Discussion:
ARP activity
(too old to reply)
Dean
2005-02-16 03:25:04 UTC
Permalink
I've been having some issues with my quality of service. For example,
lost packets and erratic round trip times to the upstream router. I
fired up a packet sniffer and found what looked to me like a lot of ARP
traffic coming through my modem. Also, some of the addresses involved
surprised me (for example, 10.*.*.*).

I'm seeing 650 ARP requests in a minute. Is this reasonable?

I can send my packet capture to someone, if desired.

dpj
Murray Watson
2005-02-16 08:09:48 UTC
Permalink
In adelphia.security-issues - article <pNGdnRdTQsyMII_fRVn-
Post by Dean
I've been having some issues with my quality of service. For example,
lost packets and erratic round trip times to the upstream router. I
fired up a packet sniffer and found what looked to me like a lot of ARP
traffic coming through my modem. Also, some of the addresses involved
surprised me (for example, 10.*.*.*).
I'm seeing 650 ARP requests in a minute. Is this reasonable?
I can send my packet capture to someone, if desired.
dpj
ARPs are the mechanism by which IP addresses are resolved to ethernet
MAC addresses, they are usually sent in broadcast mode by routers
that feed a network collision segment, It is a necessary product
of IP/ethernet infrastructure.

In this case CableModem "headends", large quantities are most likely
the result of MS Messenger** spammers sending UDP packets to all
known IP addresses in a netblock. If the router/headend doesn't have
the IP/MAC address in it's table, it sends an ARP, "who has IP
www.xxx.yyy.zzz, respond to 10.x.y.z with your MAC address". The
more sparsely populated your particular segment is, the more ARPs
you'll see.

The MAC address is the ethernet address assigned to each and every
ethernet device, the first 6 bytes are manufacturer oriented, the
second set of 6 are unique numbers to that manufacturer. Microsoft
even makes up some for it's RAS devices.

Spammers typically use UDP ports 1026, 1027, they use UDP because
it's a connectionless protocol, MS Net Messenger is not required to
respond so the originating IP can be spoofed with no consequence and
no way to track the origin. Blocking at border routers wouldn't be
noticed by many if any people on the 'Net these days other than the
ones with Win PCs directly connected to the modem where MS Net
Messenger service has not been turned off. MS Net Messenger was
intended for use on LANs.

http://support.microsoft.com/default.aspx?scid=KB;EN-US;Q330904

** MS Messenger is not to be confused by with the several other MS
"Messenger" products.

http://www.mynetwatchman.com/kb/news/2004-msgspamfrag.htm

There are several older discussions about ARPs in adelphia.powerlink.
They should still be "on the spool" if you'll go back a little way.

Or you could google for more info.
http://www.google.com/search?&q=%22large+quantity%22+ARP+packets
Dean
2005-02-16 12:26:22 UTC
Permalink
Hey, thanks for the response.

Now, I'm in a 69.166.x.x subnetwork. I know that 10.x.x.x addresses are
"private" addresses - anybody can use them, they aren't administered
through the global Internet administration, they really shouldn't be
leaving a local LAN.

Has someone just mis-configured a router to let this stuff leak out into
the cable network?

Dean
Post by Murray Watson
In adelphia.security-issues - article <pNGdnRdTQsyMII_fRVn-
Post by Dean
I've been having some issues with my quality of service. For example,
lost packets and erratic round trip times to the upstream router. I
fired up a packet sniffer and found what looked to me like a lot of ARP
traffic coming through my modem. Also, some of the addresses involved
surprised me (for example, 10.*.*.*).
I'm seeing 650 ARP requests in a minute. Is this reasonable?
I can send my packet capture to someone, if desired.
dpj
ARPs are the mechanism by which IP addresses are resolved to ethernet
MAC addresses, they are usually sent in broadcast mode by routers
that feed a network collision segment, It is a necessary product
of IP/ethernet infrastructure.
In this case CableModem "headends", large quantities are most likely
the result of MS Messenger** spammers sending UDP packets to all
known IP addresses in a netblock. If the router/headend doesn't have
the IP/MAC address in it's table, it sends an ARP, "who has IP
www.xxx.yyy.zzz, respond to 10.x.y.z with your MAC address". The
more sparsely populated your particular segment is, the more ARPs
you'll see.
The MAC address is the ethernet address assigned to each and every
ethernet device, the first 6 bytes are manufacturer oriented, the
second set of 6 are unique numbers to that manufacturer. Microsoft
even makes up some for it's RAS devices.
Spammers typically use UDP ports 1026, 1027, they use UDP because
it's a connectionless protocol, MS Net Messenger is not required to
respond so the originating IP can be spoofed with no consequence and
no way to track the origin. Blocking at border routers wouldn't be
noticed by many if any people on the 'Net these days other than the
ones with Win PCs directly connected to the modem where MS Net
Messenger service has not been turned off. MS Net Messenger was
intended for use on LANs.
http://support.microsoft.com/default.aspx?scid=KB;EN-US;Q330904
** MS Messenger is not to be confused by with the several other MS
"Messenger" products.
http://www.mynetwatchman.com/kb/news/2004-msgspamfrag.htm
There are several older discussions about ARPs in adelphia.powerlink.
They should still be "on the spool" if you'll go back a little way.
Or you could google for more info.
http://www.google.com/search?&q=%22large+quantity%22+ARP+packets
Murray Watson
2005-02-16 16:09:52 UTC
Permalink
In adelphia.security-issues - article <1oSdnX-ewehzpo7fRVn-
Post by Dean
Hey, thanks for the response.
Now, I'm in a 69.166.x.x subnetwork. I know that 10.x.x.x addresses are
"private" addresses - anybody can use them, they aren't administered
through the global Internet administration, they really shouldn't be
leaving a local LAN.
Has someone just mis-configured a router to let this stuff leak out into
the cable network?
Dean
Post by Murray Watson
In adelphia.security-issues - article <pNGdnRdTQsyMII_fRVn-
Post by Dean
I've been having some issues with my quality of service. For example,
lost packets and erratic round trip times to the upstream router. I
fired up a packet sniffer and found what looked to me like a lot of ARP
traffic coming through my modem. Also, some of the addresses involved
surprised me (for example, 10.*.*.*).
I'm seeing 650 ARP requests in a minute. Is this reasonable?
I can send my packet capture to someone, if desired.
dpj
The RFC1918 addresses are to be used internally and should not be
routed across the Internet, ie. passing through an ISP's border
routers. What you are experiencing is internal to Adelphia's
network, their "headends" operate with 10/8 addresses and your modem
is assigned a 10.x.x.x address as well. It's all quite normal.

What you are experiencing is quite normal (thank you spammers). If
you're not seeing any other protocols other than TCP/UDP/ICMP, that
would be the issue. Another scenario would be someone with a
computer/router running a routing protocol and has their gateway
device using a static IP in the 10/8 range and is configured to use a
route discovery protocol, but you would be seeing other protocols,
ARP is a lower level ethernet mechanism and is really only used by
devices on a local collision segment. You could have a neighbor
using a router in a non-bridge mode, attempting to discover and route
or some nearby screwball doing port scans.

