APMIA - - Release 24.4.2.8 ( 990008 ) - Query. We had a scenario where few linux servers were rebooted and the last running process came up as sysedge. Multiple vendors are analyzing on the root cause of the restart. In parallel, can broadcom provide some insight into this as well ?
[<id>@<host>~]$ uname -a
Linux <host> 4.18.0-513.18.1.el8_9.x86_64 #1 SMP 2024 x86_64 x86_64 x86_64 GNU/Linux
The metric screenshot shows a data gap during the server reboot, also agent connection status indicates agent was disconnected at same time. This is expected due to the server reboot.
Seems from the other vendor's analysis, it’s a kernel crash and most likely a known issue. Further analysis will be pursued.
From APM side, it’s nearly impossible for our app to cause a kernel crash like this.
Third-party analysis: This is likely a known issue with vxfs, as documented with . A mishandling of reference counts caused a block_device to be freed when it should have still been in use and open. Confirming the version of vxfs which should be used with the vendor would be recommended.