Dear All, I have two OpenVZ nodes running the following:
ploop.x86_64 1.12.1-1 ploop-lib.x86_64 1.12.1-1 vzctl.x86_64 4.7.2-1 vzctl-core.x86_64 4.7.2-1 vzkernel.x86_64 2.6.32-042stab093.4 vzkernel-devel.x86_64 vzquota.x86_64 3.1-1 I'm trying to live migrate a container which runs nfsd. Both nodes are configured in the same way, therefore nfsd module is loaded: # lsmod | grep nfs pio_nfs 17576 0 ploop 116256 32 pio_nfs,pio_direct,pfmt_raw,pfmt_ploop1 nfsd 311145 13 nfs 442468 7 pio_nfs,vzrst,vzcpt lockd 77189 3 vzrst,nfsd,nfs fscache 55684 1 nfs auth_rpcgss 44949 2 nfsd,nfs nfs_acl 2663 2 nfsd,nfs sunrpc 268245 37 pio_nfs,nfsd,nfs,lockd,auth_rpcgss,nfs_acl Whenever I try to live migrate the container, I get the following error: # vzmigrate -r yes --ssh="-p 16501" --keep-dst --online -v destination_host 161 Locked CT Starting live migration of CT 161 to destination_host OpenVZ is running... Checking for CPT version compatibility Checking for CPU flags compatibility Error: Unsupported filesystem nfsd Insufficient CPU capabilities: can't migrate Error: Error: CPU capabilities check failed! Error: Destination node CPU is not compatible Error: Can't continue live migration At that point, the lock file remains and has to be manually deleted. On the other hand, offline migration of that container works fine. The funny thing is that if I try to live migrate the container back to the original host, I get the same error (Unsupported filesystem nfsd / Insufficient CPU capabilities: can't migrate). Shall I open a bug report for this? Regards, Corrado Fiore _______________________________________________ Users mailing list [email protected] https://lists.openvz.org/mailman/listinfo/users
