Sage, I also support CivetWeb over Apache+FAST CGI. I tried HAProxy with multiple CivetWeb+RGW instances, it performs very well. It is easy to configure and gives better response time. On Thu, Feb 26, 2015 at 4:09 PM, Irek Fasikhov <malmyzh@xxxxxxxxx> wrote: > I fully support Wido. We also have no problems. > > OS: CentOS7 > [root@s3backup etc]# ceph -v > ceph version 0.80.8 (69eaad7f8308f21573c604f121956e64679a52a7) > > > 2015-02-26 13:22 GMT+03:00 Dan van der Ster <dan@xxxxxxxxxxxxxx>: >> >> Hi Sage, >> >> We switched from apache+fastcgi to civetweb (+haproxy) around one >> month ago and so far it is working quite well. Just like GuangYang, we >> had seen many error 500's with fastcgi, but we never investigated it >> deeply. After moving to civetweb we don't get any errors at all no >> matter what load we send to the gateways. >> >> Here are some details: >> - the whole cluster, radosgw included, is firefly 0.80.8 and >> Scientific Linux 6.6 >> - we have 6 gateways, each running on a 2-core VM >> - civetweb is listening on 8080 >> - haproxy is listening on _each_ gateway VM on 80 and 443 and >> proxying to the radosgw's >> - so far we've written ~20 million objects (mostly very small) >> through civetweb. >> >> Our feedback is that the civetweb configuration is _much_ easier, much >> cleaner, and more reliable than what we had with apache+fastcgi. >> Before, we needed the non-standard apache (with 100-continue support) >> and the fastcgi config was always error-prone. >> >> The main goals we had for adding haproxy were for load balancing and >> to add SSL. Currently haproxy is configured to balance the http >> sessions evenly over all of our gateways -- one civetweb feature which >> would be nice to have would be a /health report (which returns e.g. >> some "load" metric for that gateway) that we could feed into haproxy >> so it would be able to better balance the load. >> >> In conclusion, +1 from us... AFAWCT civetweb is the way to go for Red >> Hat's future supported configuration. >> >> Best Regards, Dan (+Herve who did the work!) >> >> >> >> >> On Wed, Feb 25, 2015 at 8:31 PM, Sage Weil <sweil@xxxxxxxxxx> wrote: >> > Hey, >> > >> > We are considering switching to civetweb (the embedded/standalone rgw >> > web >> > server) as the primary supported RGW frontend instead of the current >> > apache + mod-fastcgi or mod-proxy-fcgi approach. "Supported" here means >> > both the primary platform the upstream development focuses on and what >> > the >> > downstream Red Hat product will officially support. >> > >> > How many people are using RGW standalone using the embedded civetweb >> > server instead of apache? In production? At what scale? What >> > version(s) (civetweb first appeared in firefly and we've backported most >> > fixes). >> > >> > Have you seen any problems? Any other feedback? The hope is to >> > (vastly) >> > simplify deployment. >> > >> > Thanks! >> > sage >> > _______________________________________________ >> > ceph-users mailing list >> > ceph-users@xxxxxxxxxxxxxx >> > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >> _______________________________________________ >> ceph-users mailing list >> ceph-users@xxxxxxxxxxxxxx >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > > > -- > С уважением, Фасихов Ирек Нургаязович > Моб.: +79229045757 > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > -- -Pushpesh _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |