Skip to contentSkip to conversation

Size of Tasks during ddos

DISCUSSION Breno 9 participants 18 messages Sep 10, 2003 – Sep 23, 2003 2 messages not in the archive

lore
Share & tools
lore

Breno

author
#0 lore raw

Hi

One task like a httpd or named , during a ddos attack has your size in memory increased ?

att,
Breno

#1 lore raw

On Sat, Oct 11, 2003 at 07:34:28PM -0300, Breno wrote:
^^^^^^^^^^^^^^^^^

Sorry, but could you PLEASE fix the date on your workstation? :(

Signature · 1 line
Joshua Kwan
#2 lore raw
Quoted 5 lines · Joshua Kwan

On Sat, Oct 11, 2003 at 07:34:28PM -0300, Breno wrote:

   ^^^^^^^^^^^^^^^^^

Sorry, but could you PLEASE fix the date on your workstation? :(

I concur, for months his clocked has been skewed
in different directions, Mozilla sorting by date
hates you Breno.

-sb

#3 lore raw

Sorry Stan , but i connect from many machines.

My servers are in ddos attack , what i´d like to know is about size of tasks in memory during this kind of attack. I have some ideas to do in my kernel.

Someonde can talk about this situation ?

thanks
Breno
----- Original Message -----
From: "Stan Bubrouski" <[email protected]>
To: "Joshua Kwan" <[email protected]>
Cc: "Breno" <[email protected]>; "linux-kernel mailing list" <[email protected]>
Sent: Wednesday, September 10, 2003 11:10 PM
Subject: Re: [OT] Re: Size of Tasks during ddos

Quoted 20 lines
Quoted 5 lines · Joshua Kwan

On Sat, Oct 11, 2003 at 07:34:28PM -0300, Breno wrote:

   ^^^^^^^^^^^^^^^^^

Sorry, but could you PLEASE fix the date on your workstation? :(

I concur, for months his clocked has been skewed
in different directions, Mozilla sorting by date
hates you Breno.

(quoted diff) +0 -2
-sb
 
 
-

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#4 lore raw
Quoted 2 lines · Breno Silva

My servers are in ddos attack , what i´d like to know is about size of tasks in memory during this kind of attack. I have some ideas to do in my kernel.

The answer will differ depending whether (for example) you're being ICMP flooded, SYN-flooded, hit with a mass of HTTP 'GET /' commands, hit with a mass of HTTP commands that invoke a resource-intensive CGI like a database search, and so on.

We'd really need to know what the traffic involved in the DDoS is in order to be able to comment on memory usage.

#6 lore raw
Quoted 1 line · Breno

This is a Syn Flood DDoS

echo "1" >/proc/sys/net/ipv4/tcp_syncookies

End of problem.

#7 lore raw
Quoted 6 lines · Alan Cox
Quoted 1 line · Breno

This is a Syn Flood DDoS

echo "1" >/proc/sys/net/ipv4/tcp_syncookies

End of problem.

And why isn't this on by default when it's compiled in?

#8 lore raw
Quoted 9 lines · Mike Fedyk
Quoted 6 lines · Alan Cox
Quoted 1 line · Breno

This is a Syn Flood DDoS

echo "1" >/proc/sys/net/ipv4/tcp_syncookies

End of problem.

And why isn't this on by default when it's compiled in?

Syncookies protect you from DoS stuff but they have other side effects on efficiency when they are in use.

#9 lore raw
Quoted 5 lines · Alan Cox
Quoted 4 lines · Mike Fedyk
Quoted 3 lines · Alan Cox
Quoted 1 line · Breno

This is a Syn Flood DDoS

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

echo "1" >/proc/sys/net/ipv4/tcp_syncookies

End of problem.

And why isn't this on by default when it's compiled in?

Syncookies protect you from DoS stuff but they have other side effects on efficiency when they are in use.

Care to point me to a thread in the archives? I'd like to read more about this.

#10 lore raw
Quoted 5 lines · Mike Fedyk
Quoted 2 lines

Syncookies protect you from DoS stuff but they have other side effects on efficiency when they are in use.

Care to point me to a thread in the archives? I'd like to read more about this.

Not sure offhand where the thread is. The quick summary is

Syn cookies accept the SYN frame and encode sufficient information into the reply that they can avoid storing any data until the next packet arrives from the other end completing the connection.

That means squashing all the information we track (mss, window, etc) into very few bits. A modern TCP will offer large windows, selective ack and other features which we can't fit into a syn cookie so with this off a burst of traffic will cause pauses while the socket queue clears and negotiate fully featured TCP, with syncookies enabled many of the connections on the burst will not have the extra features so many not perform as well.

#11 lore raw
In reply to Mike Fedyk
Quoted 4 lines · Mike Fedyk

On Thu, Sep 11, 2003 at 07:41:10PM +0100, Alan Cox wrote:

And why isn't this on by default when it's compiled in?

there's several reasons; one of them is a bit cheap: a webserver benchmark done by a journalist looks a lot like a DoS in this respect ;)

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

author
#14 lore raw

Suppose that one task during a ddos receive much data , so it can try to alloc much memory to control this data, or to control the list of sockets in listen state.

att
Breno
----- Original Message -----
From: "Alan Cox" <[email protected]>
To: "Breno" <[email protected]>
Sent: Thursday, September 11, 2003 5:40 PM
Subject: Re: Size of Tasks during ddos

Quoted 3 lines · Breno

Alan

This is not the point. I´d like to know about size of tasks in memory .

What does a synflood attack have to do with that. There is no reason they should change

#15 lore raw
Quoted 3 lines · Breno

Suppose that one task during a ddos receive much data , so it can try to alloc much memory to control this data, or to control the list of sockets in listen state.

Syncookies dont allocate memory until the connection finishes the 3 way handshake with the other side

#16 lore raw
In reply to Breno
Quoted 3 lines · Breno

Suppose that one task during a ddos receive much data , so it can try to alloc much memory to control this data, or to control the list of sockets in listen state.

Hi Breno,

Can you please fix your clock? Thanks
--
vda

#17 lore raw

Mail header times was: Size of Tasks during ddos

Quoted 8 lines · insecure
Quoted 3 lines · Breno

Suppose that one task during a ddos receive much data , so it can try to alloc much memory to control this data, or to control the list of sockets in listen state.

Hi Breno,

Can you please fix your clock? Thanks

Sorting by received date is your friend. ;) If kmail doesn't have that, please file a bug report.

Breno, and ntp is your friend. :-D

#18 lore raw
In reply to [email protected]
Quoted 7 lines · On Thu, Sep 11, 2003 at 10:19:37AM -0400, [email protected] wrote:

The answer will differ depending whether (for example) you're being ICMP flooded, SYN-flooded, hit with a mass of HTTP 'GET /' commands, hit with a mass of HTTP commands that invoke a resource-intensive CGI like a database search, and so on.

We'd really need to know what the traffic involved in the DDoS is in order to be able to comment on memory usage.

True, but it's not a ddos unless they do everything they can to disable the target system. Sure they could just flood your net pipe, but why do that when you could have fewer senders and completely kill the box for a long time while it tries to process all of your requests (assuming you're running services accessable from the net).