Search results

do the zpool import -N -d /dev/disk/by-id/ trick to import the pool, then`exit` to continue the boot, then `update-grub` or `update-initramfs -u` (I don't remember, do both :-) after to get the /etc/zfs/zpool.cache in the boot image.

try a `update-grub` or `update-initramfs -u` (I don't remember, do both :-) after your zpool add to get the /etc/zfs/zpool.cache in the boot image

I see how to `pvecm delnode remoteIP` but I don't see how on remoteIP to add it to another cluster with another corosync cluster_name. It can zero everything in /etc/pve as there are no guests. Tx.

I saw the same with early 5.0, but it was fixed with a later 5.0 via pveceph purge ; apt-get update ; apt-get upgrade

Well, that doesn't work so well when all the OSD's journals are on a common SSD :) Thanks for listening.

Why can't a zvol be a OSD disk? root@nas:/etc/pve/rob# pveceph createosd /dev/zvol/z/ceph unable to get device info for 'zd16' root@nas:/etc/pve/rob# mkfs.xfs /dev/zvol/z/ceph specified blocksize 4096 is less than device physical sector size 8192 switching to logical sector size 512...

yup, it was that change, had I correctly upgraded to stretch on 2 of the 3 nodes that change wouldn't have been held back. Once upgraded and that patch applied this works as designed.

When one https to a node its syslog shows every 5sec Argument "U" isn't numeric in addition (+) at /usr/share/perl5/PVE/API2Tools.pm looking at the journal update /var/lib/rrdcached/db/pve2-vm/101 1490450929:2:U:4294967296:U:68719476736:0:U:U:U:U time between nodes is exactly the same...