7df5677b46
fixed Curl_freeaddrinfo() to only free addrinfo, and added Curl_freednsinfo()
...
for freeing single dns cache entries
2002-11-26 09:41:54 +00:00
ca6e770837
The test for DNS cache entries left locked is now only built if
...
AGGRESIVE_TEST is also defined, as an addition to MALLOCDEBUG. It doesn't
work for multi interface usage and should only be used with careful
consideration.
2002-11-11 22:51:09 +00:00
2cff251863
Curl_resolv() now returns a different struct, and it contains a reference
...
counter so that the caller needs to decrease that counter when done with
the returned data.
If compiled with MALLOCDEBUG I've added some extra checking that the counter
is decreased before a handle is closed etc.
2002-11-05 10:51:41 +00:00
5bc78cb724
Disable the DNS cache (by setting the timeout to 0) made libcurl leak
...
memory. Avery Fay brought the example code that proved this.
2002-10-31 13:09:11 +00:00
32c03eadd6
glibc 2.2.93 gethostbyname_r() no longer returns ERANGE if the given buffer
...
size isn't big enough. For some reason they now return EAGAIN.
Redhat 8 ships with this glibc version.
2002-10-21 13:20:30 +00:00
84fa12c885
Cris Bailiff found this flaw, gethostbyname_r() on linux returns 0 even
...
when it can't lookup the name (at least in some cases) and thus we need
to make an extra check to detect failures.
2002-09-30 08:07:16 +00:00
ba4e69bebc
updated source code boilerplate/header
2002-09-03 11:52:59 +00:00
7438c378be
prevent compiler warning
2002-08-30 11:10:50 +00:00
ca5678c8c1
attempted hpux fix for resolves
2002-08-08 22:44:43 +00:00
7f8ad34ba2
Glen Nakamura made his fix look even better!
2002-06-26 15:39:23 +00:00
a33e89b44a
Glen Nakamura's patch for Curl_getaddrinfo().
2002-06-26 08:03:46 +00:00
3c49b405de
Now uses sigsetjmp() and siglongjmp() to bail out from slow name lookups in
...
case a timeout is set. This seems to work. God knows if it is good enough
or what kind of side-effects we introduce here and now.
I'll close my eyes and cross my fingers. Hard.
2002-06-11 15:10:18 +00:00
11ba367fc9
getaddrinfo() failures now show port number too in informational output
2002-06-10 13:26:02 +00:00
0db227f55e
hostcache_fixoffset() is now corrected to work on 64bit architectures
...
Bug report #564585 .
2002-06-05 13:41:33 +00:00
9263652c6d
Fixes bug report #556930 - we need to make sure that the data is all right
...
after we've realloc() the packed hostent struct.
2002-05-17 07:49:28 +00:00
b1becd0ed5
Jacky Lam's fix to make the realloc() of the hostent data work properly
...
even when the realloc() actually gets a new memory block
2002-05-01 11:36:13 +00:00
8358505b6d
Now uses Curl_ as prefix for internal global symbols. curl_ should only be
...
used for "exported" globals.
2002-04-27 13:07:51 +00:00
8927ddec16
In order to not get problems with DNS cache pruning, we no longer store
...
any name resolved data in any curl handle struct. That way, we won't mind
if the cache entries are pruned for the next time we need them. We'll just
resolve them again instead.
This changes the Curl_resolv() proto. It modifies the SessionHandle struct
but perhaps most importantly, it'll make the internals somewhat dependent
on the DNS cache not being disabled as that will cripple operations somewhat.
Especially for persistant connections.
2002-04-25 19:00:57 +00:00
db6d4bcf47
the 80 column police narrowed this source code! B-]
2002-04-23 14:56:21 +00:00
53a9fdf078
use sclose() to close sockets
2002-04-22 23:53:49 +00:00
21fc402c01
Jacky Lam's adjust resolve-buffer size patch applied. Slightly edited
...
by Daniel.
2002-04-22 13:31:16 +00:00
381f77756d
pack_hostent does not handle 64 bit pointers correctly.
...
A Bjørn Reese patch.
2002-04-19 11:00:26 +00:00
a386562d9a
Prune old hostcache entries with each call...
...
This can be optimized a tidbit, but this is a start.
2002-04-17 20:13:55 +00:00
974f314f57
copyright string (year) update
2002-03-19 07:54:55 +00:00
fb29529a52
Jun-ichiro itojun Hagino <itojun@itojun.org>:
...
Now first check if IPv6 is supported, then use PF_UNSPEC. If not, use PF_INET.
