Hello,

We had a situation where mysql is over 90% full and services are not stopped even though the auto_stop_services_when_full option is enabled. The traffic continued to come to the collector. We ran the command support_must_gather system_db_info, and we got these results for filesystem:

Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/sda2 25671908 5868068 18476736 25% /
devtmpfs 16368372 0 16368372 0% /dev
tmpfs 16380792 28884 16351908 1% /dev/shm
tmpfs 16380792 69548 16311244 1% /run
tmpfs 16380792 0 16380792 0% /sys/fs/cgroup
/dev/sda1 499656 160228 302732 35% /boot
/dev/sda3 272096720 184453988 73797896 72% /var
tmpfs 3276160 0 3276160 0% /run/user/1607
tmpfs 3276160 0 3276160 0% /run/user/994
tmpfs 3276160 0 3276160 0% /run/user/993
tmpfs 3276160 0 3276160 0% /run/user/1600

Does anyone know what could be the reason so the services did not stop?

Best Regards,


------------------------------
Sanela Kovač
------------------------------

Hi Sanela,

Ironically this can happen if the sniffer buffer usage monitor service is not running. This service monitors the DB and disk usage of the appliance. If it is inactive then the appliance and/or DB can grow beyond 90% without stopping all the other services. In v11 this service can be started using the following cli command:
systemctl start guard-snifbufusage.service

If this is not the case here, we would need diags to see what happened on this appliance.

Thanks
Regards

------------------------------
Harold Kubannek
------------------------------