At 6:31 PM +0200 9/21/99, J. op den Brouw wrote:
> PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
> 6439 msql -25 0 50M 11M run 35:09 99.10% 97.43% htdig
>
>The size is big, round 50M (11M res).
>Any hints?
>
>By the way, retrieving with a separate htdig + conf file, only digging
>the "slow" files, works perfect, no slowdown.
At 6:56 PM +0200 9/21/99, J. op den Brouw wrote:
>Htdig is running like a mad dog after the document is received.
>I can't find a network connection with netstat corresponding
>to the retrieval of that document, so it isn't the web server.
>
>Strange thing is: URLs pointing to other servers are htdig'd
>with not slowing down, allthough they are on the same pool
>of servers....
Hmph. Sounds like there are some bugs to squash in the connection
code. Can you find the connection for that particular document in the
server log? Was the server heavily loaded at that point?
Gabriele and I are in the middle of a higher-level rewrite
(HtHTTP/Transport), but perhaps we want to revisit all the networking
code. Loic's suggestion on a test suite would help, but I'd be at a
bit of a loss for the base cases. Would we need to write/copy a TCP
sniffer, or am I missing something?
Any suggestions? Should we break the networking code out into a
separate shared library (htnet)?
-Geoff Hutchison
Williams Students Online
http://wso.williams.edu/
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED] containing the single word "unsubscribe" in
the SUBJECT of the message.
