[noCensorship] Re: Few proxies on non
- From: Hat <hat@xxxxxxxxxxxxx>
- To: nocensorship@xxxxxxxxxxxxx
- Date: 11 May 2003 21:34:25 -0000
Hello wayne,
Saturday, May 10, 2003, 8:41:59 PM, you wrote:
[...]
>> >> 202.96.1.225 :8888 P 21.6/?
>> >> 206.129.0.18 :2896 P 9.0/?
>> >> 206.129.0.19 :9999 P 9.2/?
>> >> 208.230.117.43 :8487 P 9.3/?
>> >> 209.210.176.44 :8888 P 9.0/?
>> >> 209.67.28.104 :83 P 9.0/?
>> >> 210.131.177.253 :8000 P 9.0/?
>> >
>> >Your proxy 'latency' times still seem very high.
>> >Was this all the proxies you tested at the same time?
>> >Or did you remove all the failures?
>>
>> Yes, removed all failures.
w> Was the test set much larger than the result set you posted?
w> Just trying to see why these latencies are so high.
Not really, the posted result was about 85% of the original number of
proxies tested. Other proxies failed, so I removed them from the post.
>> >If it was a lot more, I can (maybe) understand this.
>> >Or is your proxy 'latency' time always this high from there?
>>
>> Well, ya.. I noticed that the latency time is always "high" here... does
>> that tell you something?
w> Latency time from proxyTools, or by some other measure as well?
If you mean some other scanners, then comparing to Accessdiver v4.113,
the latency of SPP is high. Here is an example:
SPP test result:
66.89.113.20 :8000 P 9.3/?
66.89.59.67 :8000 P 9.0/?
66.92.232.25 :8000 P 9.0/?
66.99.219.5 :8000 P 9.6/?
67.32.1.67 :8000 P 9.1/?
68.108.188.255 :22788 P 21.9/?
68.108.79.183 :1031 P 9.0/?
68.32.59.83 :6588 P 9.3/?
80.162.18.80 :6588 P 15.9/?
80.17.157.130 :8000 P 9.0/?
81.97.126.61 :6588 P 45.4/?
Accessdiver result for same proxies:
66.89.113.20:8000 Echo= 1623
66.89.59.67:8000 Echo= 2123
66.92.232.25:8000 Echo= 1452
66.99.219.5:8000 Echo= 2203
67.32.1.67:8000 Echo= 4487
67.32.1.67:8888 Echo= 2944
68.108.188.255:22788 Echo= 2724
68.108.79.183:1031 Echo= 1592
80.162.18.80:6588 Echo= 9223
80.17.157.130:8000 Echo= 1843
81.97.126.61:6588 Echo= 11386
w> I guess it's also very high if you are using SPP to test (say) just
w> one proxy?
w> Now that I've looked more carefully, I'm getting the same latencies
w> from the USA (even for the standard port proxies).
w> I didn't notice that before. I'm seeing at least one case where
w> the latency from the UAE is about 3 seconds, while the
w> latency from the USA is about 10 seconds. And that proxy is in
w> Seattle!
w> I need to investigate this further. I guess it's SPP
w> causing the problem, but I can't see how.
w> [...]
>> >Any comments about SPP? Fast, slow? Unreliable?
>>
>> So far it looks to be fast and reliable to me. Maybe not fancy looking, but
>> that fine!
w> True.
w> Damned if I'm gonna go thru that GUI stuff again!
w> LP was bad enough.
>> >What do your normal scanners do that's cool (that I'm not doing
>> >already in SPP)?
>>
>> ehem... wish list :)
w> Yep.
>> Well, how about:
>>
>> 1. Delete bad results and timeouts
w> You don't want to know about bad ones at all?
Not all the time!
w> I need those results to 'demote' ones in hosts.xml and (I and LP
w> users) need them for merging (updating) results to their config,
w> but I see your point for posts to lists.
Yes, that's the main reason.
w> ok. I added a '-p' ('pretty print') option.
w> The ctrl-c results will still show everything so far, but the
w> final results to the screen and file will not show anything that
w> failed on test 0. If you don't do test 0, this option is ignored.
w> I'll probably rethink that last bit, and make it display *any*
w> result line with a pass. :-)
Cool!
>> 2. Remove duplicates
w> Huh? I do this already. Doesn't it work for you? Examples please.
w> I've never seen duplicate IP addresses in the results.
I was just wondering :)
>> 3. Remove Proxy gateways
w> If an address works as a proxy, it's included.
w> I guess I don't understand what you mean here.
If an address works as a gateway to another address (proxy), I see a
security reason not to use that proxy. I prefer to delete it from the
"trusted" anon proxies list at least.
>> 4. Find and remove FBI & US Army proxies
w> Safe mode (default) should already be doing this.
w> Is it not catching some? Examples please.
w> When you use SP in default mode, it uses old Craig's 'safe' proxy
w> incantations to 'safe' your list of proxies. You must specify '-u'
w> to make it behave 'unsafely'.
w> Ahh ... it won't find them if they are specified as IP addresses
w> in the list to check. You mean you want reverse resolution (IP
w> to FQDN) done before the safe check? A bit of work, but no
w> problem (you still need to put up with flakey name servers so
w> there's no guarantee of 'safeness' though).
w> Confirm if that's what you want please.
Yes, reverse resolution and done before check, so as to save our ass
before they start chasing us a terrorist :)
>> 5. DNS Lookup in the same generated file result
w> All proxies tested are forward resolved for duplicate removal.
w> Do you want the reverse resolution printed as well?
w> I guess so.
That's what I meant.
w> The problem with that was that it needed to be in the comment area
w> (the proxies logically should be listed by IP address),
w> and it was often long, so it meant lots of line wraps for people.
w> I removed it a long time ago for that reason.
Having it as an option won't harm, I guess. Specially for those who
need to know who's address they are using!
w> Also, the dns servers in most places are slow, so I would have
w> needed the parallel dns code from sortProxy to do that in a
w> reasonable time. At the moment, I'm only using single-thread,
w> non-blocking DNS code.
w> I also didn't see why it was very useful (asuming the 'safe' mode
w> works, so you know where the proxy is).
w> So ... that's the argument against it; comments?
Above. I hate to use educational or ADSL proxies, but I enjoy using
.il proxies to suck their resources :)
>> I know that many of these can be done by sortProxy. Maybe merging sortProxy
>> code in SPP will do part of the above!
w> Only the DNS resolution, and I have other reasons for not liking that.
w> I'm not familiar with the user interfaces of other proxy scanners
w> though, so I'm happy to implement stuff as required. It's all fairly
w> easy with the new fully parallel test code. That was not the case
w> before (with SP, and almost all of the proxyTools).
w> --
w> wayne@xxxxxxxxxxxxx
w> http://proxytools.sourceforge.net/
--
Regards,
Hat
===8>============== noCensorship community ===============
List's webpage: //www.freelists.org/webpage/nocensorship
List's archive: //www.freelists.org/archives/nocensorship
To unsubscribe: nocensorship-request@xxxxxxxxxxxxx with 'unsubscribe' in the
SUBJECT field.
Moderator's email: nocensorship-moderators@xxxxxxxxxxxxx
===8>============== noCensorship community ===============
Other related posts:
- » [noCensorship] Re: Few proxies on non
