Re: [pmacct-discussion] comparison is always false due to limited range of data type

2005-08-22 Thread Jamie Wilkinson
This one time, at band camp, Paolo Lucente wrote: >i've solved the warning generated by pmacct.c:605 . Rather than simply removing >the if-block, i've added a temporary (signed) int variable to do the job (which >then written into the u_int8_t one). I've done this because of the atoi() call >which

Re: [pmacct-discussion] comparison is always false due to limited range of data type

2005-08-22 Thread Paolo Lucente
Hello Jamie, i've solved the warning generated by pmacct.c:605 . Rather than simply removing the if-block, i've added a temporary (signed) int variable to do the job (which then written into the u_int8_t one). I've done this because of the atoi() call which *need* to be checked (user input from the

Re: [pmacct-discussion] [PATCH] eliminating compile warnings

2005-08-22 Thread Paolo Lucente
On Mon, Aug 22, 2005 at 06:39:33PM +1000, Jamie Wilkinson wrote: > Yeah, but surely it should only be defined in one place? [ ... ] > That's what I thought you were doing, so as it's not really a max value, but > more a cache closed threshold, perhaps you could change its name to > CACHE_CLOSED_

Re: [pmacct-discussion] [PATCH] eliminating compile warnings

2005-08-22 Thread Jamie Wilkinson
This one time, at band camp, Paolo Lucente wrote: >Hey Jamie, > >On Mon, Aug 22, 2005 at 03:44:36PM +1000, Jamie Wilkinson wrote: > >> Also the fact that this constant is defined in two places (but appears to >> only be used in sql_common.h) sets of alarm bells in my mind; and the fact >> that it's

Re: [pmacct-discussion] [PATCH] eliminating compile warnings

2005-08-22 Thread Paolo Lucente
Hey Jamie, On Mon, Aug 22, 2005 at 03:44:36PM +1000, Jamie Wilkinson wrote: > Also the fact that this constant is defined in two places (but appears to > only be used in sql_common.h) sets of alarm bells in my mind; and the fact > that it's not really the value of maximum size of a 32 bit unsigne

Re: [pmacct-discussion] [PATCH] eliminating compile warnings

2005-08-22 Thread Jamie Wilkinson
This one time, at band camp, Jamie Wilkinson wrote: >There's an error that reports that UINT32TMAX is only unsigned in ISO C99, >so here's a patch that helps it along: Also the fact that this constant is defined in two places (but appears to only be used in sql_common.h) sets of alarm bells in my

[pmacct-discussion] [PATCH] eliminating compile warnings

2005-08-22 Thread Jamie Wilkinson
There's an error that reports that UINT32TMAX is only unsigned in ISO C99, so here's a patch that helps it along: --- orig/print_plugin.h +++ mod/print_plugin.h @@ -26,7 +26,7 @@ #define DEFAULT_PRINT_REFRESH_TIME 10 #define AVERAGE_CHAIN_LEN 10 #define PRINT_CACHE_ENTRIES 16411 -#define UINT32

[pmacct-discussion] comparison is always false due to limited range of data type

2005-08-22 Thread Jamie Wilkinson
I've got a bug report about compile failures on amd64 platforms with GCC 4, so I'm going through the compiler warnings and trying to fix some of the more obvious ones. One of these is in pmacct.c:605, which causes the error in the subject line. The data structure pkt_primitives declares proto to

Re: [pmacct-discussion] -i option ignored

2005-08-22 Thread Paolo Lucente
Hey Jamie, On Mon, Aug 22, 2005 at 11:55:17AM +1000, Jamie Wilkinson wrote: > I've just found that -i on the command line no longer works; > try creating a config file without an interface: directive, and specify -i > eth1 -f config; you'll see that pcap_lookupdev gets called and chooses as if > n

[pmacct-discussion] -i option ignored

2005-08-22 Thread Jamie Wilkinson
I've just found that -i on the command line no longer works; try creating a config file without an interface: directive, and specify -i eth1 -f config; you'll see that pcap_lookupdev gets called and chooses as if no interface was specified at all. Here's a patch that just adds the debug log showin

Re: [pmacct-discussion] stats in 0.9.1

2005-08-22 Thread Paolo Lucente
Hey Jamie, On Mon, Aug 22, 2005 at 11:37:08AM +1000, Jamie Wilkinson wrote: > I've got a patch that also puts the interface name in the log, because I run > a couple of pmacctds on different interfaces. Patch accepted: it's a very good idea to add interface infos. It will appear next official r

Re: [pmacct-discussion] stats in 0.9.1

2005-08-22 Thread Jamie Wilkinson
This one time, at band camp, Paolo Lucente wrote: >Hello Jamie, > >On Thu, Aug 18, 2005 at 12:27:30PM +1000, Jamie Wilkinson wrote: > >> I'm testing out 0.9.1 at the moment, and have a question about the stats >> when pmacctd receives SIGUSR1. >> >> Is the first number in parentheses the unix time