It looks like your mkfs commands are not quite complete. The "failnode" option is deprecated and specifying only one will make that target only use that IP address. So it's timing out because it doesn't have an LNET address available to contact the MGS. It would likely work if you tried to mount from the other server.
Assuming some topology here so it might not be completely accurate, but I think the mkfs commands you want are: mkfs.lustre --fsname=ltfs1 --mgs --mdt --index=0 --servicenode=192.168.19.14@o2ib1 --servicenode=192.168.19.15@o2ib1 --reformat --mkfsoptions "-E stride=32" /dev/disk/by-id/scsi-3600b3420371420b645dde4066c0000a8 mkfs.lustre --fsname=ltfs1 --mdt --index=1 --mgsnode=192.168.19.14@o2ib1 --mgsnode=192.168.19.15@o2ib1 --servicenode=192.168.19.14@o2ib1 --servicenode=192.168.19.15@o2ib1 --reformat --mkfsoptions "-E stride=32" /dev/disk/by-id/scsi-3600b3420371420b645dde5093e0000a9 mkfs.lustre --fsname=ltfs1 --mdt --index=2 --mgsnode=192.168.19.14@o2ib1 --mgsnode=192.168.19.15@o2ib1 --servicenode=192.168.19.15@o2ib1 --servicenode=192.168.19.14@o2ib1 --reformat --mkfsoptions "-E stride=32" /dev/disk/by-id/scsi-3600b3420371420b645dde619060000aa mkfs.lustre --fsname=ltfs1 --mdt --index=3 --mgsnode=192.168.19.14@o2ib1 --mgsnode=192.168.19.15@o2ib1 --servicenode=192.168.19.15@o2ib1 --servicenode=192.168.19.14@o2ib1 --reformat --mkfsoptions "-E stride=32" /dev/disk/by-id/scsi-3600b3420371420b645dde7367f0000ab
_______________________________________________ lustre-discuss mailing list [email protected] http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
