分类目录归档：操作系统

Disable Transparent HugePages

发表于 2017 年 7 月 25 日 由 惜分飞

从redhat 6开始引入了Transparent HugePages,但是oracle一直建议disable 它,而使用标准HugePages方式.对于6和7的禁用方式有一些区别.
linux 6
修改/etc/grub.conf之后重启系统生效

vi /etc/grub.conf 

title Oracle Linux Server (2.6.32-300.25.1.el6uek.x86_64)
        root (hd0,0)
        kernel /vmlinuz-2.6.32-300.25.1.el6uek.x86_64 ro root=LABEL=/  transparent_hugepage=never
        initrd /initramfs-2.6.32-300.25.1.el6uek.x86_64.img

linux 7
修改/etc/default/grub然后执行grub2-mkconfig并重启系统生效

[root@xifenfei u01]# vi /etc/default/grub

GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL_OUTPUT="console"
GRUB_CMDLINE_LINUX="nomodeset vconsole.font=latarcyrheb-sun16 vconsole.keymap=us crashkernel=auto  
                    biosdevname=0 transparent_hugepage=never"
GRUB_DISABLE_RECOVERY="true"
~

[root@xifenfei u01]# grub2-mkconfig -o /boot/grub2/grub.cfg
Generating grub configuration file ...
Found linux image: /boot/vmlinuz-3.10.0-514.26.2.el7.x86_64
Found initrd image: /boot/initramfs-3.10.0-514.26.2.el7.x86_64.img
Found linux image: /boot/vmlinuz-3.10.0-123.el7.x86_64
Found initrd image: /boot/initramfs-3.10.0-123.el7.x86_64.img
Found linux image: /boot/vmlinuz-0-rescue-cb0b6b4de89a4fe4acfc8774c2f01486
Found initrd image: /boot/initramfs-0-rescue-cb0b6b4de89a4fe4acfc8774c2f01486.img
done

临时禁用
该方法对于linux 6和7均有效,不用重启系统

if test -f /sys/kernel/mm/transparent_hugepage/enabled; then
   echo never > /sys/kernel/mm/transparent_hugepage/enabled
fi
if test -f /sys/kernel/mm/transparent_hugepage/defrag; then
   echo never > /sys/kernel/mm/transparent_hugepage/defrag
fi
发表在 Linux | 评论关闭

ext3/ext4 超级块修复

发表于 2016 年 11 月 12 日 由 惜分飞

创建ext4文件系统

[root@localhost ~]# mkfs.ext4 /dev/sdb1
mke2fs 1.42.9 (28-Dec-2013)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
1310720 inodes, 5242624 blocks
262131 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=2153775104
160 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
        4096000

Allocating group tables: done
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

[root@localhost ~]# mkdir /sdb
[root@localhost ~]# mount /dev/sdb1 /sdb
[root@localhost ~]# df -h
Filesystem           Size  Used Avail Use% Mounted on
/dev/mapper/ol-root   36G  4.0G   32G  12% /
devtmpfs             1.8G     0  1.8G   0% /dev
tmpfs                1.8G     0  1.8G   0% /dev/shm
tmpfs                1.8G  8.9M  1.8G   1% /run
tmpfs                1.8G     0  1.8G   0% /sys/fs/cgroup
/dev/sda1            497M  195M  303M  40% /boot
tmpfs                369M     0  369M   0% /run/user/0
/dev/sdb1             20G   45M   19G   1% /sdb

准备测试数据

[root@localhost sdb]# cd /etc/sysctl.d/
[root@localhost sysctl.d]# ls
99-sysctl.conf
[root@localhost sysctl.d]# cp 99-sysctl.conf /sdb
[root@localhost sysctl.d]# more 99-sysctl.conf
# System default settings live in /usr/lib/sysctl.d/00-system.conf.
# To override those settings, enter new settings here, or in an /etc/sysctl.d/<name>.conf file
#
# For more information, see sysctl.conf(5) and sysctl.d(5).

