipment unracked, transported to the RDB and re-racked, although kept offline for now;
• Jo enployed to patch the telephone panel;
• Trunking across single-mode fibre between comms rooms established, finally! West Green switch and East Green switch communicating successfully;
• DMZ switch confirmed as 'probably dead' by Drew Hughes;
• West Green switch and East Green switch mounted in their respective rack cabinets;
• 16A plug attached to West comms cabinet's power-distribution strip;
• East Comms air conditioning leaking water;

• The main 'removals' day, lots of people about, very busy;
• Aircon in East comms was checked, appears to be working again and not dripping water;
• Most of the day spent patching the patch panels into the switches: total of about 400 ports took pretty much the whole day, with up to three people working on it at a time;
• IT Office has now got all its furniture and the fire safe in place, although will need some reorganisation - the packing crates with contents of shelves etc. are due tomorrow;
• Desktop PCs all in place in their respective offices;
• One electricity ring-main appears to be non-functioning, affecting two offices;
• All phones now installed and working;

• Final tidy-up and completion of both patch panels;
• General connectivity testing carried out, some questions over performance from West compared to East; this might be due to some flaky patch cables or Cat5e runs used to test West. This will need to be tested more thoroughly;

• First Proper Day;
• Set all non-trunk ports on both switches to be 'host' ports, thus improving performance and allowing Netware/IPX to work;
• Bring all Netware servers online, successfully tested with typical desktop PC;
• Old email server brought online, mail spools merged with those on new email server and IMAP folders copied across;
• Database server reconfigured and brought online;
• Debian proxy server, web cache reconfigured and brought online;
• Local web server reconfigured and brought online;
• Sysadmins' desktop PCs functioning normally;