Enable HA for mapreduce jobhistory server

Question

Hi Team,

From past few days I am looking for enabling HA for mapreduce(MRv2) jobhistory server .

I could get few docs ,but here my question how to install mapreduce package on multiple nodes as per this document prerequisties

"https://support.datafabric.hpe.com/s/article/Use-HA-proxy-for-accessing-Job-History-server-in-YARN?language=en_US"

Please let me know is there any other way I can enable HA

Comments

Hi@Shllpa,

Are you able to configure HA proxy in your system? I mean not for MR. How HA proxy works?

---

Hi ,

 No HA proxy I am nt able to bec that links talk about MapR env .Our deployment is open source apache hadoop 2.6.5.

Would like to know is that possible by setting HA proxy .if so any pointers to install ?Please share

---

Hi,

If you are trying to balance the load of Jobhistory server, then just mention the IP addresses in HA proxy configuration file.

---

Sory ,its like one server is down ,another to pick up .not exactly load balancing .FIrst would like to know how to set HA proxy for mapreduce .Please share documents or any pointers.Thanks for your timely help and support

---

Hi,

Check the below document. In this document, you will get an idea of how to configure HA proxy for a server.

https://www.hadoopandcloud.com/hadoop/configure-proxy-for-hiveserver2impala/

---

Thank you so much :)  .But here some more query sorry for taking your time .

Installing mapreduce daemons alone is not possible right ? especially from ambari UI 2.6.5  or can we ??

---

Hi@Shllpa,

I think it is not possible. You need to configure first. Did you try the above solution?