Cluster Status

[frantek]

Hi,

as the "current" check_mk plugin for PVE clusters relies on "pvecm" output of PVE v2 (!) I want to do a rewrite of the plugin. I think "pvesh get /cluster/ha/status/current" and "pvesh get /cluster/ha/status/manager_status" may be a good starting point. I found no detailed documentation about the output and the possible states.

What I get is:

root@pve02:~# pvesh get /cluster/ha/status/current
200 OK
[
   {
      "id" : "quorum",
      "node" : "pve02",
      "quorate" : 1,
      "status" : "OK",
      "type" : "quorum"
   },
   {
      "id" : "master",
      "node" : "pve02",
      "status" : "pve02 (active, Sun Oct  8 14:49:28 2017)",
      "timestamp" : 1507466968,
      "type" : "master"
   },
   {
      "id" : "lrm:pve01",
      "node" : "pve01",
      "status" : "pve01 (active, Sun Oct  8 14:49:24 2017)",
      "timestamp" : 1507466964,
      "type" : "lrm"
   },
   {
      "id" : "lrm:pve02",
      "node" : "pve02",
      "status" : "pve02 (idle, Sun Oct  8 14:49:34 2017)",
      "timestamp" : 1507466974,
      "type" : "lrm"
   },
   {
      "id" : "lrm:pve03",
      "node" : "pve03",
      "status" : "pve03 (idle, Sun Oct  8 14:49:34 2017)",
      "timestamp" : 1507466974,
      "type" : "lrm"
   },
   {
      "crm_state" : "started",
      "group" : "MHC",
      "id" : "service:vm:999",
      "max_relocate" : 1,
      "max_restart" : 1,
      "node" : "pve01",
      "request_state" : "started",
      "sid" : "vm:999",
      "state" : "started",
      "status" : "vm:999 (pve01, started)",
      "type" : "service"
   }
]

and

root@pve02:~# pvesh get /cluster/ha/status/manager_status
200 OK
{
   "lrm_status" : {
      "pve01" : {
         "mode" : "active",
         "results" : {
            "PxeLrpo7Xb6giaawAoSeRA" : {
               "exit_code" : 0,
               "sid" : "vm:999",
               "state" : "started"
            }
         },
         "state" : "active",
         "timestamp" : 1507467434
      },
      "pve02" : {
         "mode" : "active",
         "results" : {},
         "state" : "wait_for_agent_lock",
         "timestamp" : 1507467439
      },
      "pve03" : {
         "mode" : "active",
         "results" : {},
         "state" : "wait_for_agent_lock",
         "timestamp" : 1507467439
      }
   },
   "manager_status" : {
      "master_node" : "pve02",
      "node_status" : {
         "pve01" : "online",
         "pve02" : "online",
         "pve03" : "online"
      },
      "service_status" : {
         "vm:999" : {
            "node" : "pve01",
            "running" : 1,
            "state" : "started",
            "uid" : "YYfZMlGCrAkP8dU2KHUVmg"
         }
      },
      "timestamp" : 1507467438
   },
   "quorum" : {
      "node" : "pve02",
      "quorate" : "1"
   }
}

I need detailed information about every tag. Can some one please help?

TIA
Matthias

 

[fabian]

the PVE Admin Guide's HA chapter should explain most/all of the fields you see in the API response. if you still have questions after reading it, feel free to ask concrete questions