Hello all, hope all is well Is there a dependency or delay that needs to be added to have volumes attached via HBA and managed by multipathd/devmapper and lvm reliably mount on boot?
More detail We have some volumes connected via an hba in a Dell R630 running centos 7u2 When booting up, the volumes do not mount. Causing the server to go into emergency mode, upon login we could see that under /dev/disk/by-uuid/ sym links were not created. At this point we could run from the emergency maintenance shell, partprobe or restart systemd-udev-trigger.service. This would fully create devices under /dev/ and fully populate the links in /dev/disk/by-uuid. Then we run mount -a. and exit the emergency shell and server completes boot multipath is installed and enabled Server is in prod so I can't do further reboot testing till the weekend. How we have it boot now is treating the volumes as x-systemd.automount with noauto/nofail options. After an initial "ls" and/or "cd" the volumes are available and this prevents halts during reboots workaround: UUID=094f2d03-87a1-4c0f-9777-41d11d7e6a7d /data03 xfs noauto,nofail,x-systemd.automount,nobarrier,swalloc,inode64,noatime,nodiratime,logbufs=8,allocsize=16m 0 2 Do I need a _netdev? Apologies not sure with systemd at the helm thanks for any/all help
_______________________________________________ systemd-devel mailing list systemd-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/systemd-devel
