操作系统和informix数据库全部损坏恢复测试

主选单

技术专栏

Cheetah技术专栏

数据库管理

首页

操作系统和informix数据库全部损坏恢复测试

一，安装操作系统（redhat3u6)

1，调整内核参数
echo '
kernel.shmmax=536870912
kernel.shmmni=4096
kernel.shmall=2097152
kernel.sem=500 32000 200 256
fs.file-max=65536
net.ipv4.ip_local_port_range=1024 65000' >> /etc/sysctl.conf

2，建立用户名和组
groupadd informix
groupadd bargroup
useradd -g informix -d /home/informix -p rrlrrl informix
useradd -g bargroup -d /home/tapeback -p rrlrrl tapeback

3，分区，测试只分一个rootdbs分区就可以了
[root@test06 installfiles]# fdisk -l

Disk /dev/sda: 45.4 GB, 45492469760 bytes
255 heads, 63 sectors/track, 5530 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/sda1 * 1 13 104391 83 Linux
/dev/sda2 14 2624 20972857+ 83 Linux
/dev/sda3 2625 2755 1052257+ 82 Linux swap
/dev/sda4 2756 5530 22290187+ 5 Extended
/dev/sda5 2756 2771 128488+ 83 Linux
[root@test06 installfiles]#

4，设置raw文件
echo '
#store spaces for informix server
/dev/raw/raw1 /dev/sda5' >> /etc/sysconfig/rawdevices

5，修改/etc/init.d/rawdevices文件，设置系统启动时raw文件的属主信息
此处为修改vi /etc/init.d/rawdevices文件
case "$1" in
start)
# Assign devices
echo $"Assigning devices: "
assign_raw
sleep 1
chown informix:informix /dev/raw/*
chmod 660 /dev/raw/*
chown tapeback:bargroup /dev/nst0
echo $"done"
;;
stop)
# No action to be taken here
;;

6，设置数据库块文件
mkdir /informix.links
ln -s /dev/raw/raw1 /informix.links/rootdbs

7, 根据实际情况配置/etc/services，/etc/hosts,/home/informix/.rhosts

8，设置Informix用户环境变量
echo '
INFORMIXDIR=/home/informix
INFORMIXSERVER=ontest
DBDATE=Y4MD0
ONCONFIG=test.cfg
export LANG=en_US.ISO8859-1
export INFORMIXDIR INFORMIXSERVER DBDATE ONCONFIG
export TERMINFO=/home/informix/etc/typetab
export TERMCAP=$INFORMIXDIR/etc/termcap
PATH=$PATH:$HOME/bin:$INFORMIXDIR/bin:.
NSR_DATA_VOLUME_POOL=DBMIData
NSR_LOG_VOLUME_POOL=DBMIData
export NSR_DATA_VOLUME_POOL NSR_LOG_VOLUME_POOL PATH
'>>/home/informix/.bashrc

9,修改/etc/nsswitch.conf文件
#hosts: db files nisplus nis dns
hosts: files dns
删除dns就好了

二，安装数据库，安装networker

cp /home/informix/etc/onconfig.std /home/informix/etc/test.cfg
cp /home/informix/etc/sm_versions.std /home/informix/etc/sm_versions
cp /home/informix/etc/sqlhosts.std /home/informix/etc/sqlhosts

修改上面的3个文件后初始化数据库

建立一个库作为测试的内容

第一次备份，没有改onbar配置文件，第一次备份日志信息：
2008-03-25 09:00:44 8562 8560 /home/informix/bin/onbar_d -b -L 0
2008-03-25 09:00:44 8562 8560 Archive started on rootdbs (Requested Level 0).
2008-03-25 09:00:45 8562 8560 Begin level 0 backup rootdbs.
2008-03-25 09:00:45 8562 8560 Successfully connected to Storage Manager.
2008-03-25 09:00:47 8562 8560 Completed level 0 backup rootdbs (Storage Manager copy ID: 1206406845 1206406846).
2008-03-25 09:00:47 8562 8560 Archive on rootdbs Completed (Requested Level 0).
2008-03-25 09:00:48 8562 8560 Begin backup logical log 4.
2008-03-25 09:00:48 8562 8560 Successfully connected to Storage Manager.
2008-03-25 09:00:51 8562 8560 Completed backup logical log 4 (Storage Manager copy ID: 1206406848 1206406849).
2008-03-25 09:00:51 8562 8560 /home/informix/bin/onbar_d complete, returning 0 (0x00)

