On my 7.2 box with 2.96 and 3.3.2 compilers, the configure and build appear to
work OK - running the resulting gmond yields:
[EMAIL PROTECTED] gmond]$ ./gmond
Couldn't create socket
The test daemon then exits to shell.
On my SuSE 9.1 system (which is somewhat mroe vanilla), it configures, builds
and runs in foreground without complaint, but I don't seem to be able to
connect via localhost:8649...
Don't really have time to look into this further right now but I'll try to
answer any questions...
Matt Massie wrote:
guys-
i have a 2.6.0 test that i'd like you to run on your respective
operating systems.
download http://matt-massie.com/ganglia/ganglia-apr-2.6.0.tar.gz
&& ./configure && make
i think i'm going to scrap the libunp code from steven's book. i think
steven's is the master but getaddrinfo() is broken on too many systems
that say they support it (and the network library is based on it).
this is not a complete tarball at all. when i hear from everyone about
whether it works okay.. i'll move the rest of the gmond/gmetad etc code
into the distribution.
if you cd into the ./gmond directory you'll see a gmond daemon. it
doesn't do much but i wrote it to act like a test.
start gmond and then connect to it on port 8021. gmond will output the
address that it thinks you are coming from. to test IPv4/IPv6 sanity ..
try this
% telnet localhost 8021
and then
% telnet :: 8021
if your telnet client is IPv6 enabled. you should find that it
correctly understands where you are coming from.
the code for this new framework is based on apr.apache.org .. the apache
runtime library... i've tested it on linux ia32/ia64 and cygwin and it
works as it should. a coworker is on the machine with my freebsd
virtual machine or i'd check it there too.
martin... can you let me know what you get on your IPv6 box? i have my
laptop up and running some IPv6 interfaces. it looks pretty sane on
rh9.
since apache runs on just about every machine out there... i think this
might be the most portable approach.
let me know.
-matt
