iscsi挂载连接异常

基础运维的兄弟在一台台连接iscsi盘之后再系统日志中发现一堆错误：

[root@zlq /]# tail -f /var/log/messages
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:41:59 zlq kernel: connection2:0: detected conn error (1020)
Aug 28 15:41:59 zlq kernel: connection5:0: detected conn error (1020)
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:41:59 zlq kernel: connection4:0: detected conn error (1020)
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:41:59 zlq kernel: connection3:0: detected conn error (1020)
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:41:59 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq kernel: connection2:0: detected conn error (1020)
Aug 28 15:42:01 zlq kernel: connection5:0: detected conn error (1020)
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq iscsid: semop down failed 22
Aug 28 15:42:01 zlq kernel: connection4:0: detected conn error (1020)
Aug 28 15:42:01 zlq kernel: connection3:0: detected conn error (1020)
Aug 28 15:42:02 zlq iscsid: semop down failed 22

发现很多异常连接

ls -l /dev/disk/*
lrwxrwxrwx 1 root root  9 8月   6 15:56 ip-10.80.22.2:3260-iscsi-iqn.6173-05.storos.t-1-lun-0 -> ../../sdb
lrwxrwxrwx 1 root root  9 8月  18 10:47 ip-10.97.100.210:3260-iscsi-iqn.6173-05.storos.t-107-lun-0 -> ../../sdc
lrwxrwxrwx 1 root root  9 8月  18 13:50 ip-10.97.100.210:3260-iscsi-iqn.6173-05.storos.t-108-lun-0 -> ../../sde
lrwxrwxrwx 1 root root  9 8月  18 13:50 ip-10.97.100.210:3260-iscsi-iqn.6173-05.storos.t-109-lun-0 -> ../../sdf
lrwxrwxrwx 1 root root  9 8月  18 13:43 ip-10.97.100.210:3260-iscsi-iqn.6173-05.storos.t-110-lun-0 -> ../../sdd
lrwxrwxrwx 1 root root  9 8月  18 15:48 ip-10.97.100.210:3260-iscsi-iqn.6173-05.storos.t-120-lun-0 -> ../../sdg
lrwxrwxrwx 1 root root  9 5月   5 12:39 pci-0000:03:00.0-scsi-0:2:0:0 -> ../../sda
lrwxrwxrwx 1 root root 10 5月   5 12:40 pci-0000:03:00.0-scsi-0:2:0:0-part1 -> ../../sda1
lrwxrwxrwx 1 root root 10 5月   5 12:40 pci-0000:03:00.0-scsi-0:2:0:0-part2 -> ../../sda2

实际只有sda、sdb、sdg在用

[root@zlq /]# df -hl
文件系统              容量  已用  可用 已用%% 挂载点
/dev/mapper/vg_zlq-lv_root
                       50G  7.4G   40G  16% /
tmpfs                  16G     0   16G   0% /dev/shm
/dev/sda1             485M   37M  423M   8% /boot
/dev/mapper/vg_zlq-lv_home
                      8.0T  5.7T  2.0T  75% /home
/dev/sdb              8.1T  1.5T  6.3T  19% /databak
/dev/sdg              7.6T  177M  7.2T   1% /databak2

发现iscis异常

[root@zlq /]# iscsiadm -m discovery -t sendtargets -p 10.97.100.210                                                                                                         
iscsiadm: This command will remove the record [iface: default, target: iqn.6173-05.storos.t-109, portal: 10.97.100.210,3260], but a session is using it. Logout session then rerun command to remove record.
iscsiadm: This command will remove the record [iface: default, target: iqn.6173-05.storos.t-107, portal: 10.97.100.210,3260], but a session is using it. Logout session then rerun command to remove record.
iscsiadm: This command will remove the record [iface: default, target: iqn.6173-05.storos.t-110, portal: 10.97.100.210,3260], but a session is using it. Logout session then rerun command to remove record.
iscsiadm: This command will remove the record [iface: default, target: iqn.6173-05.storos.t-108, portal: 10.97.100.210,3260], but a session is using it. Logout session then rerun command to remove record.
10.97.100.210:3260,1 iqn.6173-05.storos.t-120

解决方法，重启iscsi，重新挂载。
ps：重启后盘符可能产生变动，如原sdg变成sdd，需要重新挂载