[LU-3085] After upgrade from 1.8.9 to 2.4, can not mount second MDT.

• Type: Bug
• Resolution: Not a Bug
• Priority: Minor
• Affects Version/s: Lustre 2.4.0
• Labels: dne
• Environment: Before upgrade server and client: 1.8.9 rhel6; After upgrade server and client: lustre-master build#1354
• Severity: 3
• Assignee: WC Triage
• Reporter: Sarah Liu

Description

[root@client-18 ~]# mkfs.lustre --fsname=lustre  --mdt --reformat --mgsnode=fat-amd-1@tcp --index=1 --device-size=10000000 --backfstype=ldiskfs /dev/sda3

   Permanent disk data:
Target:     lustre:MDT0001
Index:      1
Lustre FS:  lustre
Mount type: ldiskfs
Flags:      0x61
              (MDT first_time update )
Persistent mount opts: user_xattr,errors=remount-ro
Parameters: mgsnode=10.10.4.132@tcp

device size = 36256MB
formatting backing filesystem ldiskfs on /dev/sda3
	target name  lustre:MDT0001
	4k blocks     2500000
	options        -J size=388 -I 512 -i 2048 -q -O dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F
mkfs_cmd = mke2fs -j -b 4096 -L lustre:MDT0001  -J size=388 -I 512 -i 2048 -q -O dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F /dev/sda3 2500000
Writing CONFIGS/mountdata
[root@client-18 ~]# mkdir /mnt/mds2
[root@client-18 ~]# modprobe libcfs
[root@client-18 ~]# echo -1 > /proc/sys/lnet/debug
[root@client-18 ~]# rpm -qa|grep lustre
lustre-modules-2.3.63-2.6.32_279.19.1.el6_lustre.x86_64_ge6136f1.x86_64
lustre-tests-2.3.63-2.6.32_279.19.1.el6_lustre.x86_64_ge6136f1.x86_64
lustre-ldiskfs-3.3.0-2.6.32_279.19.1.el6_lustre.x86_64_ge6136f1.x86_64
lustre-osd-ldiskfs-2.3.63-2.6.32_279.19.1.el6_lustre.x86_64_ge6136f1.x86_64
lustre-2.3.63-2.6.32_279.19.1.el6_lustre.x86_64_ge6136f1.x86_64
kernel-2.6.32-279.19.1.el6_lustre.x86_64
[root@client-18 ~]# mount -t lustre -o acl,user_xattr /dev/sda3 /mnt/mds2
mount.lustre: mount /dev/sda3 at /mnt/mds2 failed: Transport endpoint is not connected

.........

Lustre: Lustre: Build Version: jenkins-arch=x86_64,build_type=server,distro=el6,ib_stack=inkernel-1354-ge6136f1-PRISTINE-2.6.32-279.19.1.el6_lustre.x86_64
LNet: Added LNI 10.10.4.18@tcp [8/256/0/180]
LNet: Accept secure, port 988
LDISKFS-fs (sda3): mounted filesystem with ordered data mode. quota=on. Opts: 
Lustre: srv-lustre-MDT0001: No data found on store. Initialize space
Lustre: lustre-MDT0001: new disk, initializing
LustreError: 6821:0:(osp_dev.c:516:osp_connect_to_osd()) lustre-OST0000-osc: can't locate next device: lustre-MDT0000-osd
LustreError: 6821:0:(obd_config.c:572:class_setup()) setup lustre-OST0000-osc failed (-107)
LustreError: 6821:0:(obd_config.c:1550:class_config_llog_handler()) MGC10.10.4.132@tcp: cfg command failed: rc = -107
Lustre:    cmd=cf003 0:lustre-OST0000-osc  1:lustre-OST0000_UUID  2:10.10.4.132@tcp  
LustreError: 15c-8: MGC10.10.4.132@tcp: The configuration from log 'lustre-MDT0001' failed (-107). This may be the result of communication errors between this node and the MGS, a bad configuration, or other errors. See the syslog for more information.
LustreError: 6745:0:(obd_mount.c:1849:server_start_targets()) failed to start server lustre-MDT0001: -107
LustreError: 6745:0:(obd_mount.c:2401:server_fill_super()) Unable to start targets: -107
LustreError: 6745:0:(obd_mount.c:1379:lustre_disconnect_lwp()) lustre-MDT0000-lwp-MDT0001: Can't end config log lustre-client.
LustreError: 6745:0:(obd_mount.c:2115:server_put_super()) lustre-MDT0001: failed to disconnect lwp. (rc=-2)
Lustre: Failing over lustre-MDT0001
Lustre: server umount lustre-MDT0001 complete
LustreError: 6745:0:(obd_mount.c:2989:lustre_fill_super()) Unable to mount  (-107)