Do you see any pattern in the IP's being ARP'd? If there is a
sequential pattern, it's most likely someone locally doing port
scanning on their own netblock but you'd probably also see some other
TCP/UDP probes directed at your IP, some find some amusement in that.
Spammers could be sequential but their software most likely does it
randomly to avoid IDS discovery.

I used to see a constant flow of traffic but when I was moved to my
current IP 69.164.1.153, that all ceased. This /18 netblock
"NetName: 69-164-0-0-Z2" wasn't registered until 04/2004 and the
encompasing /12 netblock "NetName: ADELPHIA-CABLE-7 netblock wasn't
registered until 08/2003 so it might not be on the spamware's CD as
being worthy of spamming.
Fred Johnson
2005-03-07 03:19:37 UTC
Permalink
----- Original Message -----
From: "Murray Watson" <***@adelphia.net>
Newsgroups: adelphia.security-issues
Sent: Wednesday, February 16, 2005 11:09 AM
Subject: Re: ARP activity
<Snipped>
Post by Murray Watson
Do you see any pattern in the IP's being ARP'd? If there is a
sequential pattern, it's most likely someone locally doing port
scanning on their own netblock but you'd probably also see some other
TCP/UDP probes directed at your IP, some find some amusement in that.
Spammers could be sequential but their software most likely does it
randomly to avoid IDS discovery.
<Snipped>

We're seeing increased arp traffic also, inside home and school lans. The
flooding looks to be a form of DoS, and originates from computers infected
with the latest SDBot worm. The amount of arp broadcasts will overload a
router in a few minutes requiring a reset.

Thoughts?
Murray Watson
2005-03-07 20:24:06 UTC
Permalink
In adelphia.security-issues - article <b_6dnbjJGYrwWrbfRVn-
***@adelphia.com>, on Sun, 6 Mar 2005 22:19:37 -0500, Fred Johnson
says...
Post by Fred Johnson
----- Original Message -----
Newsgroups: adelphia.security-issues
Sent: Wednesday, February 16, 2005 11:09 AM
Subject: Re: ARP activity
<Snipped>
Post by Murray Watson
Do you see any pattern in the IP's being ARP'd? If there is a
sequential pattern, it's most likely someone locally doing port
scanning on their own netblock but you'd probably also see some other
TCP/UDP probes directed at your IP, some find some amusement in that.
Spammers could be sequential but their software most likely does it
randomly to avoid IDS discovery.
<Snipped>
We're seeing increased arp traffic also, inside home and school lans. The
flooding looks to be a form of DoS, and originates from computers infected
with the latest SDBot worm. The amount of arp broadcasts will overload a
router in a few minutes requiring a reset.
Thoughts?
I'd try to differentiate DOS-purposed activity from DOS-like activity
of poorly written/designed, thoughtless viruses. While the activity
is similar/same, motive is different.

1) Network/Router based IDS/IPS.

2) More widely implemented virus protection.

3) More secure OS.

4) Better egress filtering.

5) Carefully thought out inbound port filtering. I'm not sure why
UDP/1026-1029 is open at Adelphia's borders, and it may not be, the
activity seen may be originating within Adelphia's network. This
should be prevented by egress filtering of invalid source IPs at the
headend. This leads to a question of what level of egress filtering
is performed by providers to prevent their networks from being
sources of abuse.

Something changed locally here over the weekend. There was an outage
for about 12hrs. Where I used to see little uninitiated modem
traffic, I now see a constant stream. I have the same IP but power
levels and SN/Dbs are different, latency is up but throughput is up.
They may have changed some head-end equipment.

I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Obfus Kataa
2005-03-07 21:22:35 UTC
Permalink
Post by Murray Watson
I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Murray,

I have been seeing TCP 1433 and pptp access attempts to my new Toshiba TiVo
(which my router drops). They have been going for five days at a low rate (a
burst, then quiet for hours), and they are all from the same UUNET address
that is not registered to anyone (other than UUNET.

I mention this because I remembered you also had a Toshiba DVR (HD40?).
--
oK+++
In the first place, God created idiots. That was for practice. Then he
made school boards.
-Mark Twain
16:18 up 20 days, 22:55, 1 user, load averages: 0.05 0.08 0.08
Murray Watson
2005-03-08 07:17:11 UTC
Permalink
In adelphia.security-issues - article
<***@rkoj.nz>, on Mon, 7 Mar 2005
16:22:35 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Murray,
I have been seeing TCP 1433 and pptp access attempts to my new Toshiba TiVo
(which my router drops). They have been going for five days at a low rate (a
burst, then quiet for hours), and they are all from the same UUNET address
that is not registered to anyone (other than UUNET.
Attempts to your Toshiba, is it in a DMZ or do you have multiple
Adelphia IPs? Connections to the TIVO service are TCP/80 based and
should pass straight through any NATed router. Mine goes through a
DLink wireless USB network adapter to a Linksys wireless access point
(WAP54G) through a Linksys router(BEFSR41).

