On 24/10/2019 16:23, Christopher Wieringa wrote: > Hello all, > > > > I’ve been using the Ceph kernel modules in Ubuntu to load a CephFS > filesystem quite successfully for several months. Yesterday, I went > through a round of updates on my Ubuntu 18.04 machines, which loaded > linux-image-5.0.0-32-generic as the kernel. I’m noticing that while the > kernel boots and the CephFS filesystem is mounted successfully, any > actual file traffic between client/servers is causing the client to hard > lock without any kernel panics seen in the logs. Downgrading to > linux-image-5.0.0-31-generic fixes the issue. > > I’m planning on trying to figure out how to get a crash kernel running > and see if I can generate a log file of some sort, but just wondering if > anyone else mounting Ubuntu 18.04 and CephFS via kernel modules could > replicate this crashing behavior in their environment. I don’t have > experience in setting up crash kernels or even in filing Ubuntu kernel > bugs, so I thought I’d ask this list to see if someone else can > replicate this. > > My Setup: > Ubuntu 18.04.3 LTS amd64 > > Kernel: linux-image-5.0.0-32-generic (installed through the > linux-generic-hwe-18.04 metapackage) > > Number of machines affected/replicated the lockup issues: >100 (three > generations of desktop computers, and multiple virtual machines as well) > Hi Chris, we have noticed the same thing on our VMs using this kernel. The previous kernel 5.0.0-31 was ok, but 5.0.0-32 seems broken when using ceph mounts and using them. Did you report the bug to Canonical yet? I think they should fix their kernel issue, but they do need to know first ;-) Cheers /Simon _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |