Skip to main content
Cl0ck
Cl0ck
Silver ⭐️⭐️⭐️
December 27, 2019
Solved

How to recover Cloudera Manager after a network migration? Hosts file updated, but still tries to connect to old database URL and fails, preventing webserver from starting up

As the title explains, we have moved our cluster from `taco.net` to `dev.taco.net`, and in the process this has made our Cloudera Manager server unable to start up because it continues to try to connect to database.taco.net instead of database.dev.taco.net. Hosts files are updated on all servers, network interfaces are configured to their new IPs.

I would imagine there is something I should update on the Cloudera Manager server, but I'm not sure where to look to update the URL. 

EricL
EricLAnswer
Platinum ⭐️⭐️
December 30, 2019
@Cl0ck

Have you updated CM's DB configuration file under /etc/cloudera-scm-server/db.properties?

You need to change the value for com.cloudera.cmf.db.host from taco.net to dev.taco.net, and see if that helps.

Cheers
Eric
View in thread

3 replies

EricL
Platinum ⭐️⭐️
EricLAnswer
Platinum ⭐️⭐️
December 30, 2019
@Cl0ck

Have you updated CM's DB configuration file under /etc/cloudera-scm-server/db.properties?

You need to change the value for com.cloudera.cmf.db.host from taco.net to dev.taco.net, and see if that helps.

Cheers
Eric
Translate
Cl0ck
Cl0ckAuthor
Silver ⭐️⭐️⭐️
January 4, 2020

Thanks Eric! That is indeed the solution I was looking for. I ended up figuring it out myself a week or so ago before I saw your comment here just now. The cool thing was that after getting CM up and running, it was able to eventually update the new FQDNs for all of the clustered hosts on it's own (even though initially it was still displaying the old hostnames from before our migration)!

Translate
EricL
Platinum ⭐️⭐️
EricL
Platinum ⭐️⭐️
January 6, 2020
@Cl0ck

Glad that it is all resolved.

cheers
Translate
Terms & ConditionsAccessibility statement