Upgrade failed: Table 'LABEL' has a multi-column primary key on [ID, LABELID]
プラットフォームについて: Server および Data Center のみ。この記事は、Server および Data Center プラットフォームのアトラシアン製品にのみ適用されます。
サーバー*製品のサポートは 2024 年 2 月 15 日に終了しました。サーバー製品を利用している場合は、アトラシアンのサーバー製品のサポート終了のお知らせページにて移行オプションをご確認ください。
*Fisheye および Crucible は除く
要約
When upgrading to Confluence 6.13.8, upgrade fails with the following ERROR:
2020-11-20 11:02:56,501 ERROR [localhost-startStop-1] [atlassian.confluence.upgrade.UpgradeLauncherServletContextListener] contextInitialized Upgrade failed, application will not start: Upgrade task com.atlassian.confluence.upgrade.upgradetask.LowerCaseUsernameReferencesUpgradeTask@20ce74e2 failed during the SCHEMA_UPGRADE phase due to: Table 'LABEL' has a multi-column primary key on [ID, LABELID]
com.atlassian.confluence.upgrade.UpgradeException: Upgrade task com.atlassian.confluence.upgrade.upgradetask.LowerCaseUsernameReferencesUpgradeTask@20ce74e2 failed during the SCHEMA_UPGRADE phase due to: Table 'LABEL' has a multi-column primary key on [ID, LABELID]When inspecting the table directly there is no extra 'ID' column, only the "LABELID, NAME, OWNER, NAMESPACE, CREATIONDATE, LASTMODDATE" columns which are expected.
環境
Confluence 5.0.2
Oracle データベース
原因
There was another Confluence schema in the DB in conflict
回避策
Create a view that restricts to just the constraints this user should see:
CREATE VIEW ALL_CONSTRAINTS AS SELECT * FROM SYS.ALL_CONSTRAINTS WHERE OWNER = UPPER('TSPACE_USER');This allows the upgrade to use the view instead of the table, and get past the issue.
a similar workaround may be applied where a view is created with name 'LABEL' for the 'LABEL' table, consisting of only the correct columns that exist in the correct target table:
CREATE VIEW schema.LABEL (LABELID, NAME, OWNER, NAMESPACE, CREATIONDATE, LASTMODDATE) AS SELECT LABELID, NAME, OWNER, NAMESPACE, CREATIONDATE, LASTMODDATE FROM schema.LABEL;