BSG: Computer Networks

[Arrow]

Of course if the comm system isn't their only access point, then shutting it down is not going to help much.

True. The Cylons could attack other sensor systems to get in. But I also have to wonder - why are external signals recieved by the Galactic allowed to execute code? The Galactic doesn't have Baltar's program - there aren't Cylon-made any back doors. And DSP software should be able to tell what those Cylon signals were, and prevented their execution.

[Arrow]

going from my personal iterpretation of the episode... the computers were all linked via a wireless network(they had to be, or else how could the cylons hack it...*my logic at work*), afterall considering the computers are probably all over the ship it would be alot easier (albiet much more dangerous) to link them via a wireless. That was what the cylons were hacking... and how did geata break the network? oh yes. he unplugged the feed from the other computers, or in your words "HE UNPLUGED THE FUCKING ANTENNIA!!!!" (at least going from what i got from the episode...

When I saw that episode, I though that one cable was a breakout cable to the other computers in the network. The didn't say what the cable was.

[phongn]

going from my personal iterpretation of the episode... the computers were all linked via a wireless network(they had to be, or else how could the cylons hack it...*my logic at work*), afterall considering the computers are probably all over the ship it would be alot easier (albiet much more dangerous) to link them via a wireless. That was what the cylons were hacking... and how did geata break the network? oh yes. he unplugged the feed from the other computers, or in your words "HE UNPLUGED THE FUCKING ANTENNIA!!!!" (at least going from what i got from the episode...

Why would Galactica carry the equipment for a WLAN? She was deliberately designed with isolated systems and it appeared that Gaeta disconnected physical cabling and not antennae from the system.

[Winston Blake]

I've only seen the first season, but if the Colonials still use normal metal cables (rather than fibre-optic or whatever), then shouldn't networking all the computers by cable essentially make one huge antenna?

I mean, if the armour is capable of keeping out all interference that could cause any harm, yet Cylon jamming techniques have since become uber, there's still the possibility of a signal getting through that is just strong enough to induce some specific currents in the cable between linked computers (like some kind of inverse-TEMPEST).

If the Cylons know the layout of Galactica and all the protocols used, they might be able strike at many different points in the cable network and succeed in getting a worm to one computer, from where it can spread.

One problem with this is that as soon as several terminals started receiving unusual signals it should be possible to cut them off them automatically.

[Koolaidkirby]

When I saw that episode, I though that one cable was a breakout cable to the other computers in the network. The didn't say what the cable was.

I just figured the cables connected the terminal where gaeta was sitting, to the wireless recievers/antenna, as since we know the computers are decentralized/isolated, they would logically be positioned in different places on the ship(not all in the CIC) so i figured "simplest solution is the right one", they would probably use a wireless connection to string together all the computers from around the ship rather than running cables through all the hallways to link them

[Azazal]

Throwing my hat in:

As phongn said, Galactica was designed and built during the height of the Cylon civil war, she was meant to have her systems isolated so that compromising one system did not lead to the failure of all. So looks like the colonials had learned to protect theirs systems, and then over time forgot the lesson.

To put into modern perspective, you only need one breach of a firewall to cause havoc. Prime example I can think of off the top of my head, email attachment viruses. Say IT gets a bulletin about a new virus from their security software provider - McAfee, Symantec, whoever, and that a patch is on the way. So while IT sets up a mail filter and sends out a posting to everyone do not open a mail message titled "THE FUCK YOU VIRUS" or some such, do not open the attachment, delete if you see it in your mailbox. Unfortunately a few copies of the message got through before the bulletin was out there. And with all the postings and warning, there is always some ass that opens the file and bam, system infection spreads through the network. So IT needs to put in the patch to stop further infections, and then scrub the infected systems. Or is you want to see you IT really go ape shit, bring a laptop form home and plug into the LAN, unprotected, unauthorized system, could be loaded with all sorts of shit. So the Cylons only need to breach the defenses one, and the infect ship is then nuked before it can analyze and recover from the virus.