第二次备份，更改Onbar配置文件，增加：
save -b ${POOL} -l full /nsr >> ${BAR_ACT_LOG} 2>&1
save -b ${POOL} -l full ${INFORMIXDIR}/etc/ixbar.* ${INFORMIXDIR}/etc/oncfg* ${INFORMIXDIR}/etc/${ONCONFIG} >> ${BAR_ACT_LOG} 2>&1
第二次备份日志信息：
2008-03-25 09:02:37 8711 8709 /home/informix/bin/onbar_d -b -L 0
2008-03-25 09:02:37 8711 8709 Archive started on rootdbs (Requested Level 0).
2008-03-25 09:02:38 8711 8709 Begin level 0 backup rootdbs.
2008-03-25 09:02:38 8711 8709 Successfully connected to Storage Manager.
2008-03-25 09:02:40 8711 8709 Completed level 0 backup rootdbs (Storage Manager copy ID: 1206406958 1206406959).
2008-03-25 09:02:40 8711 8709 Archive on rootdbs Completed (Requested Level 0).
2008-03-25 09:02:40 8711 8709 Begin backup logical log 5.
2008-03-25 09:02:40 8711 8709 Successfully connected to Storage Manager.
2008-03-25 09:02:40 8711 8709 Completed backup logical log 5 (Storage Manager copy ID: 1206406960 1206406961).
2008-03-25 09:02:41 8711 8709 /home/informix/bin/onbar_d complete, returning 0 (0x00)
save: Using test06 as server
/nsr/res/nsrla.res
/nsr/res/servers
/nsr/res/nsrwizclnt.res
/nsr/res/.nsr
/nsr/res/nsrdb/: Permission denied
/nsr/res/
/nsr/lic/res/lgtolm.res
/nsr/lic/res/lictype.res
/nsr/lic/res/
/nsr/lic/tmp/lic_res.lck
/nsr/lic/tmp/lgtolm.res.lck
/nsr/lic/tmp/ofd256
/nsr/lic/tmp/lictype.res.lck
/nsr/lic/tmp/
/nsr/lic/logs/lgtolmd.log
/nsr/lic/logs/
/nsr/lic/cores/: Permission denied
/nsr/lic/
/nsr/index/test06/: Permission denied
/nsr/index/
/nsr/mm/
/nsr/tmp/
/nsr/applogs/xbsa.messages
/nsr/applogs/
/nsr/rap/
/nsr/logs/messages
/nsr/logs/summary
/nsr/logs/daemon.log
/nsr/logs/.nsr
/nsr/logs/
/nsr/cores/: Permission denied
/nsr/debug/
/nsr/
/

save: /nsr level=full, 32 KB 00:00:00 30 files
save: Using test06 as server
/home/informix/etc/ixbar.0
/home/informix/etc/oncfg_ontest.0
/home/informix/etc/test.cfg
/home/informix/etc/
/home/informix/
/home/
/

