存储未分区直接创建文件系统

xxxxxxxxbbbb

小弟有一套磁盘柜, 基于 multipath , 分成了多个 LUN:

1. /dev/mapper/data1      2.7T  9.2G  2.6T   1% /d1
   
2. /dev/mapper/data2      2.7T  201M  2.6T   1% /d2

复制代码

这两个 LUN 均未分区, 直接 mkfs 创建了文件系统(ext4)

1. fdisk -l /dev/mapper/data1
   
2. 
   
3. Disk /dev/mapper/data1: 3000.1 GB, 3000055056896 bytes
   
4. 255 heads, 63 sectors/track, 364735 cylinders
   
5. Units = cylinders of 16065 * 512 = 8225280 bytes
   
6. Sector size (logical/physical): 512 bytes / 512 bytes
   
7. I/O size (minimum/optimal): 512 bytes / 512 bytes
   
8. Disk identifier: 0x00000000
   
9. 
   
10. Disk /dev/mapper/data1 doesn't contain a valid partition table
    
11. 
    
12. /dev/mapper/data1 on /d1 type ext4 (rw)
    
13. /dev/mapper/data2 on /d2 type ext4 (rw)
    

复制代码

目前存储数据没有出现问题, 但担心以后会出现其它问题. 特请教各位, 这种情况是否存在风险?

恳请各位指点, 不胜感激.

dengbao2001

这样用没问题的。

我当前环境和这个差不多

LUN过来做了多路径，访问mpath下面的，不过我后面还做LVM，然后格式化LV了。

xxxxxxxxbbbb

谢谢 "dengbao2001"

不幸的是小弟昨天发贴询问题, 昨晚存储就出了问题:

1. [5426593.159564] EXT4-fs error (device dm-7): ext4_ext_find_extent: bad header/extent in inode #29623215: invalid magic - magic 9e02, entries 42213, max 14989(0), depth 21024(0)
   
2. [5426593.160092] EXT4-fs error (device dm-7): ext4_ext_remove_space: bad header/extent in inode #29623215: invalid magic - magic 9e02, entries 42213, max 14989(0), depth 21024(0)
   
3. [5426593.160586] EXT4-fs error (device dm-7): file system corruption: inode #29623215 logical block 1602 mapped to 222654511637520 (size 1)
   
4. [5426593.161076] EXT4-fs error (device dm-7): ext4_ext_remove_space: bad header/extent in inode #29623215: invalid magic - magic 9e02, entries 42213, max 14989(0), depth 21024(0)
   
5. 
   
6. 
   
7. [5442142.543223] This should not happen!!  Data will be lost
   
8. [5442319.017307] INFO: task jbd2/dm-7-8:3934 blocked for more than 120 seconds.
   
9. [5442319.017387] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
   
10. [5442319.017534] jbd2/dm-7-8   D ffff8810276ec900     0  3934      2    0 0x00000080
    
11. [5442319.017543]  ffff881026195d10 0000000000000046 00000000ffffffff 0000000000000008
    
12. [5442319.017555]  000000000001e240 000000000001e240 ffff88102c4d8000 000000000001e240
    
13. [5442319.017565]  000000000001e240 ffff8810276eceb8 ffff881026195fd8 ffff881026195fd8
    
14. [5442319.017575] Call Trace:
    
15. [5442319.017586]  [<ffffffff81095ace>] ? prepare_to_wait+0x4e/0x80
    
16. [5442319.017612]  [<ffffffffa00b981f>] jbd2_journal_commit_transaction+0x19f/0x14b0 [jbd2]
    
17. [5442319.017622]  [<ffffffff81009780>] ? __switch_to+0xd0/0x320
    
18. [5442319.017632]  [<ffffffff8107e4ac>] ? lock_timer_base+0x3c/0x70
    
19. [5442319.017643]  [<ffffffff810957e0>] ? autoremove_wake_function+0x0/0x40
    
20. [5442319.017657]  [<ffffffffa00bf3b8>] kjournald2+0xb8/0x220 [jbd2]
    
21. [5442319.017664]  [<ffffffff810957e0>] ? autoremove_wake_function+0x0/0x40
    
22. [5442319.017677]  [<ffffffffa00bf300>] ? kjournald2+0x0/0x220 [jbd2]
    
23. [5442319.017684]  [<ffffffff81095206>] kthread+0x96/0xa0
    
24. [5442319.017691]  [<ffffffff8100c20a>] child_rip+0xa/0x20
    
25. [5442319.017698]  [<ffffffff81095170>] ? kthread+0x0/0xa0
    
26. [5442319.017704]  [<ffffffff8100c200>] ? child_rip+0x0/0x20
    
27. [5442319.017719] INFO: task flush-253:7:4067 blocked for more than 120 seconds.

复制代码

beyondfly

不懂，帮顶

xxxxxxxxbbbb

回复 4# beyondfly

谢谢.

冬瓜头

不分区一样用，没问题。报错不是因为这个导致

xxxxxxxxbbbb

回复 6# 冬瓜头

哇, 是冬瓜头呀. 谢谢! 没想到这么长时间的帖子竟被翻了出来.

这个问题的后续, 存储故障后, 在未调整 Multipath 及 LUN 的情况下, 分区并创建文件系统, 至目前止未再出现存储问题.