in Monitor servers - Operations Manual "Table 10.13. Measured values, accessed via the status endpoint"
How do we access these checks? In particular my goal is to monitor the health of the cluster from a single URL if possible. In the table above there is allServersAreHealthy which looks like something I'd be interested in.
Version: neo4j-enterprise 3.5.2 Ubuntu 16.04
Thanks in advance!
What is your actual question?
Those endpoints can be retrieved with any http client including ones from load balancers.
Via the configured http(s) port of your server(s)
Apologies - how do I actually get at the value of allServersAreHealthy ? I've been hunting around with URLs and cannot seem to find it. Thank you
What full URL have you been using?
curl 10.211.162.48:7474/db/manage/server/causalclustering/status
and I get back:
{"lastAppliedRaftIndex":-1,"votingMembers":["45afaec8-005f-48a4-95c3-bf343a02f1b9","a829b2e3-cde0-40bf-8eab-9474ef8951a8","e8a1aa2b-41f0-4f57-a740-7dc1cc9fe2e1"],"memberId":"45afaec8-005f-48a4-95c3-bf343a02f1b9","leader":"a829b2e3-cde0-40bf-8eab-9474ef8951a8","millisSinceLastLeaderMessage":1002,"participatingInRaftGroup":true,"core":true,"healthy":true}
Where do the checks in table 10.13 show up?
Hi Mark,
You will not find them in the response of the status endpoint, or elsewhere.
If you take attention to the paragraph just above the table 10.13, you will see that this table is a list of verification (check) to do if you want to do a rolling-upgrade.
So for your use-case, you need to call the status endpoint of every node in the cluster.
Cheers
Got it. Thank you for your help!
One more thing - what is the purpose of that table at all, if those values "accessed via the status endpoint" cannot actually be accessed?
Like I said it's a list of verification you have to do to make a rolling release.
Those values can be computed from the different status endpoints, and that's why this table has a Method of calculation column (to know how to compute them)