Anyway, back to Galactica, since the computer are decentralized, the Cylons could have decided to forgo trying their virus attack, then when detecting that the three mainframes were linked, decided to try the attack then. As for how they were able to transmit and log in, given the depth of infiltration that the Cylons were able to achieve, could there be a tap hidden somewhere in Galactica? Remember the device that was connected to the consol in the CIC, the one the Baltar discovered, the one that no one really noticed or paid attention to. What if there are other devices through out Galactica, hidden in the miles and miles of conduit. Could be a simple receiver that is in the engine area, or plugged into part of the com system. Something that just passively receives signals, low EM, hard to see on sensors.

And for Gaeta pulling the plugs, I though he was severing the cabling connecting the mainframes.

[Arrow]

I've only seen the first season, but if the Colonials still use normal metal cables (rather than fibre-optic or whatever), then shouldn't networking all the computers by cable essentially make one huge antenna?

I mean, if the armour is capable of keeping out all interference that could cause any harm, yet Cylon jamming techniques have since become uber, there's still the possibility of a signal getting through that is just strong enough to induce some specific currents in the cable between linked computers (like some kind of inverse-TEMPEST).

If the Cylons know the layout of Galactica and all the protocols used, they might be able strike at many different points in the cable network and succeed in getting a worm to one computer, from where it can spread.

One problem with this is that as soon as several terminals started receiving unusual signals it should be possible to cut them off them automatically.

How well a cable acts as an antenna depends alot the materials and construction of the cable, and the frequency you're trying to use. Coax cable can act as an antenna, such as if you have a long run of it in your attic.

And lets not forget that the Galactic was hit by a nuke and the computer systems didn't even flinch - the ship's hull is obviously capable of blocking an extremely wide frequency spectrum, so the kind of attack you propose should not be possible.

[phongn]

I just figured the cables connected the terminal where gaeta was sitting, to the wireless recievers/antenna, as since we know the computers are decentralized/isolated, they would logically be positioned in different places on the ship(not all in the CIC) so i figured "simplest solution is the right one", they would probably use a wireless connection to string together all the computers from around the ship rather than running cables through all the hallways to link them

But only the beefy computers need be near or in CIC. The rest are probably piddly machines designed for specific purposes and not really capable of heavy number crunching. A WLAN still flies in the face of Galactica's totally non-networked design.

[StarshipTitanic]

I was watching some scenes in the miniseries and the tour guide Cylon, while he was touring, mentioned that computers were kept primitive because Cylons were also excellent hackers.

[Zac Naloen]

I haven't managed to read through the whole thread but apologies if this has been addressed, but in regards to tempest attacks.


One reason why they may only be able to control a system if networked is due to the overwhelming amount of individual "chatter" that must be coming out of seperately networked computeres, but wire them together and suddenly things begin to become a whole lot clearer to and outside observer.

[DocHorror]

Its possible computers on old Battlestars were isolated in case of a Cylon boarding action. The Cylons wouldn't be able to take over the ships systems from one terminal.

Remember the colonial defenses went down when the Cylons sent a signal that activated Baltars backdoors & shut off the Colonial systems.

I'm not sure but I don't remember a virus being mentioned in the mini.

[Ironwolf]

Arrow I believe you said something about just unplugging the antenna or hooking up a swtich? I have a few comments on that.

1) Lets think about all the antennas that could be on a Battlestar:

Voice Comm for CIC figure on the low stide of two each upper and lower, UHV, VHF, HF and thats only 1 radio each.

Thats 6,

Now add 1 radio each for backup,

Now at 12,

LSO (thats Landing Signal Officer) Comm on 1 flight pod has to have at least 2 UHF and 2 VHF, figuring 1 antenna each,

Thats 16,

See the pattern? Now add antennas for Glide Slope, Radar, Figure they have to have some sort of homing beacon or TACAN, some type of IFF system, some type of naviagtion system (can't all be done by the stars considering I haven't seen anyone with a sectant), most likely some kind of ECM system, all of which need antennas. Now add to the fact that this ship probably has a lot of redundant systems. That equals a whole lot of antennas.

