One or more Bitbucket Data Center nodes does not start successfully
プラットフォームについて: Data Center のみ - この記事は、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。
この KB は Data Center バージョンの製品用に作成されています。Data Center 固有ではない機能の Data Center KB は、製品のサーバー バージョンでも動作する可能性はありますが、テストは行われていません。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
After restarting one of the Bitbucket server nodes, the application does not come up for that node.
環境
- Bitbucket Data Center
- Multiple cluster nodes are in use
診断
There are no messages being added to the application log file from the missing node.
Checking the logs from an up and running node, we see the following:
2021-04-15 15:02:52,826 WARN [hz.hazelcast.cached.thread-15] c.h.n.t.TcpIpConnectionErrorHandler [10.150.90.150]:5701 [stash-cluster] [3.11.1] Removing connection to endpoint [10.150.90.149]:5701 Cause => java.net.SocketException {Connection refused to address /10.150.90.149:5701}, Error-Count: 5
2021-04-15 15:02:52,837 WARN [hz.hazelcast.cached.thread-15] c.h.i.cluster.impl.MembershipManager [10.150.90.150]:5701 [stash-cluster] [3.11.1] Member [10.150.90.149]:5701 - 25831d29-2a10-42bd-94d8-3f778di8d645 is suspected to be dead for reason: No connection
2021-04-15 15:02:52,925 INFO [hz.hazelcast.event-3] c.a.s.i.c.HazelcastClusterService Node '/10.150.90.149:5701 (stash1)' was REMOVED from the cluster. Updated cluster:
[/10.150.90.150:5701 master this name='stash2' uuid='208d5fac-5ff4-452b-84ce-1f39c32e5b3l' vm-id='3e936bd8-dc81-4218-82c2-2061fe50b41f']
The problematic node 10.150.90.149 was removed from the cluster as seen by the message above. An important part of the message is Connection refused to address /10.150.90.149:5701}, Error-Count: 5. Hazelcast performs a heartbeat check 5 times and if there is no response, the unresponsive node is ejected from the cluster.
原因
Usually, this message is a consequence of one of the following behaviors:
- Network related issues.
- Automation tools are not properly configured to run Bitbucket.
ソリューション
ソリューション 1
Check the network connectivity between nodes, specifically on port 5701. You can use telnet or ping to check if the nodes are communicating with each other.
ソリューション 2
Check the automaton tools configurations used to start Bitbucket.