Hi Ray,
I tried your config and it works fine (only telnet localhost 7000 is accepted).
Check the log if the status port connection really comes from 127.0.0.1.
Regarding compilation - it's tricky if you want to build everything more than
once from different versions. And sometimes your OS comes with preinstalled
PWLib/OpenH323 libraries/headers. And building a broken executable happens
even to me sometimes, when I am not careful.
The 2.2.x line seems to be a really stable stuff now, so go ahead with tests:)
Regards, Michal
----- Original Message ----- From: "Ray Jackman" <ray@xxxxxxxxx>
Sent: Friday, February 04, 2005 10:41 PM
hi all.
ive compiled 2.2.1 from last nites CVS and have it running on my test machine.
i wil make some comments about the compile but first i have a quick question.
has the ruleset changed to allow access from telnet to status port. i used
existing config but it denies me access. i read the docs (2.2.1) and did not
see any change. the rule is as follows.
[GkStatus::Auth] rule=explicit 127.0.0.1=allow default=forbid
now i will make some comments regarding the compile.
my environ is debian 2.4.22 testing release (bit old but only runs GK and has
been stable so far) building optnoshared versions.
pwlib and openh323 is from CVS
i experienced problems as mentioned from several posts regarding compile
errors (mostly do to LD linker) and when compiled i also had seqmentation
faults. some things i learned were as follows.
for pwlib (make first) and openh323 always make clean, make optdepend and run
configure and i also make both optnoshared. if there are complaints during
these compiles, gnugk will also not compile correctly.
compile on the target machine. a build i made on one machine produced a seq
fault on another.
take things out you dont use, like --disable-pqsql if your not using postgress
for example. building things you dont use makes the executable larger and
relies on more libs.
its mostly better to compile the kernel with the same libs on the machine. its
very common to update GCC and libs but not compile a new kernel. in this case
i found is some cases caused probs. re-compiling a kernel with the existing
libs and compilers was mostly safer for me.
otherwise im ready to test 2.2.1 and then put in production on a GK doing production calls to stress test it.
thanks as always and regards ray
------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________________
List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549 Homepage: http://www.gnugk.org/