Set up as a NATed DHCP client but I use DHCP from an NT server on my
internal network instead of the router so my internal network gets
WINS support, I also gave it a MAC reservation so it's IP remains
constant and it has a name.
Post by Obfus Kataa
I mention this because I remembered you also had a Toshiba DVR (HD40?).
HD400, 80 minutes at basic quality. Did you get the new one? IIRC
120 minutes w/recorder?
Obfus Kataa
2005-03-08 22:04:45 UTC
Permalink
Post by Murray Watson
In adelphia.security-issues - article
16:22:35 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Murray,
I have been seeing TCP 1433 and pptp access attempts to my new Toshiba TiVo
(which my router drops). They have been going for five days at a low rate (a
burst, then quiet for hours), and they are all from the same UUNET address
that is not registered to anyone (other than UUNET.
Attempts to your Toshiba, is it in a DMZ or do you have multiple
Adelphia IPs? Connections to the TIVO service are TCP/80 based and
should pass straight through any NATed router. Mine goes through a
DLink wireless USB network adapter to a Linksys wireless access point
(WAP54G) through a Linksys router(BEFSR41).
It is NATed. The destination shows up as 192.168.0.12 (my Toshiba). My
Netgear drops it and logs it. Yes the SRC port is 080 on 204.176.49.2
I don't know which protocol is being used because the filter rule I have for
the two ports that have been being tickled denies access "from any to any
for any protocol." It's a rule late in the filter process after any local
addresses (i.e. on the LAN) would have been forwarded if they were trying to
speak to each other. Since the Netgear separates filters for LAN->LAN,
WAN->LAN, LAN->WAN, I can tell this is coming from outside trying to get in
to the Toshiba. I thought it might be a response to a request from my TiVo
to TiVo central for update, but the address does not seen to be one of
theirs, and updates seem to be coming through fine. TiVo doesn't mention
those ports as being used by them. Their block is both in the lower < 1024
and upper range (> 8000).
Post by Murray Watson
Set up as a NATed DHCP client but I use DHCP from an NT server on my
internal network instead of the router so my internal network gets
WINS support, I also gave it a MAC reservation so it's IP remains
constant and it has a name.
Post by Obfus Kataa
I mention this because I remembered you also had a Toshiba DVR (HD40?).
HD400, 80 minutes at basic quality. Did you get the new one? IIRC
120 minutes w/recorder?
Yes I got the TR 20 with 120GB drive and a DVD recorder. So far it has been
good. I record on highest quality, and the signal through the Toshiba is
better than the one I used to get when the Sony VCR was between the box and
the TV. It may be just my imagination, but I think the quality is a bit
better when I use the RF signal input on the DVR rather than using the Cable
box RCA outputs. Wish the Motorola's serial port was active. But going way
off topic for this group, here.
--
oK+++
16:19 up 21 days, 22:56, 1 user, load averages: 0.48 0.50 0.58
Murray Watson
2005-03-09 09:29:06 UTC
Permalink
In adelphia.security-issues - article
<***@surivxeljes.gdh>, on Tue, 8 Mar
2005 17:04:45 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
In adelphia.security-issues - article
16:22:35 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Murray,
I have been seeing TCP 1433 and pptp access attempts to my new Toshiba TiVo
(which my router drops). They have been going for five days at a low rate (a
burst, then quiet for hours), and they are all from the same UUNET address
that is not registered to anyone (other than UUNET.
Attempts to your Toshiba, is it in a DMZ or do you have multiple
Adelphia IPs? Connections to the TIVO service are TCP/80 based and
should pass straight through any NATed router. Mine goes through a
DLink wireless USB network adapter to a Linksys wireless access point
(WAP54G) through a Linksys router(BEFSR41).
It is NATed. The destination shows up as 192.168.0.12 (my Toshiba). My
Netgear drops it and logs it. Yes the SRC port is 080 on 204.176.49.2
I don't know which protocol is being used because the filter rule I have for
the two ports that have been being tickled denies access "from any to any
for any protocol." It's a rule late in the filter process after any local
addresses (i.e. on the LAN) would have been forwarded if they were trying to
speak to each other. Since the Netgear separates filters for LAN->LAN,
WAN->LAN, LAN->WAN, I can tell this is coming from outside trying to get in
to the Toshiba. I thought it might be a response to a request from my TiVo
to TiVo central for update, but the address does not seen to be one of
theirs, and updates seem to be coming through fine. TiVo doesn't mention
those ports as being used by them. Their block is both in the lower < 1024
and upper range (> 8000).
204.176.49.0/24 is indeed the IP range I show my unit contacting.
Are you getting program guide updates?

Tivo, like many such services, uses port 80 to easily get through
proxies or otherwise limited interfaces, ie. if you can browse
through the connection, Tivo will connect... Unless you've blocked
either or both of the source/destination addresses. The Tivo device
initiates a TCP connection from it's next available port (>1024) to
port 80/8080 or UDP/123 (I would suspect NTP time updates) of the
Tivo server, responses are sent back from their receiving port to
your initiating port. Basic IP transport. It would appear that if
there are connection failures, it does a failover using 8080 instead
of 80.

NAT doesn't allow for outside initiated connections to NATted
addresses that aren't specifically forwarded, someone would have to
hijack a packet of an already initiated transmission, knowing the
correct packet sequence number, injected into your stream before the
true source's packet arrives. Something that's quite difficult to
do, I'd relax a little on firewalling internal devices, focus on your
external device and monitor NATed client activity for devices that
may be trojaned.

What model Netgear router is passing uninitiated packets to NATed
devices where they aren't designated as a forwarded-to device or in a
DMZ? How would it know to send something to 192.168.0.12 if the NAT
connection table isn't tracking it? I seriously doubt packets
addressed to 192.168.0.12 are making it to the external interface of
your Netgear and if they were, the Netgear should be dropping them as
irrelevant chaff on the network, it should only be listening for
packets addressed to the Adelphia assigned address on the WAN side.

Something's really foobar'd if you're seeing 1433 going to your Tivo,
you need to put a magnifying glass to those rules you have set up or
for flaws with your router, firmware possibly??

I suppose it's possible that the Tivo source code could be
compromised and is initiating stuff on 1433 but I doubt it. IIRC,
it's Linux running on PowerPC based boards.