Now, simpley putting a switch on the antenna isn't a good idea for a few reasons, you introduce a source of noise, a lot more wiring (we are talking a few pounds at least if the antenna isn't near the switch), plus you have now added something else to break and/or possible voltage to a system not equipted to handle a whole lot.

Simply unplugging the antenna or turning it off with a switch is unrealistic in a C130 let alone something the size of a Battlestar.

Just my oppinion.

[Arrow]

Arrow I believe you said something about just unplugging the antenna or hooking up a swtich? I have a few comments on that.

1) Lets think about all the antennas that could be on a Battlestar:

Voice Comm for CIC figure on the low stide of two each upper and lower, UHV, VHF, HF and thats only 1 radio each.

Thats 6,

Now add 1 radio each for backup,

Now at 12,

LSO (thats Landing Signal Officer) Comm on 1 flight pod has to have at least 2 UHF and 2 VHF, figuring 1 antenna each,

Thats 16,

See the pattern? Now add antennas for Glide Slope, Radar, Figure they have to have some sort of homing beacon or TACAN, some type of IFF system, some type of naviagtion system (can't all be done by the stars considering I haven't seen anyone with a sectant), most likely some kind of ECM system, all of which need antennas. Now add to the fact that this ship probably has a lot of redundant systems. That equals a whole lot of antennas.

Now, simpley putting a switch on the antenna isn't a good idea for a few reasons, you introduce a source of noise, a lot more wiring (we are talking a few pounds at least if the antenna isn't near the switch), plus you have now added something else to break and/or possible voltage to a system not equipted to handle a whole lot.

Simply unplugging the antenna or turning it off with a switch is unrealistic in a C130 let alone something the size of a Battlestar.

Just my oppinion.

You do know they make low noise switches? And that modern RF switches are rated for millions of switching operations?

The switch can be mounted in the receiver the antenna plugs into, or into an RF Distrobution box. The can be electrically controlled, requiring only a single wire to turn them on or off. Right now one of my coworkers has a four antenna switch that is two inches in a diameter and about half an inch high; this one is used to route signals to a single receiver. A single on/off antenna switch is can be much smaller. Since the Cylons like to hack into electronic sensors (and this includes comm systems), you would think these would be designed into those systems. The switches can be placed into devices (such as the receiver, signal processor, RFD or amplifier boxes) where power and control signals are readily available.

Using switches is hardly an unrealisitc option.

[Ironwolf]

You do know they make low noise switches? And that modern RF switches are rated for millions of switching operations?

Yes I am aware of low noise switches. And I am also aware of the ratings of those switches and the real world application as well.

The switch can be mounted in the receiver the antenna plugs into, or into an RF Distrobution box. The can be electrically controlled, requiring only a single wire to turn them on or off.

Now multiple that "one wire" by how many different systems that these switches will be connected to. It adds up.

Right now one of my coworkers has a four antenna switch that is two inches in a diameter and about half an inch high; this one is used to route signals to a single receiver.

I work with these a lot and if I remember right it's not an on/off switch, it's an (in my case) upper/lower switch. You can't shut it off.

Since the Cylons like to hack into electronic sensors (and this includes comm systems), you would think these would be designed into those systems. The switches can be placed into devices (such as the receiver, signal processor, RFD or amplifier boxes) where power and control signals are readily available.

Look at the number of systems your talking about disconnecting. If comm gets hacks (or attempted) yes, it would be easy to tell, however what if a digital systems gets hacked? How will you know unless it's to late. A network could tell you it's been compromized but how about other systems? IFF? If it's a system that is supposed to recieve a signal, how can it tell untill it recieves and analizes that signal? By then it's to late.

