Hi Everyone,
I upgraded to 17.17 from 17.16 on my wazo cluster after following all directions in the wazo documentation on the process. Everything went fine on both hosts but about 20 minutes after both upgrades completed, a handful of wazo services become disabled and asterisk is NOT running on the secondary HA host. The odd thing is that I can run wazo-service enable, and all services start and all seems well, but out of the blue asterisk closes cleanly and all wazo services are stopped and then disabled except for the following services: wazo-admin-ui, wazo-plugind, wazo-webhookd, xivo-sysconfd, xivo-confgend, xivo-confd, wazo-auth.
I can re-enable all services with wazo-service enable, and then Im able to start all services successfully but after about 5 minutes of running without errors or anything looking odd or wrong, the following services stop with no errors or reported troubles; in-fact Asterisk exists cleanly and just stops its service along with the following services: wazo-call-logd, xivo-dxtora, xivo-provd, xivo-agid, asterisk, xivo-amid, xivo-agentd, xivo-ctid, xivo-dird, xivo-dird-phoned, xivo-ctid-ng, and xivo-websocketd. All other services stay online during this weird state...
wth
I cannot find an error/warning or anything anywhere that would be insightful on why the above is happening only on my passive node. My primary node is running without issues and does NOT exhibit the same behavior. Also, just an FYI but we also rebuilt the second node with a fresh image and joined to the cluster again but all ending with same result as above. If I turn off HA functionality on node 2 all services start normally and never shutdown. Once HA is reenabled though within a few mins, the same happens as noted above.
I really HOPE someone else has seen this behavior and knows the fix or at-least a direction to be pointed would be GREATLY appreciated!!
Thank you all in advance for any assistance! This is preventing us from deploying to a production system so anything would be wonderful!
I upgraded to 17.17 from 17.16 on my wazo cluster after following all directions in the wazo documentation on the process. Everything went fine on both hosts but about 20 minutes after both upgrades completed, a handful of wazo services become disabled and asterisk is NOT running on the secondary HA host. The odd thing is that I can run wazo-service enable, and all services start and all seems well, but out of the blue asterisk closes cleanly and all wazo services are stopped and then disabled except for the following services: wazo-admin-ui, wazo-plugind, wazo-webhookd, xivo-sysconfd, xivo-confgend, xivo-confd, wazo-auth.
I can re-enable all services with wazo-service enable, and then Im able to start all services successfully but after about 5 minutes of running without errors or anything looking odd or wrong, the following services stop with no errors or reported troubles; in-fact Asterisk exists cleanly and just stops its service along with the following services: wazo-call-logd, xivo-dxtora, xivo-provd, xivo-agid, asterisk, xivo-amid, xivo-agentd, xivo-ctid, xivo-dird, xivo-dird-phoned, xivo-ctid-ng, and xivo-websocketd. All other services stay online during this weird state...
wthI cannot find an error/warning or anything anywhere that would be insightful on why the above is happening only on my passive node. My primary node is running without issues and does NOT exhibit the same behavior. Also, just an FYI but we also rebuilt the second node with a fresh image and joined to the cluster again but all ending with same result as above. If I turn off HA functionality on node 2 all services start normally and never shutdown. Once HA is reenabled though within a few mins, the same happens as noted above.
I really HOPE someone else has seen this behavior and knows the fix or at-least a direction to be pointed would be GREATLY appreciated!!
Thank you all in advance for any assistance! This is preventing us from deploying to a production system so anything would be wonderful!
Phone System 