ok, thank you for merge I was roughly through Drvr.GetRemoteIP and since there are no checks of returned pointer I just wanted to make sure we'll never send a NULL to printf (which anyway should not be a problem)
Best regards Radoslad Bodo Dne 08/26/14 15:04, Rainer Gerhards napsal(a): > FYI: I have merged the patch now, but removed the "N/A" check. You can see > the commits on github. Again, if you see these "N/A", we need to dig into > what's going on. > > Thanks again! > Rainer > > > On Tue, Aug 26, 2014 at 1:52 PM, Rainer Gerhards <[email protected]> > wrote: > >> I mean the N/A case... >> >> Sent from phone, thus brief. >> Am 26.08.2014 13:45 schrieb "bodik" <[email protected]>: >> >> On 08/26/14 13:27, Rainer Gerhards wrote: >>>> On Tue, Aug 26, 2014 at 12:49 PM, bodik <[email protected]> wrote: >>>> >>>>> we have no problem with ASL2.0 >>>>> >>>>> if you merge our proposal please use the latest version from >>>>> >>>>> >>>>> >>> http://home.zcu.cz/~bodik/metasw/rsyslog2/metapatches/tcpsrv-peeraddr-logging1.patch >>>>> >>>>> which has just cosmetic difference ("NA" vs "N/A") >>>>> >>>>> >>>> do you have actual cases where this happens? I ask because this would >>> be a >>>> bug that needs different addressing. The property needs to be set at >>> that >>>> point of processing. >>> >>> we see those log lines in production from time to time, and since we are >>> unable >>> to tell which node is failing to connect, we choose to enhance logging. >>> also we >>> are able to simulate it during our QA testing phase using simulated >>> network >>> failures ( >>> http://home.zcu.cz/~bodik/metasw/rsyslog2/test02/test_clients.sh) >>> >>> >>> >>> I also tried a different approach with getpeername() and getnameinfo(), >>> but it >>> generaly does not work in every situation since I cann't get proper >>> information >>> on failed socket [ENOTCONN] >>> >>> >>> >>> So I choosed to use property which should be present and set at the very >>> begining of the connection (both tcp or gssapi) >>> >>> >>> http://git.adiscon.com/?p=rsyslog.git;a=blob;f=tcpsrv.c;h=3770a02cfda7ce8f372ce9f6040acef645f799c8;hb=HEAD#l456 >>> >>> and whole thing is based on actual rsyslog code >>> >>> >>> http://git.adiscon.com/?p=rsyslog.git;a=blob;f=tcpsrv.c;h=3770a02cfda7ce8f372ce9f6040acef645f799c8;hb=HEAD#l556 >>> >>> >>> >>> If there's something I'm missing we could elaborate further on, the goal >>> is to >>> get the IP address on the server side in case of failing reading from >>> socket >>> >>> >>> >>> ----------------------------- >>> the real world example: >>> >>> Aug 26 07:59:59 127.0.0.1 xxx Aug 26 07:59:59 rsyslogd-2101: netstream >>> session >>> 0x7f67d8658570 will be closed due to error >>> Aug 26 08:00:04 127.0.0.1 xxx Aug 26 08:00:04 rsyslogd-2101: netstream >>> session >>> 0x7f67d83f3d20 will be closed due to error >>> Aug 26 08:02:58 127.0.0.1 xxx Aug 26 08:02:58 rsyslogd-2101: netstream >>> session >>> 0x1862e50 will be closed due to error >>> >>> _______________________________________________ >>> rsyslog mailing list >>> http://lists.adiscon.net/mailman/listinfo/rsyslog >>> http://www.rsyslog.com/professional-services/ >>> What's up with rsyslog? Follow https://twitter.com/rgerhards >>> NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad >>> of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you >>> DON'T LIKE THAT. >>> >> > _______________________________________________ > rsyslog mailing list > http://lists.adiscon.net/mailman/listinfo/rsyslog > http://www.rsyslog.com/professional-services/ > What's up with rsyslog? Follow https://twitter.com/rgerhards > NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of > sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T > LIKE THAT. > _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