Can you cut off individual antennas/systems? Yes. But you would either be flipping switches for a while or hook them all up to one switch and blind yourself to the outside world. All of the displays on the CIC that have radar contacts show an object with information from a lot of different systems. IFF, Radar, Radio comms, RF and EM emissions to name a few. All of these things require different systems to recieve and process signals.

Again, in my oppinion, it's too much to loose. Whats the use of a CIC that can't see or hear anything?

[Battlehymn Republic]

I read the following in a fanfic author's response to a review of his very excellent story, Lost Kin.

I will concede that I’m probably giving the Cylons an almost magic ability to get into computer systems. But I have a couple of defenses.

1 – They are code. In other words they can adapt to even unknown systems in a way no one else can. Some of this will be touched on in later chapters and will comprise a key part of the sequel.

2 – In my opinion the show is even more magical. Gaeta hooks up three standalone computers in the second episode of season two and suddenly the Cylons can infiltrate them. They were hardwire connections. Basically he took three desk top machines and hooked them up to make a LAN. Where was the external connection that allowed the Cylons access to the network? For that matter the Cylons were able to plant a virus in that network that affected systems all over the Galactica. How? The big deal about the Galactica is that NONE of its systems are connected. So the only systems that should have been affected by the Cylon virus were the three that were hooked up. And yet we saw power systems, communications and environmental controls go haywire.

And let’s not forget the specific point made in the Pilot movie that the airlocks on the Galactica were manual – not computer controlled. But in the show the environmental systems (not one of the computers plugged into Gaeta’s LAN) were able to AUTOMATICALLY close and seal the blast doors and AUTOMATICALLY depressurize the compartment.

Its pretty close to magic to be able to hack into a system that has no external links and affect systems that are designed to be operated manually with no automatic or computer controls. Think of it as the diagnostic computer in your car suddenly being able to turn the steering wheel and apply the parking brake.

Are these fair criticisms?

[Xon]

Are these fair criticisms?

Feel the fucking wank.

[Arrow]

Right now one of my coworkers has a four antenna switch that is two inches in a diameter and about half an inch high; this one is used to route signals to a single receiver.

I work with these a lot and if I remember right it's not an on/off switch, it's an (in my case) upper/lower switch. You can't shut it off.

No, feed the thing all zeros, and its off. Its one of the features we like about the switch, because it allows to make sure the attentuators are set properly before allowing a signal through.

Look at the number of systems your talking about disconnecting. If comm gets hacks (or attempted) yes, it would be easy to tell, however what if a digital systems gets hacked? How will you know unless it's to late. A network could tell you it's been compromized but how about other systems? IFF? If it's a system that is supposed to recieve a signal, how can it tell untill it recieves and analizes that signal? By then it's to late.

Can you cut off individual antennas/systems? Yes. But you would either be flipping switches for a while or hook them all up to one switch and blind yourself to the outside world. All of the displays on the CIC that have radar contacts show an object with information from a lot of different systems. IFF, Radar, Radio comms, RF and EM emissions to name a few. All of these things require different systems to recieve and process signals.

Again, in my oppinion, it's too much to loose. Whats the use of a CIC that can't see or hear anything?

And whats the point of allowing your systems to be used by the enemy? If they're comprised, you can't use them - they can. The cut-offs should be one of the most important aspects of the design, since it would be the last ditch safety feature. They crew is obvious trained to detect Cylon intrusions (notice how Gaeta did when he set up the network), and they would have a procedures to follow for when its time to cut the signal feeds. As the for the number of wires - its a ship that's 4640'x1820'x580' (as listed on Scifi's site), you have the room for them. This isn't a C130, were space is at an absolute premium.

[Arrow]

Right now one of my coworkers has a four antenna switch that is two inches in a diameter and about half an inch high; this one is used to route signals to a single receiver.

