On Wed, 19 Mar 2003 20:33:48 +0100 (CET)
  Daniel Stenberg <daniel_at_haxx.se> wrote:
>On Wed, 19 Mar 2003, Nisan Bloch wrote:
>
>> I cannot make it happen, but within a few 100 requests I
>>get the segfault,
>> allways in the same place.
>
>Are they all using different host names in the URLs?
>It is obviously a crash in the pruning code that removes
>cashed DNS entries.
>

ech thread runs through a database and doest status
callbacks to our clients, so yes at time many many
different host names and at others batches of the same
hostname over and over.

>You have the timeout set to 0, which should make one
>entry to remove on every
>call, since the previous one should be pruned then.
>

yup.. thats what I thought.

>Do you think you could check out what the DNS cache
>contains at the moment it
>crashes?

How do I do that?

>If you increast the DNS timeout, does the
>problem rate change?
>

yes

>I know it is painful to work like this, but as I can't
>repeat this myself I
>can only sit here and suggest things...
>

Thats ok.. At least you are helping.. And a lot better
than some other open source projects

ta

Nisan
>--
> Daniel Stenberg -- curl, cURL, Curl, CURL. Groks URLs.
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: Does your code think
>in ink?
>You could win a Tablet PC. Get a free Tablet PC hat just
>for playing.
>What are you waiting for?
>http://ads.sourceforge.net/cgi-bin/redirect.pl?micr5043en
>

-------------------------------------------------------
This SF.net email is sponsored by: Does your code think in ink?
You could win a Tablet PC. Get a free Tablet PC hat just for playing.
What are you waiting for?
http://ads.sourceforge.net/cgi-bin/redirect.pl?micr5043en
Received on 2003-03-21