https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=250770

            Bug ID: 250770
           Summary: AWS EC2 system freezes up possibly associated with NFS
                    (EFS)
           Product: Base System
           Version: 12.2-RELEASE
          Hardware: Any
                OS: Any
            Status: New
          Severity: Affects Some People
          Priority: ---
         Component: kern
          Assignee: b...@freebsd.org
          Reporter: r...@gusw.net

I use light-weight small cloneable FreeBSD systems built from the basic AWS EC2
AMI, first with 12.0-RELEASE and just now re-built with 12.2-RELEASE version.
The idea is to have a minimal system rootfs while the bulk of the system disk,
the /usr tree, is mounted from EFS (NFSv4). The AMI out of the box for 12.0
required 8 GB system disk, and now for 12.2 it's 10 GB, if I can do it in 1 GB
I have some significant cost savings in redundant EBS volumes while also an
easier time to distribute application software updates to my worker and server
farm. I have described the principle here:
https://forums.freebsd.org/threads/amazon-aws-ec2-super-light-configuration.69084/

The problem is that no matter what I do, the system locks up after about 4
hours. And it's not a kernel panic, it's not associated with any error, and I
cannot see it at all. Nothing on the console screenshot, nothing in the logs
and nothing in the /var/log/messages upon force-stop & re-start. 

Moreover, AWS EC2 console thinks the system is up in its status check. But any
attempt to do anything results in hangs. SSH connection hangs. telnet to port
22, no response at all (not even connection reset, etc.) just hangs. If I
happen to be logged in on the system whatever shell command I am doing will
just hang, unable to CTRL-C break out nor CTRL-Z to the background. Just hang.

So I have nothing to show you. But I can provide anyone who wants to see it
with a setup and would appreciate if someone would help out here. 

Here is my /etc/fstab:

# Custom /etc/fstab for FreeBSD VM images
/dev/gpt/rootfs  /        ufs     rw      1       1
/dev/gpt/varfs   /var     ufs     rw      1       1
fdesc            /dev/fd  fdescfs rw      0       0
proc             /proc    procfs  rw      0       0
/usr             /.usr    nullfs  rw      0       0
fs-xxxxxx.efs.us-xxxxx-x.amazonaws.com:/ /usr nfs
rw,nfsv4,minorversion=1,oneopenown,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport,late,bg
0 0

It is freezing up within around 4 hours, even just idling around doing nothing,
and it doesn't seem like heavy work activity would accelerate the lock-up
situation. Also doesn't depend on which instance type, specifically I have
tried t3.nano, t3.micro, c5.large, m5a.large, it happens on all of them about
the same time.

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
freebsd-bugs@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "freebsd-bugs-unsubscr...@freebsd.org"

Reply via email to