I received a software update to my Tivo unit the other day and
haven't noticed either before or since, activity other than
destination port 80, 123 or 8080 originating from my Tivo unit. If
one of my NATed devices doesn't initiate it, my Linksys won't return
any traffic to it. Basic NAT stuff.
Obfus Kataa
2005-03-09 12:04:06 UTC
Permalink
I am keeping the thread background here. Comments and replies are
interleaved with your questions.
Post by Murray Watson
In adelphia.security-issues - article
2005 17:04:45 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
In adelphia.security-issues - article
16:22:35 -0500, Obfus Kataa says...
Post by Obfus Kataa
Post by Murray Watson
I haven't plugged up a sniffer yet but it could be ARPs, my router is
reporting very little if any of it, I've seen an dramatic increase of
UDP1026-1029/Messenger spam as well as TCP1433/SQLSpida and
UDP1434/SQLSlammer worms, some from Adelphia addresses but most from
China.
Murray,
I have been seeing TCP 1433 and pptp access attempts to my new Toshiba TiVo
(which my router drops). They have been going for five days at a low rate (a
burst, then quiet for hours), and they are all from the same UUNET address
that is not registered to anyone (other than UUNET.
Attempts to your Toshiba, is it in a DMZ or do you have multiple
Adelphia IPs? Connections to the TIVO service are TCP/80 based and
should pass straight through any NATed router. Mine goes through a
DLink wireless USB network adapter to a Linksys wireless access point
(WAP54G) through a Linksys router(BEFSR41).
It is NATed. The destination shows up as 192.168.0.12 (my Toshiba). My
Netgear drops it and logs it. Yes the SRC port is 080 on 204.176.49.2
I don't know which protocol is being used because the filter rule I have for
the two ports that have been being tickled denies access "from any to any
for any protocol." It's a rule late in the filter process after any local
addresses (i.e. on the LAN) would have been forwarded if they were trying to
speak to each other. Since the Netgear separates filters for LAN->LAN,
WAN->LAN, LAN->WAN, I can tell this is coming from outside trying to get in
to the Toshiba. I thought it might be a response to a request from my TiVo
to TiVo central for update, but the address does not seen to be one of
theirs, and updates seem to be coming through fine. TiVo doesn't mention
those ports as being used by them. Their block is both in the lower < 1024
and upper range (> 8000).
204.176.49.0/24 is indeed the IP range I show my unit contacting.
Are you getting program guide updates?
Yes, but no software updates. However, TiVo's site says that the new
DVD/Series2s from Toshiba and others will not be updated with the new
software until later this year. I tried to install the new SDK this weekend
on my desktop and discovered that the functions were not yet supported for
units with my TiVo Service number.
Post by Murray Watson
Tivo, like many such services, uses port 80 to easily get through
proxies or otherwise limited interfaces, ie. if you can browse
through the connection, Tivo will connect... Unless you've blocked
either or both of the source/destination addresses. The Tivo device
initiates a TCP connection from it's next available port (>1024) to
port 80/8080 or UDP/123 (I would suspect NTP time updates) of the
Tivo server, responses are sent back from their receiving port to
your initiating port. Basic IP transport. It would appear that if
there are connection failures, it does a failover using 8080 instead
of 80.
NAT doesn't allow for outside initiated connections to NATted
addresses that aren't specifically forwarded, someone would have to
hijack a packet of an already initiated transmission, knowing the
correct packet sequence number, injected into your stream before the
true source's packet arrives. Something that's quite difficult to
do, I'd relax a little on firewalling internal devices, focus on your
external device and monitor NATed client activity for devices that
may be trojaned.
What model Netgear router is passing uninitiated packets to NATed
devices where they aren't designated as a forwarded-to device or in a
DMZ?
If I understand the question, the answer is that it is not passing the
packets. The logs are from the router's syslog. The packets are not
getting to the TiVo. It's an RT314 with the ZyXEL firmware upgrades that
make it rather close to the old Prestige.
Post by Murray Watson
How would it know to send something to 192.168.0.12 if the NAT
connection table isn't tracking it? I seriously doubt packets
addressed to 192.168.0.12 are making it to the external interface of
your Netgear and if they were, the Netgear should be dropping them as
irrelevant chaff on the network, it should only be listening for
packets addressed to the Adelphia assigned address on the WAN side.
I have seen similar packets from DeviantArt computers in the past, using the
same ports and trying to connect to another computer on the LAN--one that
accesses DA regularly. But that computer gets daily spyware, antivirus
checks along with regular w2k critical updates. My main computer is Mac OSX
and has its own firewall running also. I blocked the port because of a
MySQL alert perhaps a year ago which coincided with a flood of incoming
traffic from DA computers, dozens of machines practically non-stop. In fact
it created logs of multiple megabytes per day and slowed our access
considerably. We scoured the DA user's computer and found no spyware. I
found no evidence in the logs that the DA user's computer was responding to
the packets.
Post by Murray Watson
Something's really foobar'd if you're seeing 1433 going to your Tivo,
you need to put a magnifying glass to those rules you have set up or
for flaws with your router, firmware possibly??
It doesn't get to the TiVo, but I wonder whether I should open that rule to
allow access only to the TiVo and monitor what it gets. perhaps I should
add a rule to let the TiVo go it on its own for a while and log every packet
it sends and receives.
Post by Murray Watson
I suppose it's possible that the Tivo source code could be
compromised and is initiating stuff on 1433 but I doubt it. IIRC,
it's Linux running on PowerPC based boards.
I received a software update to my Tivo unit the other day and
haven't noticed either before or since, activity other than
destination port 80, 123 or 8080 originating from my Tivo unit. If
one of my NATed devices doesn't initiate it, my Linksys won't return
any traffic to it. Basic NAT stuff.
I suspect the LAN specific address showing in the log as destination may be
just a convenience because the destination in the rule says "any address."
Most of the rules for incoming packets say "any address in my address space"
so you could tell which machines are being accessed by the rule itself.

The old RT314 has a rather arcane rule-writing approach.

By the way. I am also seeing packets from 204.176.49.2 aimed at ports 1723,
1900, and 1512.

All of these ports show attempts coming from other sources in addition to
(and including) 204.176.49.2 and trying to get to two other computers on the
LAN, but never to the Mac. They all originate with port 80 (from the
external computer). Only one is from a computer registered to a (former?)
Adelphia company: 209.18.34.41

Others are APNIC registered sites that are firewalled at the server I
maintain on a co-located server farm where we have seen quite a bit of MySQL
traffic and gotten three MySQL security updates relatively recently. The
APNIC ones are UDP and aimed at the Adelphia-assigned public address. So
they appear to be different. The others are TCP.
--
oK+++
"YAY!" cheered all the little people. "HOORAY FOR HE WHO IS OVERLY CRITICAL
OF OTHERS! HE IS FULFILLING HIS STATION IN LIFE AS ONLY HE CAN! NOW LET'S
GO KILL HE WHO IT IS OKAY TO MURDER!"
-James "Kibo" Parry, in "Spot Buys A CD-I(R)"
6:14 up 22 days, 12:51, 1 user, load averages: 1.82 1.47 1.25
Murray Watson
2005-03-09 14:48:34 UTC
Permalink
In adelphia.security-issues - article
<***@dlwnjh.mmo.hn>, on Wed, 9 Mar 2005
07:04:06 -0500, Obfus Kataa says...

[snip]

Do you know how NAT/PAT is supposed to work? If computers are being
addressed on your LAN, it's your router/rules doing it, it's contrary
for NAT to forward packets from an IP where it hasn't initiated the
conversation.

Can you save your Netgear configuration to a file? If so, I'd
suggest doing that, then resetting the Netgear to it's factory
defaults and seeing what you have.
Post by Obfus Kataa
I suspect the LAN specific address showing in the log as destination may be
just a convenience because the destination in the rule says "any address."
Most of the rules for incoming packets say "any address in my address space"
so you could tell which machines are being accessed by the rule itself.
Typically 0.0.0.0 is used to mean "any address" in that type of
reporting. Irregardless, they shouldn't make it past the WAN
interface if they aren't from an address/port that was initiated on
the LAN side. The NAT table in the router tracks that.
Post by Obfus Kataa
The old RT314 has a rather arcane rule-writing approach.
By the way. I am also seeing packets from 204.176.49.2 aimed at ports 1723,
1900, and 1512.
Are you seeing them only at the WAN side of the router or aimed at
specific NATed computers on your LAN.
Post by Obfus Kataa
All of these ports show attempts coming from other sources in addition to
(and including) 204.176.49.2 and trying to get to two other computers on the
LAN, but never to the Mac. They all originate with port 80 (from the
external computer). Only one is from a computer registered to a (former?)
Adelphia company: 209.18.34.41
Either the NAT code for your router is foobar'd or you have a port
forwarding rule that's causing it. It's contrary for NAT to send
traffic to an IP that hasn't initiated it.