It'll solve both the "slow name lookup" problem on IPv4 and still work fine on
IPv6 hosts.
Bug report #530204 has more details:
http://sourceforge.net/tracker/?func=detail&atid=100976&aid=530204&group_id=976
2002-03-15 09:54:30 +00:00
3cd2673077
bug report #530204 correctly identified that revision 1.52 broke ipv6
...
functionality and this change reverts this.
However, with this revert we bring back problems on (some/all?) non-IPv6
enabled Linux machines that have getaddrinfo().
2002-03-15 08:45:09 +00:00
721b05e343
Nico Baggus' VMS tweaks
2002-02-20 13:46:53 +00:00
3d5732d4e0
Rick Richardson's getaddrinfo() usage fix to speed up name resolves
2002-02-18 23:12:37 +00:00
b795929858
INADDR_NONE should be in_addr_t to work with 64bit archs better.
...
Really, we should only #define this in one file, not both here and in
connect.c!
2002-02-18 22:59:26 +00:00
3474ec4ecb
_num_chars did wrong when called with a number that starts with 1!
2002-01-31 07:51:06 +00:00
6212e6990a
someone should have me punished, but this bug made curl bug seriously
...
on IPv4-linux machines
2002-01-28 19:23:18 +00:00
e74b20926d
prevents gcc -Wcast-align from complaining
2002-01-18 12:59:33 +00:00
edcbf4350b
include our own sprintf() prototype to make it return sensible data on
...
all platforms, I also edited a few data types slightly to prevent my
compiler from warning on comparisions between signed and unsigned values
2002-01-17 08:03:48 +00:00
9289ea471f
Get this working, still need to check for leaks and such, but should be
...
fine..
2002-01-17 07:38:25 +00:00
7d06185aa6
Make the keys for hostcache entries be in the format::
...
host:port, so accessing curl.haxx.se on port 80 would yield a key value
of ::
curl.haxx.se:80
2002-01-17 06:55:37 +00:00
fefc7ea600
a memory leak when name lookup failed is now removed
2002-01-14 23:14:24 +00:00
75bba0da92
added two typecasts to prevent compiler (gcc3) warnings
2002-01-08 07:06:07 +00:00
c0bfe7be15
1) the dns_cache_timeout should be an integer, not a bool
...
2) in the curl_dns_cache_entry structure, timestamp should be
a time_t instead of an integer (although I doubt it matters).
2002-01-08 04:30:59 +00:00
22ac08e06d
Add support for DNS cache timeouts via the CURLOPT_DNS_CACHE_TIMEOUT option.
...
The default cache timeout for this is 60 seconds, which is arbitrary and
completely subject to change :)
2002-01-08 04:26:47 +00:00
bd878756fc
Probably not necessary, but good practice.
2002-01-07 20:55:35 +00:00
8d7f402efb
Make cach'ing work with threads now, there are now three cases:
...
- Use a global dns cache (via setting the tentatively named,
CURLOPT_DNS_USE_GLOBAL_CACHE option to true)
- Use a per-handle dns cache, by default
- Use a pooled dns cache when in the "multi" interface
2002-01-07 20:52:32 +00:00
6de7dc5879
Sterling Hughes' provided initial DNS cache source code.
2002-01-03 10:22:59 +00:00
fe37fb5921
Philip Gladstone's 64-bit sparc native compiler compatibility issues fixed.
2002-01-02 10:06:47 +00:00
436d147925
Eric's #include fixes for better macos compiles
2001-12-04 13:03:08 +00:00
4e276b1b68
pack_hostent() now aligns the data properly on 64bit boundaries to work on
...
more CPU architectures
2001-10-23 12:11:43 +00:00
010044e03c
the malloc debug system only logs data if the logfile FILE * is set, which
...
makes it easier to disable debug output when built with debug functions
2001-10-17 12:33:35 +00:00
8e91d5de8e
looks nicer and is better compatible with older vim versions
2001-10-11 09:32:19 +00:00
47def8091c
made sure the correct pieces of code are compiled on IPv4-only hosts, and
...
the pack_hostent() is only compiled if gethostbyname_r() isn't present.
2001-10-04 13:36:11 +00:00
5d9ae88f58
getaddrinfo() cleanups
2001-10-04 13:25:12 +00:00
9d066935e5
Keith McGuigan's excellent fix that makes a cloned copy of the hostent struct
...
for when gethostbyname() is used so that we have the memory of the struct
allocated. This turns out to be needed if the curl handled is passed between
threads on Windows and possibly other operating systems where we use that
function.
2001-10-03 21:42:04 +00:00
Daniel Stenberg