破坏ext4文件系统

[root@localhost ~]#  dd if=/dev/zero of=/dev/sdb1 bs=1024 count=5
5+0 records in
5+0 records out
5120 bytes (5.1 kB) copied, 0.00270838 s, 1.9 MB/s
[root@localhost ~]# mount /dev/sdb1 /sdb
mount: unknown filesystem type '(null)'

日志报错

[ 8868.362628] sd 32:0:1:0: [sdb] Cache data unavailable
[ 8868.362632] sd 32:0:1:0: [sdb] Assuming drive cache: write through
[ 8868.363714]  sdb: sdb1
[ 8868.390297] sd 32:0:1:0: [sdb] Cache data unavailable
[ 8868.390301] sd 32:0:1:0: [sdb] Assuming drive cache: write through
[ 8868.391462]  sdb: sdb1
[ 8900.130143] EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: (null)
[ 8900.130163] SELinux: initialized (dev sdb1, type ext4), uses xattr
[ 8902.803966] sdb1: WRITE SAME failed. Manually zeroing.

fsck修复

[root@localhost ~]# fsck -t ext4 /dev/sdb1
fsck from util-linux 2.23.2
e2fsck 1.42.9 (28-Dec-2013)
ext2fs_open2: Bad magic number in super-block
fsck.ext4: Superblock invalid, trying backup blocks...
/dev/sdb1 was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong for group #1 (31740, counted=31739).
Fix<y>? yes
Free blocks count wrong (5116302, counted=5116301).
Fix<y>? yes
Free inodes count wrong for group #0 (8181, counted=8180).
Fix<y>? yes
Free inodes count wrong (1310709, counted=1310708).
Fix<y>? yes

/dev/sdb1: ***** FILE SYSTEM WAS MODIFIED *****
/dev/sdb1: 12/1310720 files (0.0% non-contiguous), 126323/5242624 blocks

测试修复结果

[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# mount /dev/sdb1 /sdb
[root@localhost ~]# df -h
Filesystem           Size  Used Avail Use% Mounted on
/dev/mapper/ol-root   36G  4.0G   32G  12% /
devtmpfs             1.8G     0  1.8G   0% /dev
tmpfs                1.8G     0  1.8G   0% /dev/shm
tmpfs                1.8G  8.9M  1.8G   1% /run
tmpfs                1.8G     0  1.8G   0% /sys/fs/cgroup
/dev/sda1            497M  195M  303M  40% /boot
tmpfs                369M     0  369M   0% /run/user/0
/dev/sdb1             20G   45M   19G   1% /sdb
[root@localhost ~]# cd /sdb
[root@localhost sdb]# ls
99-sysctl.conf  lost+found
[root@localhost sdb]# more 99-sysctl.conf
# System default settings live in /usr/lib/sysctl.d/00-system.conf.
# To override those settings, enter new settings here, or in an /etc/sysctl.d/<name>.conf file
#
# For more information, see sysctl.conf(5) and sysctl.d(5).

ext4文件系统修复

[root@localhost ~]# mkfs.ext3 /dev/sdb1
mke2fs 1.42.9 (28-Dec-2013)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
1310720 inodes, 5242624 blocks
262131 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
160 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
        4096000

Allocating group tables: done
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done

[root@localhost ~]# mount /dev/sdb1 /sdb
[root@localhost ~]# df -h
Filesystem           Size  Used Avail Use% Mounted on
/dev/mapper/ol-root   36G  4.0G   32G  12% /
devtmpfs             1.8G     0  1.8G   0% /dev
tmpfs                1.8G     0  1.8G   0% /dev/shm
tmpfs                1.8G  8.9M  1.8G   1% /run
tmpfs                1.8G     0  1.8G   0% /sys/fs/cgroup
/dev/sda1            497M  195M  303M  40% /boot
tmpfs                369M     0  369M   0% /run/user/0
/dev/sdb1             20G   45M   19G   1% /sdb
[root@localhost ~]# dd if=/dev/zero of=/dev/sdb1 bs=1024 count=5
5+0 records in
5+0 records out
5120 bytes (5.1 kB) copied, 0.0138915 s, 369 kB/s
[root@localhost ~]# fsck -t ext3 /dev/sdb1
fsck from util-linux 2.23.2
e2fsck 1.42.9 (28-Dec-2013)
ext2fs_open2: Bad magic number in super-block
fsck.ext3: Superblock invalid, trying backup blocks...
/dev/sdb1 was not cleanly unmounted, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information