As you describe it, something is not operating as it should there.
Have you checked that the offending packets aren't originating on
your LAN and using forged IPs? If they're not, I'd be looking at
replacing that router or correcting the rules quickly.
Post by Obfus Kataa
Others are APNIC registered sites that are firewalled at the server I
maintain on a co-located server farm where we have seen quite a bit of MySQL
traffic and gotten three MySQL security updates relatively recently. The
APNIC ones are UDP and aimed at the Adelphia-assigned public address. So
they appear to be different. The others are TCP.
Do you have an IP tunnel of some sort established to this co-located
server?
Obfus Kataa
2005-03-09 21:44:04 UTC
Permalink
Post by Murray Watson
In adelphia.security-issues - article
07:04:06 -0500, Obfus Kataa says...
[snip]
Do you know how NAT/PAT is supposed to work? If computers are being
addressed on your LAN, it's your router/rules doing it, it's contrary
for NAT to forward packets from an IP where it hasn't initiated the
conversation.
Can you save your Netgear configuration to a file? If so, I'd
suggest doing that, then resetting the Netgear to it's factory
defaults and seeing what you have.
Post by Obfus Kataa
I suspect the LAN specific address showing in the log as destination may be
just a convenience because the destination in the rule says "any address."
Most of the rules for incoming packets say "any address in my address space"
so you could tell which machines are being accessed by the rule itself.
Typically 0.0.0.0 is used to mean "any address" in that type of
reporting. Irregardless, they shouldn't make it past the WAN
interface if they aren't from an address/port that was initiated on
the LAN side. The NAT table in the router tracks that.
Actually, they aren't the logs are reporting what the Motorola modem is
sending to the WAN side. The router is dropping the packets because those
ports are not allowed for my LAN machines to receive from the WAN. But the
router is also logging them. ZyXEL firmware supports logging of individual
rules on a rule by rule basis.

I added a rule to watch the TiVo. I discovered that it is initiating the
transactions. I also tracked the packets back and forth when I initiated an
update manually through the TiVo center. It chose a different machine for
the transaction but in the same /24 block.

The conversation between the TiVo and its PhoneHomeOwner occurs every 15
minutes (+- 30 seconds). Here is a transaction sequence:
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:17, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
2005-03-09, 16:03:17, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:17, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
2005-03-09, 16:03:19, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
2005-03-09, 16:03:19, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
2005-03-09, 16:03:19, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:19, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:19, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF

S04 is a rule tied to the WAN side and only activates for traffic on that
side of the router. S02 is a similar rule that only activates on the LAN
side for traffic headed OUT of the LAN. I do not have any rules that apply
within the LAN, except on the computers themselves.

192.168.0.1 is the router (the reporting source for the syslog).
192.168.0.12 is the TiVo. After I opened access, the first transaction (at
07:32:20) used port 2275. Each transaction moves to a new port.
Post by Murray Watson
Post by Obfus Kataa
The old RT314 has a rather arcane rule-writing approach.
By the way. I am also seeing packets from 204.176.49.2 aimed at ports 1723,
1900, and 1512.
Are you seeing them only at the WAN side of the router or aimed at
specific NATed computers on your LAN.
Post by Obfus Kataa
All of these ports show attempts coming from other sources in addition to
(and including) 204.176.49.2 and trying to get to two other computers on the
LAN, but never to the Mac. They all originate with port 80 (from the
external computer). Only one is from a computer registered to a (former?)
Adelphia company: 209.18.34.41
Either the NAT code for your router is foobar'd or you have a port
forwarding rule that's causing it. It's contrary for NAT to send
traffic to an IP that hasn't initiated it.
As you describe it, something is not operating as it should there.
Have you checked that the offending packets aren't originating on
your LAN and using forged IPs? If they're not, I'd be looking at
replacing that router or correcting the rules quickly.
Yes, I am reasonably sure the packets that say Src=EXTERNAL-IP
Dst=192.168.0.X are from the WAN side. I looked at my rules, and for those
which use the 0.0.0.0/0.0.0.0 as source AND 0.0.0.0/0.0.0.0 as destination
and have logging turned on, they show up this way (i.e. the log contains the
local IP address of the LAN destination the outsider wished to contact.
Those that have 192.168.0.0/255.255.255.0 as the destination show up as the
Adelphia external address when triggered.
Post by Murray Watson
Post by Obfus Kataa
Others are APNIC registered sites that are firewalled at the server I
maintain on a co-located server farm where we have seen quite a bit of MySQL
traffic and gotten three MySQL security updates relatively recently. The
APNIC ones are UDP and aimed at the Adelphia-assigned public address. So
they appear to be different. The others are TCP.
Do you have an IP tunnel of some sort established to this co-located
server?
No, I mentioned the co-located server just as an aside. I noted that the
address trying to get through from Korea (which was showing up as
Dst=myadelphia-assigned-ip) was in an address block that I had locked out of
access to my (INNOWAYATTAACEDTOADELPHIASONOTBREAKINGTHETOS) server's mysql
applications. The records show the computers from that block had shown
signs of one of the earlier MySQL "parasites."
--
oK+++
Murray Watson
2005-03-10 00:53:05 UTC
Permalink
In adelphia.security-issues - article
<***@xxksc.ollmltjoed.hzm>, on Wed, 9
Mar 2005 16:44:04 -0500, Obfus Kataa says...
Post by Obfus Kataa
The conversation between the TiVo and its PhoneHomeOwner occurs every 15
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=192.168.0.12 Dst=204.176.49.2 TCP spo=02317 dpo=00080]}S02>R03mF
2005-03-09, 16:03:16, Local1.Notice, 192.168.0.1, cratylus: IP[Src=204.176.49.2 Dst=192.168.0.12 TCP spo=00080 dpo=02317]}S04>R04mF
They're reporting the ultimate destination, somewhat odd from my 'Nix
experience.

Your Tivo sent a packet from port 2317 to Tivo's service port 80
Tivo's service replied from their port 80 to your Tivo's port 2317.

Somewhere in there, there's another packet in your router/gateway
that has a different port number, I don't know if the packet they are
reporting is the one on the LAN side or WAN side.

There's a table in your router that maps which ports were used for
each source[port]/destination[port] conversation. The router's
programming examines each packet arriving and if the sequence numbers
match, responses from w.x.y.z[port] to your.adelphia.ip.number[port]
are forwarded to 192.168.0.12[port].

After a specific period of time(unique to implementation), entries
are dropped from the translation table and packets from w.x.y.z[port]
are simply dropped.

Your router will continue to increment the port number until
[specified time] has occured and it can begin reusing previously used
ports. That's the way NAT/PAT works.

In other words, your computer may be sending out on port 2345 but the
port your router sends to the WAN destination(ie. Tivo service) could
be 3456.
W. Mark Herrick, Jr.
2005-03-07 20:40:03 UTC
Permalink
Post by Fred Johnson
----- Original Message -----
Newsgroups: adelphia.security-issues
Sent: Wednesday, February 16, 2005 11:09 AM
Subject: Re: ARP activity
<Snipped>
Post by Murray Watson
Do you see any pattern in the IP's being ARP'd? If there is a
sequential pattern, it's most likely someone locally doing port
scanning on their own netblock but you'd probably also see some other
TCP/UDP probes directed at your IP, some find some amusement in that.
Spammers could be sequential but their software most likely does it
randomly to avoid IDS discovery.
<Snipped>
We're seeing increased arp traffic also, inside home and school lans. The
flooding looks to be a form of DoS, and originates from computers infected
with the latest SDBot worm. The amount of arp broadcasts will overload a
router in a few minutes requiring a reset.
Thoughts?
Fred,

ARP activity is a necessary precursor to communications between
devices as it provides the data link layer with the necessary mappings
between the IP and MAC addresses of communicating devices.

