JIRA encounters error on startup (new install): "ERROR [c.a.j.appconsistency.db.PostgresSchemaConfigCheck] The schema name 'null' in your dbconfig.xml file contains upper case characters and JIRA only supports lower case schemas in POSTGRES."

[no_longer_in_sudoers_file]

This morning, while working on an internal JIRA instance, I encountered the following error:

ERROR [c.a.j.appconsistency.db.PostgresSchemaConfigCheck] The schema name 'null' in your dbconfig.xml file contains upper case characters and JIRA only supports lower case schemas in POSTGRES.

Searching for an answer on this, I found that no one really had documented this error message and I figured I would create something so that others (and my future self) don't have to approach this from a blank-slate.

This error occurred on JIRA 7.0 Service Desk 3.xx with a dbconfig.xml file generated by config.sh.

The dbconfig.xml (with some information redacted) appears as follows:

<?xml version="1.0" encoding="UTF-8"?>
<jira-database-config>
  <name>defaultDS</name>
  <delegator-name>default</delegator-name>
  <database-type>postgres72</database-type>
  <jdbc-datasource>
    <url>jdbc:postgresql://***REDACTED***/***REDACTED***</url>
    <driver-class>org.postgresql.Driver</driver-class>
    <username>***REDACTED***</username>
    <password>***REDACTED***</password>
    <pool-min-size>***REDACTED***</pool-min-size>
    <pool-max-size>***REDACTED***</pool-max-size>
    <pool-max-wait>***REDACTED***</pool-max-wait>
    <time-between-eviction-runs-millis>***REDACTED***</time-between-eviction-runs-millis>
    <pool-max-idle>***REDACTED***</pool-max-idle>
    <pool-prepared-statements>***REDACTED***</pool-prepared-statements>
    <pool-remove-abandoned>***REDACTED***</pool-remove-abandoned>
    <pool-remove-abandoned-timeout>***REDACTED***</pool-remove-abandoned-timeout>
    <pool-test-while-idle>***REDACTED***</pool-test-while-idle>
    <validation-query-timeout>***REDACTED***</validation-query-timeout>
  </jdbc-datasource>
</jira-database-config>

[no_longer_in_sudoers_file · Accepted solution]

(1) This error appears to be a trivial issue.  Not specifying a schema appears to have no impact on JIRA starting as NULL schema just reverts to the default (e.g. public) schema...at least in my installation.  Nonetheless, I don't like errors in my logs. 

(2) I manually added "<schema-name>public</schema-name>" to the dbconfig.xml to fix this issue.  This fixes my issue.

(3) But what if you wanted to connect to a different schema?  Well, first you could list all of the available schemas.  The following SQL statement will tell you what schemas are available (in postgresql only):

sam_test-> select schema_name
sam-test-> from information_schema.schemata;
    schema_name     
--------------------
 public
 pg_catalog
 information_schema
 global_thermonuclear_war
 a_thing_that_should_work
(5 rows)

Hopefully that helps someone else, but what about the long term?  Why doesn't config.sh do this?

Researching a bit further, I found that @Nathan Peters already identified this issue (See [Jira config.sh creates invalid database settings file]).  I'm not sure why I didn't find this earlier in my morning (not enough coffee, perhaps).  Nathan raises a valid point in my humble opinion.  While this is a trivial error with a known fix, I have submitted a bug report (https://jira.atlassian.com/browse/JRA-47663)  this morning.

[Gennadiy]

This issue helped me to resolve own problem during start Jira.

When Jira was starting, I got warning message   “The ‘schema-name’ property in your dbconfig.xml file contain upper case letters”.  Actually, my Jira instance dbconfig.xml file do not contain < schema-name> section too and I could not guess why I see this warning. This configuration xml file created in 7.0.11 version and 7.2.13, 7.8.2 .bin installer did not add < schema-name> section while application update.

Line was added manually, as Sam suggested above, and warning message disappeared.