save: /home/informix/etc/ level=full, 18 KB 00:00:01 7 files

第三次备份，备份先修改一下nsr目录的属性：
chown informix:bargroup -R /nsr
chmod g+rx -R /nsr
备份的日志信息：
2008-03-25 09:06:17 8837 8835 /home/informix/bin/onbar_d -b -L 0
2008-03-25 09:06:18 8837 8835 Archive started on rootdbs (Requested Level 0).
2008-03-25 09:06:18 8837 8835 Begin level 0 backup rootdbs.
2008-03-25 09:06:18 8837 8835 Successfully connected to Storage Manager.
2008-03-25 09:06:20 8837 8835 Completed level 0 backup rootdbs (Storage Manager copy ID: 1206407178 1206407179).
2008-03-25 09:06:20 8837 8835 Archive on rootdbs Completed (Requested Level 0).
2008-03-25 09:06:20 8837 8835 Begin backup logical log 6.
2008-03-25 09:06:20 8837 8835 Successfully connected to Storage Manager.
2008-03-25 09:06:20 8837 8835 Completed backup logical log 6 (Storage Manager copy ID: 1206407180 1206407181).
2008-03-25 09:06:20 8837 8835 /home/informix/bin/onbar_d complete, returning 0 (0x00)
save: Using test06 as server
/nsr/res/nsrla.res
/nsr/res/servers
/nsr/res/nsrwizclnt.res
/nsr/res/.nsr
/nsr/res/nsrdb/00/0a00ed1f0f4de847ac190174
/nsr/res/nsrdb/00/1400ed1f0f4de847ac190174
/nsr/res/nsrdb/00/1e00ed1f0f4de847ac190174
/nsr/res/nsrdb/00/2800ed1f0f4de847ac190174
/nsr/res/nsrdb/00/3200ed1f0f4de847ac190174
/nsr/res/nsrdb/00/3c00ed1f0f4de847ac190174
/nsr/res/nsrdb/00/4600ed1f0f4de847ac190174
/nsr/res/nsrdb/00/5000ed1f0f4de847ac190174
/nsr/res/nsrdb/00/5a00ed1f0f4de847ac190174
/nsr/res/nsrdb/00/6400ed1f0f4de847ac190174
/nsr/res/nsrdb/00/
/nsr/res/nsrdb/01/0b00ed1f0f4de847ac190174
/nsr/res/nsrdb/01/1500ed1f0f4de847ac190174
/nsr/res/nsrdb/01/6500ed1f0f4de847ac190174
/nsr/res/nsrdb/01/4700ed1f0f4de847ac190174
/nsr/res/nsrdb/01/5100ed1f0f4de847ac190174
/nsr/res/nsrdb/01/
/nsr/res/nsrdb/02/0c00ed1f0f4de847ac190174
:$
/nsr/res/nsrdb/09/0900ed1f0f4de847ac190174
/nsr/res/nsrdb/09/1300ed1f0f4de847ac190174
/nsr/res/nsrdb/09/1d00ed1f0f4de847ac190174
/nsr/res/nsrdb/09/2700ed1f0f4de847ac190174
/nsr/res/nsrdb/09/3100ed1f0f4de847ac190174
/nsr/res/nsrdb/09/4500ed1f0f4de847ac190174
/nsr/res/nsrdb/09/4f00ed1f0f4de847ac190174
/nsr/res/nsrdb/09/5900ed1f0f4de847ac190174
/nsr/res/nsrdb/09/6300ed1f0f4de847ac190174
/nsr/res/nsrdb/09/
/nsr/res/nsrdb/
/nsr/res/
/nsr/lic/res/lgtolm.res
/nsr/lic/res/lictype.res
/nsr/lic/res/
/nsr/lic/tmp/lic_res.lck
/nsr/lic/tmp/lgtolm.res.lck
/nsr/lic/tmp/ofd256
/nsr/lic/tmp/lictype.res.lck
/nsr/lic/tmp/
/nsr/lic/logs/lgtolmd.log
/nsr/lic/logs/
/nsr/lic/cores/lgtolmd/.nsr
/nsr/lic/cores/lgtolmd/
/nsr/lic/cores/
/nsr/lic/
/nsr/index/test06/
/nsr/index/
/nsr/mm/
/nsr/tmp/
/nsr/applogs/xbsa.messages
/nsr/applogs/
/nsr/rap/
/nsr/logs/messages
/nsr/logs/summary
/nsr/logs/daemon.log
/nsr/logs/.nsr
/nsr/logs/
/nsr/cores/nsrexecd/.nsr
/nsr/cores/nsrexecd/
/nsr/cores/nsrd/.nsr
/nsr/cores/nsrd/
/nsr/cores/nsrmmdbd/.nsr
/nsr/cores/nsrmmdbd/
/nsr/cores/nsrindexd/.nsr
/nsr/cores/nsrindexd/
/nsr/cores/nsrck/.nsr
/nsr/cores/nsrck/
/nsr/cores/nsrmmd/.nsr
/nsr/cores/nsrmmd/
/nsr/cores/
/nsr/debug/
/nsr/
/

save: /nsr level=full, 80 KB 00:00:00 139 files
save: Using test06 as server
/home/informix/etc/ixbar.0
/home/informix/etc/oncfg_ontest.0
/home/informix/etc/test.cfg
/home/informix/etc/
/home/informix/
/home/
/

save: /home/informix/etc/ level=full, 18 KB 00:00:01 7 files

三，恢复测试
1，首先关闭数据库，停止networker,移动informix目录和/nsr演示数据库坏和networker坏
2，删除networker重新安装，重新安装informix

安装完networker后，放进去磁带，此时的状态为：
[root@test06 installfiles]# nsrmm -m
nsrmm: RAP error: 4mm 20GB tape test.01 not in media index
[root@test06 installfiles]# mminfo -m
mminfo: no matches found for the query
[root@test06 installfiles]# mt -f /dev/nst0 status
SCSI 2 tape drive:
File number=10, block number=2, partition=0.
Tape block size 0 bytes. Density code 0x25 (DDS-3).
Soft error count since last status=0
General status bits on (1010000):
ONLINE IM_REP_EN
[root@test06 installfiles]#