I work with these a lot and if I remember right it's not an on/off switch, it's an (in my case) upper/lower switch. You can't shut it off.

No, feed the thing all zeros, and its off. Its one of the features we like about the switch, because it allows to make sure the attentuators are set properly before allowing a signal through.

BTW, the switch I'm talking about is a General Microwave Series 91 (F9140 should be the correct part number).

[Arrow]

Ok, here's a thought that's be nagging at me: Why don't the Cylons hack the Galactic's Comm and Sensor sensors everytime they engage each other? Sure, they're not networked, but the Cylons could cause all sorts of havoc by hacking those systems alone, preventing coordination with the Vipers and civilian fleet, and generating false targets. Why wait for the systems to be networked? How can Cylons detect when the computers are networked together before starting their attack?

[FOG3]

Admiteddly I haven't seen the series due to not having cable, but the initial attack used backdoors build in to the Command Navigation Program. Additionally we know they've planted devices like those on the Galactica Six pointed out.

If Baltar developed the new program, one would think he might have access to the old program, which one would think would be run be the biggest computers. What else would you need the computational power for? Surely if he even remotely had access Six would have sent it to her superiors. Plus they've undoubtably been doing other things in infiltration, which may have won them knowledge of older code. Perhaps under a premise of historical purposes, as newer ships aren't using them. Newer code possibly being modified by them in development in hard to pick up ways they could use.

Perhaps they've managed to plant enough devices to access, but nowhere near all, some of the Galactica's computers and can use what knowledge they've won from infiltrating the Colonials to try to access. As long as they're not hooked up, their devices can only mess with one system, and it's backups can compensate and then restore it. If they're networked using their knowledge of the Colonial's systems and planted devices they can seize a lot more. If there's no backup system to compensate after that you just lost an entire system and whatever it's connected to.

[Turin]

This is a bit of an out-there idea, but bear with me. Maybe the Cylon virus already existed within the Galactica's computers (because of infiltrators/Baltar), but just doesn't work unless it's in a networked environment? Maybe the code is deliberately written in such a way that its components are seperated into different subsystem computers.

Why would the cylons do such a silly thing? Because it would make the viral code harder to detect using whatever rudimentary virus scanning software the Colonial Fleet was using prior to the attack. This way all the code is installed and in place, just awaiting a "master command" that is transmitted by the Cylons. This causes the controlling piece of code (probably installed in the communications subsystems) to "search out" and connect to the other pieces of code and then become a sort of gestalt virus.

So in this case, when Gaeta connected the computers into a network, that allowed the pieces of viral code (which maybe have been actively searching for their mates since the original attack) to link up and become the active virus.

Probably some big gaping flaws in that theory, but it explain some of it, at least. It still doesn't work at explaining why non-computerized systems can be affected by the virus (like the airlock on the firing range), but I'm willing to waive SoD in that single case because I don't think it was well thought-through by the writers.

[Arrow]

Both of those are fairly plausible explainations, and high-light just how bad the Colonials are at computer security. But I think the biggest reason for the Cylon's super hacker abilities is general ignorance on the writers part. I'm not bashing them for; they're only end users, not IT techs, IT managers, or software/hardware developers. But in reality, a computer network can be locked down so tight that attempting to hack into it just isn't worth the effort in a combat situation.

I would had perferred if Six had used Baltar to get information on fleet deployments (under the guise that he needed real world information to test his navigation program). Then, when the war started, the Cylons could have sabotaged command and control bases, while the Cylon fleet, knowing precisely where the Colonial battlegroups are located, jump in and destroy the Colonial fleet in a surprise attack. But that's just me.

[Uraniun235]

I think what this really stems from is Ron Moore's desperate desire for Galactica to be a more "down to Earth" starship with handsets and manual controls. He wants a way to make everyone think that it makes sense for it not to look like bleeding-edge technology and so he makes the Cylons these super-hackers against whom bleeding-edge technology becomes a liability, not an asset.