-
Suggestion
-
Resolution: Unresolved
-
None
-
None
-
19
-
1
-
- mentioned in
-
Page Failed to load
[JRASERVER-68430] Switch Data Center node from UI
We discovered that initiating a full Reindex forces the load balancer to kick you to the other active node, not a great option during high usage times, but an option none the less.
I keep seeing mention of connecting to a specific node in Atlassian's data center documentation, without a way to do it.
As an example, on this page (for Confluence): https://confluence.atlassian.com/conf85/cache-statistics-1283361246.html, it says ...
If you're running Confluence in a cluster, this screen shows the statistics for the node you're currently on.
To view cache statistics for another node in the cluster:
- Go to Administration
> General Configuration > Clustering.
- Select
More options> Cache statistics next to the node you want to view.
You will only be able to view the statistics. To flush a cache or adjust the size, you'll need to access the Cache Management screen on each node directly.
... but then doesn't provide any way to do that. Come on Atlassian, give us the tools we need to administer your data center products.
We hop user nodes by opening a new incognito browser tab/window and hoping for the load balancer to do the rest. It can take a few tries but usually gets the job done. Switching from the UI would definitely help though
It is so common (after upgrades especially) to experience issues with something on a single node. That's a difficult thing to quickly catch when we can't switch nodes from the UI.
If you're using NGINX, you can you do something like this, then you just add "?direct=X" to the end of the URL where X is the node you want to connect to. Its not a great solution as it only works for the one page, as soon as you navigate away it will put you back on your original node. It also doesn't work if you're doing an action like clearing Flushing the Mail Queue, you'll get an XRSF token error. Its not perfect but it can be helpful.
Update your NGINX Config:
upstream jira_cluster { ip_hash; server ipAddress:8080; server ipAddress:8080; server ipAddress:8080; } upstream jira_cluster_1 { ip_hash; server ipAddress:8080; } upstream jira_cluster_2 { ip_hash; server ipAddress:8080; } upstream jira_cluster_3 { ip_hash; server ipAddress:8080; } server { listen 443 ssl; ssl_certificate /etc/nginx/certs/atlassian-20231114.cer; ssl_certificate_key /etc/nginx/certs/atlassian-20231116.key; server_name jira.sierraspace.com; client_max_body_size 300M; location / { add_header Set-Cookie "nginxServer=1" always; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; #proxy_pass http://jira_cluster; proxy_connect_timeout 900; proxy_send_timeout 900; proxy_read_timeout 900; send_timeout 900; proxy_next_upstream error timeout http_502 http_503 http_504; if ($arg_direct = 1) { set $upstream http://jira_cluster_1; } if ($arg_direct = 2) { set $upstream http://jira_cluster_2; } if ($arg_direct = 3) { set $upstream http://jira_cluster_3; } if ($arg_direct !~ "[123]") { set $upstream http://jira_cluster; } proxy_pass $upstream; }