One of the nice things about cloud environments that use K8s (among other things) is that they can restart services automatically, even in the event of a complete pod hardware failure. Many folks rely on this capability in lieu of broker-specific HA. If this suits your needs, then that's what I recommend. This keeps the broker configuration & management extremely simple.
If that doesn't suit your needs then the next best approach I'd recommend is 2 brokers using shared storage (e.g. on a persistent volume claim) HA configuration. After that, 2 brokers using replication with ZooKeeper would be my recommendation. For what it's worth, I brought this up previously because you said, "We want to configure Artemis for HA using 'replication' strategy." I'm not sure what you mean by a "triple-replica-set." I'm currently working on a big documentation update that will hopefully make a lot of this stuff more clear. Justin On Sun, Jan 7, 2024 at 5:55 PM John Lilley <john.lil...@redpointglobal.com.invalid> wrote: > Sorry, one more thing, you mentioned “single pair of brokers with > replication”… does that imply that a triple-replica-set would be immune > from the split-brain issue? > > > > john > > > > > [image: rg] <https://www.redpointglobal.com/> > > John Lilley > > Data Management Chief Architect, Redpoint Global Inc. > > 34 Washington Street, Suite 205 Wellesley Hills, MA 02481 > > *M: *+1 7209385761 <+1%207209385761> | john.lil...@redpointglobal.com > > *From:* John Lilley > *Sent:* Sunday, January 7, 2024 4:11 PM > *To:* users@activemq.apache.org > *Subject:* RE: Looking for HA/replication boilerplate broker.xml and > advice > > > > Justin, > > > > Regarding your zookeeper advice we are on Kubernetes -- is there further > configuration doc/advice/example when we add K8S into the mix? This is > only Azure AKS for now. > > > > And we are not wedded to the replication strategy. I while back I * > *thought** I read it was preferred to shared-storage, but perhaps I’ve > misremembered. Is there a “recommended best practice” for Artemis HA on > K8S? We’re not after the very best performance, our goals are more like, > ease of configuration and management combined with resilience. > > > > Thanks > > john > > > > *From:* Justin Bertram <jbert...@apache.org> > *Sent:* Friday, December 22, 2023 1:38 PM > *To:* users@activemq.apache.org > *Subject:* Re: Looking for HA/replication boilerplate broker.xml and > advice > > > > **** [Caution] This email is from an external source. Please use caution > responding, opening attachments or clicking embedded links. **** > > > > Which configuration is "better" really depends on your use-case. Failback > is a nice feature when you have, for example, invested more money into the > hardware of your primary broker since failback ensures the primary will > take over as soon as it comes back online. > > > > Keep in mind that running a single pair of brokers with replication is not > recommended due to the risk of split brain. The recommended approach to > mitigate split brain is by using ZooKeeper. See the > "zookeeper-single-pair-failback" example for more details on that. > > > > > > Justin > > > > On Fri, Dec 22, 2023 at 2:13 PM John Lilley < > john.lil...@redpointglobal.com.invalid> wrote: > > Thanks again for the help! > > > FYI I had to add this line to the master’s broker.xml from the example you > mentioned: > > > > <check-for-live-server>true</check-for-live-server> > > > > Otherwise, a failed master that came back would start to fight with the > still-active slave. > > > > I think I could have also told the slave to “fail-back”? Is there a > recommendation for which is better? > > > > John > > > > > > [image: rg] > <https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.redpointglobal.com%2f&c=E,1,5XwtFHoe_7j3xickhrC1SclF26aQUzGQ0J29kY8MWbiROJpJYHoK1iBBFbGN7Hq7nM9QaXP7xw27D-CApgn18_NYHaqXexG91GvTMH7omSBqy6Zaxg,,&typo=1> > > *John Lilley * > > *Data Management Chief Architect, Redpoint Global Inc. * > > 34 Washington Street, Suite 205 Wellesley Hills, MA 02481 > > *M: *+1 7209385761 <+1%207209385761> | john.lil...@redpointglobal.com > > *From:* Justin Bertram <jbert...@apache.org> > *Sent:* Thursday, December 21, 2023 3:42 PM > *To:* users@activemq.apache.org > *Subject:* Re: Looking for HA/replication boilerplate broker.xml and > advice > > > > **** [Caution] This email is from an external source. Please use caution > responding, opening attachments or clicking embedded links. **** > > > > > ...is it possible to run both master and slave on the same host, for > dev/test? > > > > Yes. > > > > > Like… configure them with different ports? > > > > Yes. This is what the examples do which I mentioned in my previous email. > > > > You can also use the --port-offset switch with the create command. > > > > > > Justin > > > > On Thu, Dec 21, 2023 at 4:36 PM John Lilley < > john.lil...@redpointglobal.com.invalid> wrote: > > Follow up question: is it possible to run both master and slave on the > same host, for dev/test? > > Like… configure them with different ports? > > Thanks > > John > > > > [image: rg] > <https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fwww.redpointglobal.com%2f&c=E,1,d_k8uFQLvdcbwVXF9dBwVRuLOnnnAIAoM3Jv3uUG5zr_o_vvFSzAuxMjv0GmhSyd2QND75uDoO8klh83_xBW-O68X2u6v9hYK06kHtBxXML-5OTwo3xNxo8,&typo=1> > > *John Lilley * > > *Data Management Chief Architect, Redpoint Global Inc. * > > 34 Washington Street, Suite 205 Wellesley Hills, MA 02481 > > *M: *+1 7209385761 <+1%207209385761> | john.lil...@redpointglobal.com > > > PLEASE NOTE: This e-mail from Redpoint Global Inc. (“Redpoint”) is > confidential and is intended solely for the use of the individual(s) to > whom it is addressed. If you believe you received this e-mail in error, > please notify the sender immediately, delete the e-mail from your computer > and do not copy, print or disclose it to anyone else. If you properly > received this e-mail as a customer, partner or vendor of Redpoint, you > should maintain its contents in confidence subject to the terms and > conditions of your agreement(s) with Redpoint. > > > PLEASE NOTE: This e-mail from Redpoint Global Inc. (“Redpoint”) is > confidential and is intended solely for the use of the individual(s) to > whom it is addressed. If you believe you received this e-mail in error, > please notify the sender immediately, delete the e-mail from your computer > and do not copy, print or disclose it to anyone else. If you properly > received this e-mail as a customer, partner or vendor of Redpoint, you > should maintain its contents in confidence subject to the terms and > conditions of your agreement(s) with Redpoint. > > > PLEASE NOTE: This e-mail from Redpoint Global Inc. (“Redpoint”) is > confidential and is intended solely for the use of the individual(s) to > whom it is addressed. If you believe you received this e-mail in error, > please notify the sender immediately, delete the e-mail from your computer > and do not copy, print or disclose it to anyone else. If you properly > received this e-mail as a customer, partner or vendor of Redpoint, you > should maintain its contents in confidence subject to the terms and > conditions of your agreement(s) with Redpoint. >