That being said, normal ARP activity within a CMTS can get quite
active, especially as more and more hosts connect within a given area
and, as you have noticed, if a given host becomes infected with a worm
that utilizes local areas as a primary source for scanning.

You did not mention what kind of router that you are using, but it's
been my experience that you should be able to, in most common (e.g.,
Cisco, Juniper) router brands, block other ARP requests that did not
originate from your space, or limit the ARP traffic that you are
seeing.

As always, if you are seeing a good deal of ARP traffic from a
particular host, you're free to fire off a traffic log to my staff at
***@adelphia.net, and we'll investigate.

-MH



W. Mark Herrick, Jr.
Director - Data and Network Security
Adelphia Communications
5619 DTC Parkway
Greenwood Village, CO 80111
Murray Watson
2005-03-08 07:17:11 UTC
Permalink
In adelphia.security-issues - article
<***@4ax.com>, on Mon, 07 Mar 2005
13:40:03 -0700, W. Mark Herrick, Jr. says...
Post by W. Mark Herrick, Jr.
Post by Fred Johnson
We're seeing increased arp traffic also, inside home and school lans. The
flooding looks to be a form of DoS, and originates from computers infected
with the latest SDBot worm. The amount of arp broadcasts will overload a
router in a few minutes requiring a reset.
Thoughts?
Fred,
ARP activity is a necessary precursor to communications between
devices as it provides the data link layer with the necessary mappings
between the IP and MAC addresses of communicating devices.
That being said, normal ARP activity within a CMTS can get quite
active, especially as more and more hosts connect within a given area
and, as you have noticed, if a given host becomes infected with a worm
that utilizes local areas as a primary source for scanning.
High numbers of ARPs are increased in a sparsely populated network as
their ARP table entries are culled more quickly. Messenger spammers
and pinging worms that address entire netblocks cause the high number
of requests, hosts in a consumer network don't normally initiate
ARPs.
Post by W. Mark Herrick, Jr.
You did not mention what kind of router that you are using, but it's
been my experience that you should be able to, in most common (e.g.,
Cisco, Juniper) router brands, block other ARP requests that did not
originate from your space, or limit the ARP traffic that you are
seeing.
As always, if you are seeing a good deal of ARP traffic from a
particular host, you're free to fire off a traffic log to my staff at
-MH
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.

Windows messenger spammers operate by sequentially but sometimes
randomly blind broadcasting to entire 'B' sized networks, they
generate a high number of ARPs as the respective headend doesn't have
an ARP table entry (since there's no response from an un-used IP),
each time the unused IPs are addressed by a windows messenger spammer
an ARP goes out to discover the MAC for the non-existant IP. I'm
noticing sequential transmissions to 1026 and 1027 which may result
in 2 arps per unused target IP.

Since they're UDP and no response is required, they're usually
spoofed source IPs and are nothing but trouble. You'd do well to
egress block those ports as well. I notice Baldwin filters them out
via his MNW client software so you don't see any reports on them.

If no one is clamoring to keep them open (can't think of a valid
reason) that's a quick fix for a port that has less than 0 value.

Curious, do the headends have the capability to egress filter not-
endemic IPs? If so, implementing that could offer some protection
within Adelphia's network from internally originated UDP/IP spoofs.
I'm also curious if the vendors are beginning to offer any stateful
IDS/IPS oriented features.
W. Mark Herrick, Jr.
2005-03-08 17:48:31 UTC
Permalink
On Tue, 8 Mar 2005 02:17:11 -0500, Murray Watson
Post by Murray Watson
Post by W. Mark Herrick, Jr.
As always, if you are seeing a good deal of ARP traffic from a
particular host, you're free to fire off a traffic log to my staff at
-MH
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.
Noone has specifically requested those be open, but since the posts
are above 1024, there's a higher liklihood that they'd be used for
random UDP traffic flows, yes?
Post by Murray Watson
Curious, do the headends have the capability to egress filter not-
endemic IPs? If so, implementing that could offer some protection
within Adelphia's network from internally originated UDP/IP spoofs.
I'm also curious if the vendors are beginning to offer any stateful
IDS/IPS oriented features.
Headends - no. CMTS's (which may or may not be in the headend) - yes.
I can assume that's what you meant, though. :)

Is it implemented? No - since we have no correlation between our
billing systems and CMTS's (directly), which would allow us to send a
signal to the CMTS, saying "this MAC is provisioned so it's good -
this traffic came from an unregistered MAC so drop it". Obviously,
this opens up some theft of service issues, but we've got other ways
of dealing with that.

We are currently working with some vendors to impelement some rather
interesting tools at the CMTS'ish level, that will (hopefully) be able
to ID and mitigate these types of malicious traffic, signal back to
the source (if it's internal), and then place the infected system into
a quarantine area.

Real cool stuff. :)

-MH
W. Mark Herrick, Jr.
Director - Data and Network Security
Adelphia Communications
5619 DTC Parkway
Greenwood Village, CO 80111
Murray Watson
2005-03-10 00:53:06 UTC
Permalink
In adelphia.security-issues - article
<***@4ax.com>, on Tue, 08 Mar 2005
10:48:31 -0700, W. Mark Herrick, Jr. says...
Post by W. Mark Herrick, Jr.
On Tue, 8 Mar 2005 02:17:11 -0500, Murray Watson
Post by Murray Watson
Post by W. Mark Herrick, Jr.
As always, if you are seeing a good deal of ARP traffic from a
particular host, you're free to fire off a traffic log to my staff at
-MH
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.
Noone has specifically requested those be open, but since the posts
are above 1024, there's a higher liklihood that they'd be used for
random UDP traffic flows, yes?
You have a point but while I'm probably not representative of most
consumers, I don't use any of the consumer oriented IM clients and
stuff like that, the only thing I show registering outbound on UDP in
1024-1029 are queries by some of my NTP clients. Out of 700 entries
over the last 6 months, 15 weren't originated from port 1024.
Any applications using UDP aren't too concerned about transport, the
kicker is a potential for DNS lookups to falter for a moment as they
skip over those ports.

So, it's a trade-off between a bazillion ARPs and an occasional
delay'd DNS lookup. But... one could either exclude Adelphia's DNS
servers from the rule or apply only at the borders.

In case this is not representative of what you're seeing in the glass
tower, here's a view from here. I have placed 5000 records of a
tcpdump on my web page.
http://users.adelphia.net/~mcwatson/watch.txt .
It is unretouched and contains a couple of internal packets.

