You observe a sla_loss alarm on your Exinda and wondering whether these SLAs can be ignored.
You have multiple sites configured in the Exinda and some are triggering the sla_loss alarm. You are seeing a high latency threshold message "Latency threshold exceeded for SLA object 'SLA_SW_Core_<ip address>" and the following errors in the log file:
[statsd.NOTICE]: alarm 'sla_loss_SLA_SW_Core_Server': triggered for rising error for event sla_loss
[statsd.NOTICE]: alarm 'sla_loss_SLA_Wan_Carrier_XXX.XXX.XXX.XXX': triggered for rising error for event sla_loss
SLA objects are configured to monitor the availability of the internet sites/hosts, the sla_loss alarm means that the site/host was unreachable at that time. If you are still getting an sla_loss alarm for only some hosts:
- Ping the host from Exinda for a response.
- If there is no response, fix the connectivity with the help of your network team.
- If the site/host is no longer part of the Exinda network, disable the relevant SLA objects and ignore the errors.
Please refer to the below knowledge base articles for a better understanding of the SLA objects and monitoring of servers.