Dear sir,

    Hi.  This is imacat from Taiwan.  I'm the current list manager of TLUG
(Taiwan Linux Users' Group).

    Our server was crashed last night at 23:30pm.  After I reboot it I found
immediately 30-40 active web connections from a single host, and the
User-Agent from the Apache access_log is:

Mozilla/4.5 (compatible; HTTrack 3.0x; Windows 98)

    This, I believe, is your product HTTrack.

    Since our web site runs MySQL, that means simutanuously 30-40 MySQL
queries.  Our server was not expecting that amount of simutanuous requests. 
The request starts from 9:30pm.  After 2 hours of hard working our server
can't stand anymore and crashed at 11:30pm.  I got notified and reboot the
server at 1:00am, Sunday midnight.  The server was off for 2 hours.  We have
lots of clients on that server.  All of them are shut down just for HTTrack.

    Now I consider HTTrack as dangerous.  I have apply some filter to all our
websites specificly on HTTrack, and have spread this alert to all the
newsgroups, mailing lists I'm attending to.

    We welcome offline browsers, like wget.  Wget requests one resource at a
time.  That is polite.  The server can always handle it.  I'm in the hope that
all offline browsers are polite.  I cannot expect the users to be polite since
I have no clue who they are at all, and even if I have, mostly the problem has
already ceased before I reach them.  I wish that HTTrack can enforce the limit
of simutanuous connections and the user politeness.  Can you help on this? 
Until then, I shall consider HTTrack as safe, release the restriction on
HTTrack on all our websites and notify everyone about this change.