You'll note that 5000 records were obtained in something just under
1.5 minutes, somewhere around 55/second. It's been like this since
the weekend, where before I rarely if ever saw any modem traffic that
wasn't initiated here. An occasional virus probe but not the steady
flow. It's possible that the messenger spammers updated their
netblock lists coinciding with the outage/changes noted here in
Mooresville NC.

That's 55 (packets/second)
x60 (3300/minute)
x60 (198,000/hour)
x24 (4,752,000/day)
x7 (33,264,000/week)
x4.3(143,035,200/month)
x365(1,729,728,000/year)
At ~1500 bytes/packet, that 660kbs of broadcast download bandwidth
consumed by ARPs

You'll note the somewhat sequential nature and double entry of the
ARPs, indicating transmission to 2 ports sequentially for each IP
(which corresponds with the 1026/1027 messenger spam sequences I'm
seeing in my logs) at sequential IPs at these times :
17:52:33.311831
17:52:39.071831
these are definitely Windows Messenger spammers and the list goes on
and on.

If your CMTS/routers with IDS/IPS could identify and partition the
sources, you'd help us recoup some of the bandwidth and traffic on
your backbone.

I think a good starting point would be to block UDP/1026-1029 from
APNIC netblocks at the border routers. Give that a try and see what
you end up blocking. China's hosts has been heavily compromised by
the Sql worms and are probably the biggest source of the messenger
spams, quite possibly not even forged IPs but one would have to do a
MAC trace to determine that, a tool I don't have in my arsenal.

You can see from the tcpdump would appear to contain several
neblocks, spanning more than my 69.164.0.0/18, more than
69.164.0.0/16
ie. 69.164.0.0 - 69.174.255.0.
not quite the 69.160.0.0/12 netblock and the 69.174.0.0 seems to be
small in number whereas 69.164.0.0/18 seems to be appropriately
represented.

Maybe the scope of the IPs is attributable to a typo in a
configuration. If Adelphia made some CMTS changes over the weekend,
I figured there would be some kinks to work out and it may take a
little while. Is the scope of these ARPs typical?

This rings as odd:
17:52:33.981831 arp who-has 67.21.36.246 tell 67.21.36.245
17:52:33.981831 arp who-has 67.21.36.246 tell 67.21.36.245
It would appear to be a router in Atlanta with an identity crisis.
Post by W. Mark Herrick, Jr.
Post by Murray Watson
Curious, do the headends have the capability to egress filter not-
endemic IPs? If so, implementing that could offer some protection
within Adelphia's network from internally originated UDP/IP spoofs.
I'm also curious if the vendors are beginning to offer any stateful
IDS/IPS oriented features.
Headends - no. CMTS's (which may or may not be in the headend) - yes.
I can assume that's what you meant, though. :)
Yes. My use of headend is a generic MSO-end equipment word. I'll
start using the correct terminology.

Back to the meat of the question then, they are capable of
identifying an offending source based on rules or input from a
honeypot type device and either blocking or rate-limiting?
Post by W. Mark Herrick, Jr.
Is it implemented? No - since we have no correlation between our
billing systems and CMTS's (directly), which would allow us to send a
signal to the CMTS, saying "this MAC is provisioned so it's good -
this traffic came from an unregistered MAC so drop it". Obviously,
this opens up some theft of service issues, but we've got other ways
of dealing with that.
Not at a MAC level, IP level. Certainly the CMTSs are IP smart, ie.
CMTS 123 manages 24.48.32.0/21 (and/or other netblocks). Giving it
the ability to reject any traffic initiated on the modem side with a
source IP != 24.48.32.0/21?? If it's not at the CMTS level then
surely at the next router level up.

While it's less significant and I can't say that I've seen it on
Adelphia but I recall an incident on RR where I was seeing traffic
from 1.2.3.4, an obviously invalid IP, one that ARIN had not and will
not delegated. Those packets should never have been seen, blocked at
the border routers, CMTS or other MSO equipment.
Post by W. Mark Herrick, Jr.
We are currently working with some vendors to impelement some rather
interesting tools at the CMTS'ish level, that will (hopefully) be able
to ID and mitigate these types of malicious traffic, signal back to
the source (if it's internal), and then place the infected system into
a quarantine area.
Real cool stuff. :)
Good to hear they are proactively working at cleaning up what has
become a big mess over the last 10 years. With what I expect to be a
fair amount of processing power available on the modem, I'd think
they would be looking a pushing some of the IDS/IPS functionality
down to the modem level, via something like a configuration file.

Thanks for your attention and feedback, it certainly facilitates
learning for all participants.
Murray Watson
2005-03-20 01:11:47 UTC
Permalink
In adelphia.security-issues - article
<***@news1.news.adelphia.net>, on Wed, 9 Mar
2005 19:53:06 -0500, Murray Watson says...
Post by Murray Watson
In case this is not representative of what you're seeing in the glass
tower, here's a view from here. I have placed 5000 records of a
tcpdump on my web page.
http://users.adelphia.net/~mcwatson/watch.txt .
It is unretouched and contains a couple of internal packets.
You'll note that 5000 records were obtained in something just under
1.5 minutes, somewhere around 55/second. It's been like this since
the weekend, where before I rarely if ever saw any modem traffic that
wasn't initiated here. An occasional virus probe but not the steady
flow. It's possible that the messenger spammers updated their
netblock lists coinciding with the outage/changes noted here in
Mooresville NC.
Following up, I'm pleased to say the 3/18/2005(UTC) went without any
inbound TCP probes from Adelphia netspace...

But it would appear that ARPs are on the increase.
5000 records in ~37 seconds, ~137/sec
http://users.adelphia.net/~mcwatson/tcpdump20050319.txt
Murray Watson
2005-03-22 04:17:15 UTC
Permalink
In adelphia.security-issues - article
<***@news1.news.adelphia.net>, on Sat, 19 Mar
2005 20:11:47 -0500, Murray Watson says...
Post by Murray Watson
In adelphia.security-issues - article
2005 19:53:06 -0500, Murray Watson says...
Post by Murray Watson
In case this is not representative of what you're seeing in the glass
tower, here's a view from here. I have placed 5000 records of a
tcpdump on my web page.
http://users.adelphia.net/~mcwatson/watch.txt .
It is unretouched and contains a couple of internal packets.
You'll note that 5000 records were obtained in something just under
1.5 minutes, somewhere around 55/second. It's been like this since
the weekend, where before I rarely if ever saw any modem traffic that
wasn't initiated here. An occasional virus probe but not the steady
flow. It's possible that the messenger spammers updated their
netblock lists coinciding with the outage/changes noted here in
Mooresville NC.
Following up, I'm pleased to say the 3/18/2005(UTC) went without any
inbound TCP probes from Adelphia netspace...
But it would appear that ARPs are on the increase.
5000 records in ~37 seconds, ~137/sec
http://users.adelphia.net/~mcwatson/tcpdump20050319.txt
Seems to leveled out
5000 records in ~38 seconds, ~131/sec
http://users.adelphia.net/~mcwatson/tcpdump20050321.txt

Murray Watson
2005-03-15 06:31:31 UTC
Permalink
In adelphia.security-issues - article <MPG.1c971b62e6659ba8989c45
@news1.news.adelphia.net>, on Tue, 8 Mar 2005 02:17:11 -0500, Murray
Watson says...
Post by Murray Watson
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.
My observations about the high number of ARPs has focused on Windows
Messenger spam, I should note that in a review of my logs, a
significant number of SQL Spida probes have been noted from Adelphia
IPs, IIRC, it performs sequential IP scanning looking for targets.

