![]() |
Re: More port blocking
Quote:
|
Re: More port blocking
I personally think the best way to aproach this problem, is for NTL to place a virus protector on their side, so that all downloads are virus free! That way if a virus is detected, it would get bashed on the head before the unsuspecting public get the hassles. Surely that would be far easier than blocking ports and having to try and educate some of the 'know it all's' there are around, who think they are not ever likely to be infected and refuse point blank to have any virus/firewall on their computer. A lot of computer suppliers are now making sure they are sold with it anyway.(if they are reputable that is.)
|
Re: More port blocking
Quote:
incog. :kiss: |
Re: More port blocking
I don't have a major problem with blocking the microsoft networking related ports (135, 137, 445 etc.) but question blocking ports above 1024. I don't think many home users run MS SQL, so why is this considered a risk. The other port mentioned in NTLs news item appears to relate to a port used by trojan software. Since trojans can be written to use any port, surely this approach will result in an increasing level of denial of service to customers as more ports are blocked.
Alan :( |
Re: More port blocking
I do have a problem with these ports being blocked, some games run on ports above 1024, does this mean that i won't be able to have a game of good ol Delta Force????
NTL dont seem to get it trough their heads :knock: §talk |
Re: More port blocking
That particular trojan (Sub seven) is quite old now and I think newer versions of it use different ports.
I see very few instances of that port turning up in my firewall logs now (on my non NTL servers). |
Re: More port blocking
Quote:
Exactly what reduction in service have you noticed ? |
Re: More port blocking
Quote:
But, for what it's worth, I will notice it when, due to NTL blocking ports above 1024, I get the occasional "page not found" error. And if I ever sign up for distance learning and expect to connect to the university systems using any of the methods now disbarred by NTL (note, this is still a legitimate residential use), I would sure as eggs is eggs notice it. Even NTL has acknowledged that some customers will be affected, fraz, you really don't have to try to fight the battle for them. |
Re: More port blocking
Quote:
The number of internet users now online that simply buy their PC and expect to use it in the same manner as they use a Microwave (or any other item of white goods) now far exceeds the number of enthusiasts using the net. ISPs (& I'm not singling out NTL) are changing the way that they provide a product to their customers and Blaster/Welchia gave all ISPs a short sharp shock. |
Re: More port blocking
Quote:
On the other hand blocking ports above 1024 is only effective as a short term solution as virus writers are likely to write future trojans to use ports that aren't blocked by major ISP. Hence this becomes an arms race, with more & more ports becoming blocked. I an wondering what happens when I buy software over the internet, which allows only 2 or 3 download attempts. Normal FTP could fail & if a second attempt also gets a blocked port, it is possible that it could be difficult to recover the situation. I am unsure how ports are allocated, but if it is based on cyclicly searching for the next free port starting from the last allocated port, then blocking port ranges could be a problem. Anyone know if this is actually likely to happen. Lastly, it occurs to me that it would be possible to write a trojan that hops ports on a time basis (just like frequency hopping on older secure radio links). Provided the time between hops is fairly long and the victim's PC's clock is set correctly, then the virus writer would have no difficulty determining the open port at any time. And what about the approaches taken by peer-to-peer networking software. I don't know how those work, but I have heard that port blocking is ineffective on those due to non-fixed port usage. So I say again, port blocking above 1024 is an unwinable approach and can at best be considered a temporary measure. To end on a more pleasant note, I would like to reiterate that blocking 135, 137, 139 & 445 seems a good idea to me. Well done NTL. Alan :banghead: |
Re: More port blocking
Quote:
|
Re: More port blocking
Quote:
I expect the real problem the sql & subseven ports this solve is the number of Mr Angry e-mails abuse get about PFW entries. That NTL are prepared to randomly break customer applications for that is appalling. If they really must block these ports why couldn't or wouldn't NTL use stateful filtering i.e. just stop inbound SYNs? Router performance? I hope so because at least that implies a limit to how much they can break. |
Re: More port blocking
Quote:
|
Re: More port blocking
Quote:
|
Re: More port blocking
Quote:
Originally you asked, 'have you even noticed?' Pem's answer to that is clearly 'yes'. You can't just go changing the question when your original point proves wobbly. |
| All times are GMT. The time now is 15:56. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2026, vBulletin Solutions Inc.
All Posts and Content are © Cable Forum