/dev/sdb1: ***** FILE SYSTEM WAS MODIFIED *****
/dev/sdb1: 11/1310720 files (0.0% non-contiguous), 126322/5242624 blocks
[root@localhost ~]# mount /dev/sdb1 /sdb
[root@localhost ~]# df -h
Filesystem           Size  Used Avail Use% Mounted on
/dev/mapper/ol-root   36G  4.0G   32G  12% /
devtmpfs             1.8G     0  1.8G   0% /dev
tmpfs                1.8G     0  1.8G   0% /dev/shm
tmpfs                1.8G  8.9M  1.8G   1% /run
tmpfs                1.8G     0  1.8G   0% /sys/fs/cgroup
/dev/sda1            497M  195M  303M  40% /boot
tmpfs                369M     0  369M   0% /run/user/0
/dev/sdb1             20G   45M   19G   1% /sdb

fsck修复危险性较大,建议先备份对应的分区(dd命令备份分区)然后再处理,有导致分区数据全部或者部分丢失的风险,如果超级块彻底损坏无法恢复,请联系我们，提供专业ORACLE数据库恢复技术支持
Phone:17813235971    Q Q:107644445QQ咨询惜分飞    E-Mail:dba@xifenfei.com

发表在 Linux | 标签为 , , | 评论关闭

hp平台rdisk中磁盘丢失导致asm启动报ORA-15042恢复

发表于 2016 年 8 月 27 日 由 惜分飞

有老朋友找到我,说一个客户的数据库异常,问题是asm无法正常mount,提示是缺少两块磁盘.问我是否可以恢复.因为是内网环境,通过他那边发过来的零零散散的信息,大概分析如下
asm alert日志报错
ERROR: diskgroup DGROUP1 was not mounted

Fri Aug 12 16:03:12 EAT 2016
SQL> alter diskgroup DGROUP1 mount 
Fri Aug 12 16:03:12 EAT 2016
NOTE: cache registered group DGROUP1 number=1 incarn=0xf6781b5c
Fri Aug 12 16:03:12 EAT 2016
NOTE: Hbeat: instance first (grp 1)
Fri Aug 12 16:03:16 EAT 2016
NOTE: start heartbeating (grp 1)
Fri Aug 12 16:03:16 EAT 2016
NOTE: cache dismounting group 1/0xF6781B5C (DGROUP1) 
NOTE: dbwr not being msg'd to dismount
ERROR: diskgroup DGROUP1 was not mounted

前台尝试mount asm 磁盘组报错ORA-15042
ORA-15042

从这里可以明显的看出来asm 磁盘组无法正常mount,是由于缺少asm disk 15,16.如果想恢复asm,最好的方法就是找出来这两个磁盘.通过kfed对现在的磁盘进行分析,最后我们发现asm disk 14对应的磁盘为disk160,,asm disk 17对应的disk163,根据第一感觉很可能是disk161和disk161两块盘异常,让机房检查硬件无任何告警

OS层面分析
省略和本次结论无关的记录

ls -l /dev/rdisk
crw-rw----   1 oracle     dba         13 0x000070 Jan  1  2016 disk160
crw-rw----   1 oracle     dba         13 0x000073 Jan  1  2016 disk163

ls -l /dev/disk
brw-r-----   1 bin        sys          1 0x000070 Jan 13  2015 disk160
brw-r-----   1 bin        sys          1 0x000071 Jan 13  2015 disk161
brw-r-----   1 bin        sys          1 0x000072 Jan 13  2015 disk162
brw-r-----   1 bin        sys          1 0x000073 Jan 13  2015 disk163

