Hi all,
I've had a system running for a year or so now, it's sitting on ESXi as a VM with several jails inside of a main system. There is virtual networking where I port forward from the main 'host' system to the internal jails for SSH and other services. I run a server on one of the jails, apache on another, etc
Twice within the past month, the system has gone in to a bizarre state. SSH quits responding on the main host, but continues to work on the jails. However, inside the jails, 'sleep 1' will block forever. My server runs in a timed loop with nanosleep(..), and it blocks here. Rebooting the VM fixes it, can't find any other solution
I was on FreeBSD 8.2. After this happened once, I upgraded to 8.3. This problem never happened until about a month ago - the VM server admin/owner is my friend, the server is in his basement, he hasn't upgraded ESXi, he hasn't touched the configuration of anything, and he hasn't changed the hardware at all. According to him, nothing has been touched
There's no sign of foul play. The 'host' system only runs sshd. The jails have no network path to each other or back to the main system. It would seem that there's an issue with the timer/clock hardware, but I can't find a reason for it. I am considering upgrading to FreeBSD 9.x, but I am running the latest 8.3 patches and it seems that I should be fine (especially since I was fine for so long)..
Thoughts on this would be greatly appreciated!
I've had a system running for a year or so now, it's sitting on ESXi as a VM with several jails inside of a main system. There is virtual networking where I port forward from the main 'host' system to the internal jails for SSH and other services. I run a server on one of the jails, apache on another, etc
Twice within the past month, the system has gone in to a bizarre state. SSH quits responding on the main host, but continues to work on the jails. However, inside the jails, 'sleep 1' will block forever. My server runs in a timed loop with nanosleep(..), and it blocks here. Rebooting the VM fixes it, can't find any other solution
I was on FreeBSD 8.2. After this happened once, I upgraded to 8.3. This problem never happened until about a month ago - the VM server admin/owner is my friend, the server is in his basement, he hasn't upgraded ESXi, he hasn't touched the configuration of anything, and he hasn't changed the hardware at all. According to him, nothing has been touched
There's no sign of foul play. The 'host' system only runs sshd. The jails have no network path to each other or back to the main system. It would seem that there's an issue with the timer/clock hardware, but I can't find a reason for it. I am considering upgrading to FreeBSD 9.x, but I am running the latest 8.3 patches and it seems that I should be fine (especially since I was fine for so long)..
Thoughts on this would be greatly appreciated!
