[ClusterLabs] IPaddr2 clone deprecated
Hi guys, im trying to create a IPaddr2 cloned resource for one of my project. i need some kind of simple but effective loadbalancer for my rabbitmq cluster managed by pacemaker. My current SO is Almalinux 8.6. seems IPaddr2 coned resource is not working / supported anymore, probably because CLUSTERIP is not viable in the latest releases. if i remeber well on centos7 no issues related. could u explain me a little bit more ? below the errors i got: ClusterIP_start_0 on pcsnode1 'unknown error' (1): call=40, status=complete, exitreason='iptables failed', ERROR: iptables failed Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: Result of start operation for rabbitmq-vip on rabbitvm1: error (iptables failed) Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: rabbitmq-vip_start_0@rabbitvm1 output [ iptables v1.8.4 (nf_tables): chain name not allowed to start with `-'\n\nTry `iptables -h' or 'iptables --help' for more information.\nocf-exit-reason:iptables failed\n ] ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
Re: [ClusterLabs] IPaddr2 clone deprecated
Hi Damiano, The clusterip functionality was removed from the kernel, and we havent currently been able to find a way to use the suggested replacement to perform the same kind of logic: https://wiki.nftables.org/wiki-nftables/index.php/Supported_features_compared_to_xtables#cluster Oyvind On 20/10/23 11:49 +0200, Damiano Giuliani wrote: Hi guys, im trying to create a IPaddr2 cloned resource for one of my project. i need some kind of simple but effective loadbalancer for my rabbitmq cluster managed by pacemaker. My current SO is Almalinux 8.6. seems IPaddr2 coned resource is not working / supported anymore, probably because CLUSTERIP is not viable in the latest releases. if i remeber well on centos7 no issues related. could u explain me a little bit more ? below the errors i got: ClusterIP_start_0 on pcsnode1 'unknown error' (1): call=40, status=complete, exitreason='iptables failed', ERROR: iptables failed Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: Result of start operation for rabbitmq-vip on rabbitvm1: error (iptables failed) Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: rabbitmq-vip_start_0@rabbitvm1 output [ iptables v1.8.4 (nf_tables): chain name not allowed to start with `-'\n\nTry `iptables -h' or 'iptables --help' for more information.\nocf-exit-reason:iptables failed\n ] ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/ ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/
Re: [ClusterLabs] IPaddr2 clone deprecated
Thanks for clarify it. so right now there is now way have a simple LB for active/active resource on Pacemaker :/ Il giorno ven 20 ott 2023 alle ore 11:59 Oyvind Albrigtsen < oalbr...@redhat.com> ha scritto: > Hi Damiano, > > The clusterip functionality was removed from the kernel, and we havent > currently been able to find a way to use the suggested replacement to > perform the same kind of logic: > > https://wiki.nftables.org/wiki-nftables/index.php/Supported_features_compared_to_xtables#cluster > > > Oyvind > > On 20/10/23 11:49 +0200, Damiano Giuliani wrote: > >Hi guys, > > > >im trying to create a IPaddr2 cloned resource for one of my project. > >i need some kind of simple but effective loadbalancer for my rabbitmq > >cluster managed by pacemaker. > >My current SO is Almalinux 8.6. > >seems IPaddr2 coned resource is not working / supported anymore, probably > >because CLUSTERIP is not viable in the latest releases. > >if i remeber well on centos7 no issues related. > > > >could u explain me a little bit more ? > > > >below the errors i got: > > > >ClusterIP_start_0 on pcsnode1 'unknown error' (1): call=40, > >status=complete, exitreason='iptables failed', > > > >ERROR: iptables failed > >Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: Result of > >start operation for rabbitmq-vip on rabbitvm1: error (iptables failed) > >Oct 18 10:18:18 rabbitvm1 pacemaker-controld[32192]: notice: > >rabbitmq-vip_start_0@rabbitvm1 output [ iptables v1.8.4 (nf_tables): > >chain name not allowed to start with `-'\n\nTry `iptables -h' or > >'iptables --help' for more information.\nocf-exit-reason:iptables > >failed\n ] > > >___ > >Manage your subscription: > >https://lists.clusterlabs.org/mailman/listinfo/users > > > >ClusterLabs home: https://www.clusterlabs.org/ > > ___ > Manage your subscription: > https://lists.clusterlabs.org/mailman/listinfo/users > > ClusterLabs home: https://www.clusterlabs.org/ > ___ Manage your subscription: https://lists.clusterlabs.org/mailman/listinfo/users ClusterLabs home: https://www.clusterlabs.org/