这里我们发现在hp unix中/dev/disk下面磁盘都存在,但是/dev/rdisk下面丢失，通过ioscan相关命令继续分析

ioscan -fNnkC disk
disk    160  64000/0xfa00/0x70  esdisk   CLAIMED     DEVICE       HP      OPEN-V
                      /dev/disk/disk160   /dev/rdisk/disk160
disk    161  64000/0xfa00/0x71  esdisk   CLAIMED     DEVICE       HP      OPEN-V
                      /dev/disk/disk161
disk    162  64000/0xfa00/0x72  esdisk   CLAIMED     DEVICE       HP      OPEN-V
                      /dev/disk/disk162
disk    163  64000/0xfa00/0x73  esdisk   CLAIMED     DEVICE       HP      OPEN-V
                      /dev/disk/disk163   /dev/rdisk/disk163

这里我们基本上可以确定是/dev/rdisk下面的盘发生丢失.进一步分析,因为rdisk是聚合后的盘符,那我们分析聚合前的盘符是否正常

ioscan -m dsf
/dev/rdisk/disk160       /dev/rdsk/c29t12d4
                         /dev/rdsk/c28t12d4
/dev/rdisk/disk163       /dev/rdsk/c29t12d7
                         /dev/rdsk/c28t12d7

ls -l /dev/rdsk
crw-r-----   1 bin        sys        188 0x1dc000 Apr 22  2014 c29t12d0
crw-r-----   1 bin        sys        188 0x1dc100 Apr 22  2014 c29t12d1
crw-r-----   1 bin        sys        188 0x1dc300 Jan 13  2015 c29t12d3
crw-r-----   1 bin        sys        188 0x1dc400 Jan 13  2015 c29t12d4
crw-r-----   1 bin        sys        188 0x1dc500 Jan 13  2015 c29t12d5
crw-r-----   1 bin        sys        188 0x1dc600 Jan 13  2015 c29t12d6
crw-r-----   1 bin        sys        188 0x1dc700 Jan 13  2015 c29t12d7

crw-r-----   1 bin        sys        188 0x1cc100 Apr 22  2014 c28t12d1
crw-r-----   1 bin        sys        188 0x1cc300 Jan 13  2015 c28t12d3
crw-r-----   1 bin        sys        188 0x1cc400 Jan 13  2015 c28t12d4
crw-r-----   1 bin        sys        188 0x1cc500 Jan 13  2015 c28t12d5
crw-r-----   1 bin        sys        188 0x1cc600 Jan 13  2015 c28t12d6
crw-r-----   1 bin        sys        188 0x1cc700 Jan 13  2015 c28t12d7

通过这里我们基本上可以大概判断出来/dev/rdsk/c28t12d5,/dev/rdsk/c28t12d6,/dev/rdsk/c29t12d5,/dev/rdsk/c29t12d6就是我们需要找的/dev/rdisk/disk161和disk162的聚合之前的盘符.也就是说,现在我们判断只有/dev/rdisk下面的字符设备有问题,其他均正常.

通过系统命令修复异常

insf -e -H 64000/0xfa00/0x71
insf -e -H 64000/0xfa00/0x72

hp-asm-disk

现在已经可以正常看到/dev/rdisk/disk161和/dev/rdisk/disk162盘符,初步判断,os层面盘符已经恢复正常.修改磁盘权限和所属组

chmod 660 /dev/rdisk/disk161
chmod 660 /dev/rdisk/disk162
chown oracle:dba /dev/rdisk/disk161
chown oracle:dba /dev/rdisk/disk162

正常启动asm,mount磁盘组,open数据库
asm-mount

这次的恢复,主要是从操作系统层面判断解决问题,从而实现数据库完美恢复,数据0丢失.有类似恢复案例:分区无法识别导致asm diskgroup无法mount
如果您遇到此类情况,无法解决请联系我们，提供专业ORACLE数据库恢复技术支持
Phone:17813235971    Q Q:107644445QQ咨询惜分飞    E-Mail:dba@xifenfei.com

发表在 Oracle ASM, Unix | 标签为 , , | 评论关闭