Bugtraq mailing list archives
Re: a cheesy Apache / IIS DoS vuln (+a question)
From: Michal Zalewski <lcamtuf () dione ids pl>
Date: Thu, 4 Jan 2007 09:18:23 +0100 (CET)
On Wed, 3 Jan 2007, William A. Rowe, Jr. wrote:
Seriously, HTTP pipelining can accomplish EXACTLY the same thing with minimal pain.
No, it can't. Client-side pipelining using simultaneous sessions with keep-alives is usually severely restricted on server-side (exactly for the reason they can be abused to DoS the server), and the overhead is much higher (you need to send hundreds of bytes to request a copy of a static page). Here, requesting an extra copy costs you 3 bytes, and that is in my opinion notable - because you can do it thousands of times in a single short request.
If you have an issue with this behavior, of HTTP, then you have an issue with the behavior under FTP or a host of other protocols.
Not really; see above. These are typically well known, preventable by configuring server-side limits, and require a much higher overhead. /mz
Current thread:
- a cheesy Apache / IIS DoS vuln (+a question) Michal Zalewski (Jan 03)
- Re: a cheesy Apache / IIS DoS vuln (+a question) William A. Rowe, Jr. (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Michal Zalewski (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) William A. Rowe, Jr. (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Michal Zalewski (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Michal Zalewski (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Michal Zalewski (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) William A. Rowe, Jr. (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Gadi Evron (Jan 08)
- Re: a cheesy Apache / IIS DoS vuln (+a question) Rob Sherwood (Jan 04)
- Re: a cheesy Apache / IIS DoS vuln (+a question) William A. Rowe, Jr. (Jan 09)
- Re: a cheesy Apache / IIS DoS vuln (+a question) bugtraq (Jan 10)