Look in your abuse desk logs from MNW on a daily basis for IPs that
are infected.

Mark, I know you won't use MNW reports as evidence but they can be an
excellent tool to ferret out problems on your network, don't cut off
your nose to spite your face.

I'm showing 5 active Adelphia IP originated TCP1433-Spida Worms
hitting me 3/13 & 3/14.

69.160.191.187 *
69.162.253.23 **
69.164.143.205 **
69.164.28.111
69.164.75.190

* MNW shows your ACK on 3/10 but I'm still getting hit on 3/14
** Of particular note - duration
W. Mark Herrick, Jr.
2005-03-15 23:22:59 UTC
Permalink
On Tue, 15 Mar 2005 01:31:31 -0500, Murray Watson
Post by Murray Watson
In adelphia.security-issues - article <MPG.1c971b62e6659ba8989c45
@news1.news.adelphia.net>, on Tue, 8 Mar 2005 02:17:11 -0500, Murray
Watson says...
Post by Murray Watson
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.
My observations about the high number of ARPs has focused on Windows
Messenger spam, I should note that in a review of my logs, a
significant number of SQL Spida probes have been noted from Adelphia
IPs, IIRC, it performs sequential IP scanning looking for targets.
Look in your abuse desk logs from MNW on a daily basis for IPs that
are infected.
Mark, I know you won't use MNW reports as evidence but they can be an
excellent tool to ferret out problems on your network, don't cut off
your nose to spite your face.
I think I said previously that I wouldn't weight MNW reports as
urgent, as they once were. We still use them, accept them, and act
upon them.

Best,
-MH


W. Mark Herrick, Jr.
Director - Data and Network Security
Adelphia Communications
5619 DTC Parkway
Greenwood Village, CO 80111
Murray Watson
2005-03-16 09:12:06 UTC
Permalink
In adelphia.security-issues - article
<***@4ax.com>, on Tue, 15 Mar 2005
16:22:59 -0700, W. Mark Herrick, Jr. says...
Post by W. Mark Herrick, Jr.
On Tue, 15 Mar 2005 01:31:31 -0500, Murray Watson
Post by Murray Watson
In adelphia.security-issues - article <MPG.1c971b62e6659ba8989c45
@news1.news.adelphia.net>, on Tue, 8 Mar 2005 02:17:11 -0500, Murray
Watson says...
Post by Murray Watson
Has anyone requested UDP ports 1026-1029 be kept open? If not, you
could cut down on a lot of the ARPs we see by blocking them at the
borders. Clears up a lot of blinking lights and some network
bandwidth.
My observations about the high number of ARPs has focused on Windows
Messenger spam, I should note that in a review of my logs, a
significant number of SQL Spida probes have been noted from Adelphia
IPs, IIRC, it performs sequential IP scanning looking for targets.
Look in your abuse desk logs from MNW on a daily basis for IPs that
are infected.
Mark, I know you won't use MNW reports as evidence but they can be an
excellent tool to ferret out problems on your network, don't cut off
your nose to spite your face.
I think I said previously that I wouldn't weight MNW reports as
urgent, as they once were. We still use them, accept them, and act
upon them.
Best,
-MH
W. Mark Herrick, Jr.
Director - Data and Network Security
Adelphia Communications
5619 DTC Parkway
Greenwood Village, CO 80111
I believe your stated scenario was that you wouldn't put your
department in a position to use them as evidence because of the
partial IP munging.
Giganews has a pretty long retention, try article
<***@4ax.com>
Or
Subject: Re: Adelphia Abuse : Infected host
Mon, 30 Aug 2004 13:19:01 -0600


I'm not trying to be an ass but I'm trying to understand a few
things.

I show fairly constant SQL Spida (TCP/1433) activity from
69.160.191.187 for the last 30 days. Not heavy, 1 hit per indicated
day.

2005-02-14
2005-02-15
2005-02-27
2005-02-28
2005-03-02
2005-03-03
2005-03-04
2005-03-14

As have 15 MNW "agents" most of which appear to be Adelphia
customers.
http://www.mynetwatchman.com/LID.asp?IID=149545158

I know I reported this IP via email to Adelphia abuse 3/5/2005 a good
2 weeks after MNW would have started reporting it, which indicates
2/27 as a "re-escalation" report, is the abuse desk running that far
behind?
I did receive an auto-ACK from Adelphia [AB-C7134092W]

Tell me if I'm using the wrong subject line
"Subject: Security issue - Source IP: 69.160.191.187"
I'm going by the Adelphia web site reference to MNW's page on how to
construct a report.

I try to facilitate the expedient handling of viruses and worms,
while I'm not going to cast any aspersions of what is worse, spam or
virus/worms, viruses and worms tend to be damaging. My thought was
your priorities may be towards "security issues" and they would be
queued as such.

For received virus reports I use :
"Subject: [email][virus][virus.name][originatingIP]"
I'm glad to say I receive few viruses originating from Adelphia's
netspace.

You may want to rethink the statement in the abuse ACK
"** PLEASE NOTE: Since most incidents are investigated and resolved
within three (3) business days, any reports older than three business
days may not be processed."

3 business days doesn't seem to apply here.

Please advise if I am doing something incorrectly. If I've uncovered
a crack in Adelphia's system, I hope you'll take corrective action.
If you can plug the hole on Slammer and Spida, that will certainly
take a little heat off the high ARP activity situation, certainly
your routers have better things to do than issue ARPs for non-
existent IPs.

I know your department is working, they disconnected me based on a
report I had sent. When talking to your person, he started reading
to me the report and I finished reading it for him. He had me turned
up in moments while I was still on the phone, a very polite and
apologetic guy.

My last question to your tech would have been who, my first questions
were what ports and when, I could then look for tell-tell signs in my
logs and/or programs. Your guy gave me those and after verifying a
no-virus environment and an absence of the indicated ports in my
logs, I asked him to re-examine the complaint which is when he
started reading my emailed complaint.

I know the variety of difficulties you face which is why I am
dumbfounded by a lack of use of automation. How much would it cost
to divert MNW reports, have a script constructed to parse them to
"individualize" each incident, giving the offending IP the requisite
"incident" volume to bring it to your attention. How much more
quickly could you resolve issues and how much time would that free
your staff up to handle the others, ones requiring human
interpretation? I think your ROI/TTP would be fairly significant.
Loading...