-----Original Message----- From: Rainer Gerhards <[email protected]> Reply-To: rsyslog-users <[email protected]> Date: Tuesday, April 8, 2014 at 1:45 AM To: rsyslog-users <[email protected]> Subject: Re: [rsyslog] elasticsearch RPM for el5
>On Tue, Apr 8, 2014 at 1:53 AM, Mike Hoskins (michoski) ><[email protected]>wrote: > >> -----Original Message----- >> From: Andre Lorbach <[email protected]> >> Reply-To: "[email protected]" <[email protected]>, rsyslog-users >> <[email protected]> >> Date: Monday, April 7, 2014 at 10:19 AM >> To: rsyslog-users <[email protected]> >> Subject: Re: [rsyslog] elasticsearch RPM for el5 >> >> >Elasticsearch requires libuuid and libcurl,. Either one or both of >>these >> >libs were outdated and could not be rebuild from newer sources easily. >>So >> >I had to remove support for elasticsearch on EHEL5. >> >> I'm almost entirely on CentOS 6.x now so don't care personally, but it's >> ashame...since RHEL/CentOS 5.x is not "old" by any means, 5.10 was just >> released end of last year, and it's pretty reasonable to expect larger >> enterprises -- which are probably also the ones who would fund projects >>-- >> > >I like this "theory of funding". Actually, nobody objects trying to work >on >this if the project is funded, and if a sponsor would come up and request >that this and that feature must be available on RHEL 5 AND fund that work, >we'd be more than happy to do that. The plain fact is that this is a myth. >Nobody wants to fund such work, which IMHO is also a very good conclusion >that nobody *seriously* wants it. Yep, fully understand... A colleague and I have spent the past six months building a PoC, writing up our justification and getting required budget approval to fund development work for patches to OpenLDAP (various bugs and new features in pcache and translucent) with a contract to return all work to OSS community. So it is possible to get funding, and do it right, but it's rather arduous even on the inside. :-( As I've mentioned in the past, I was somewhat recently handed a collection of environments using logstash. All I (well, my users) really care about is getting data into elasticsearch. I try not to rip and replace just because I'm more comfortable with something, but for years all log infra I've built has been rsyslog based (since replacing syslog-ng). I love rsyslog, and logstash has given me several real issues (performance/stability) so I am waiting to find free time to build a rsyslog based PoC that accomplishes the same functionality as what my colleague built (he's conveniently reassigned and unavailable to help...big pond). So it's quite possible I'll be able to arrange funding similar to what we've done for OpenLDAP...but I need time to "play" and find out. That said, I'd be deploying everything on latest OS versions, so funding would be more bug fix/feature request oriented. Certainly can't complain about free work, just pointing out that no matter how "current" (or not) people think a given RHEL release happens to be...large enterprises will often be using current-1 and basing a lot of infrastructure on it (similar thought process to how Solaris versions used to be adopted) -- although many of those will also be paying millions for commercial log solutions so probably moot. :-) _______________________________________________ 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.