3，scanner重建磁带媒体和索引信息
[root@test06 installfiles]# scanner -i /dev/nst0
scanner: scanning 4mm 20GB tape test.01 on /dev/nst0
scanner: adding 4mm 20GB tape test.01 to pool DBMIData
scanner: ssid 4293414589: scan complete
scanner: ssid 4293414589: 5023 KB, 1 file(s)
scanner: ssid 4276637376: scan complete
scanner: ssid 4276637376: 1865 KB, 1 file(s)
scanner: ssid 4259860270: scan complete
scanner: ssid 4259860270: 5023 KB, 1 file(s)
scanner: ssid 4243083056: scan complete
scanner: ssid 4243083056: 19 KB, 1 file(s)
scanner: ssid 4226305842: scan complete
scanner: ssid 4226305842: 32 KB, 30 file(s)
scanner: ssid 4209528626: scan complete
scanner: ssid 4209528626: 18 KB, 7 file(s)
scanner: ssid 4192751626: scan complete
scanner: ssid 4192751626: 5023 KB, 1 file(s)
scanner: ssid 4175974412: scan complete
scanner: ssid 4175974412: 17 KB, 1 file(s)
scanner: ssid 4159197198: scan complete
scanner: ssid 4159197198: 80 KB, 139 file(s)
scanner: ssid 4142419982: scan complete
scanner: ssid 4142419982: 18 KB, 7 file(s)
scanner: done with 4mm 20GB tape test.01

[root@test06 installfiles]#

常用的scanner参数：
-m 设备为存储卷内的所有内容重建媒体数据库项
-i 设备为存储卷内的所有内容重建媒体数据库项和客户机文件索引项。
-S 存储集ID -m 设备为存储卷内的特定存储集重建媒体数据库项
-S 存储集ID -i 设备为存储卷内的特定存储集重建媒体数据库项和客户机文件索引项。
-c client_name 只处理来自指定NetWorker客户机的存储集。
-N 存储集名只处理有存储集名指定的存储集。
-S 存储集ID 向标准输出发送一个特定的存储集
-x 针对每个存储集引入进一步的处理命令。
-B 在卷上定位引导数据库存储集。

4，此时可以加载磁带信息：
[root@test06 installfiles]# nsrmm -m
4mm 20GB tape test.01 mounted on /dev/nst0, write protected
[root@test06 installfiles]# watch
volume written (%) expires read mounts capacity
test.01 17 MB 0.1% 03/26/09 0 KB 1 20 GB
Current tape test.01 MB is used!
[root@test06 installfiles]#

5，用recover恢复备份的配置文件
[root@test06 installfiles]# recover /home/informix/etc
recover: Current working directory is /home/informix/etc/
recover> list
0 file(s) marked for recovery
recover> lf
ixbar.0 oncfg_ontest.0 test.cfg
recover> add ixbar.0
1 file(s) marked for recovery
recover> add oncfg_ontest.0
2 file(s) marked for recovery
recover> add test.cfg
3 file(s) marked for recovery
recover> recover
recover: Total estimated disk space needed for recover is 24 KB.
Recovering 3 files into their original locations
Volumes needed (all on-line):
test.01 at /dev/nst0
Requesting 3 file(s), this may take a while...
./ixbar.0
./oncfg_ontest.0
./test.cfg
Received 3 file(s) from NSR server test06'
Recover completion time: Tue Mar 25 09:33:18 2008
recover> quit
[root@test06 installfiles]#

onbar恢复失败，查看bar_act.log文件，信息如下：
2008-03-25 09:42:48 22836 22834 /home/informix/bin/onbar_d -r
2008-03-25 09:42:48 22836 22834 (-43078) Open or close failed on file '/home/informix/etc/ixbar.0', errno = 13 .
Permission denied
2008-03-25 09:42:53 22836 22834 /home/informix/bin/onbar_d complete, returning 142 (0x8e)

修改这些文件的属主信息：
[root@test06 etc]# chown informix:informix ixbar.0
[root@test06 etc]# chown informix:informix oncfg_ontest.0
[root@test06 etc]# chown informix:informix test.cfg

重新恢复也是失败，查看bar_act.log得到日志信息：
2008-03-25 09:46:40 22940 22938 /home/informix/bin/onbar_d -r
2008-03-25 09:46:40 22940 22938 Successfully connected to Storage Manager.
2008-03-25 09:47:04 22940 22938 Begin cold level 0 restore rootdbs (Storage Manager copy ID: 1206407178 1206407179).
2008-03-25 09:47:04 22940 22938 Unable to write storage space restore data to the database server: Error opening file /informix.links/rootdbs.
errno=13.
2008-03-25 09:47:04 22940 22938 (-43391) Skipped backup/restore of space 'rootdbs'.
2008-03-25 09:47:09 22940 22938 (-43140) Due to the previous error, logical restore will not be attempted.
2008-03-25 09:47:09 22940 22938 /home/informix/bin/onbar_d complete, returning 131 (0x83)
~

原因：
刚才删除infomrix用户后重新建立了，此时用户的uid和gid与现在的不同，造就/dev/raw/raw1文件的权限不对，修改：
[root@test06 etc]# chown informix.informix /dev/raw/raw1
[root@test06 etc]# chmod 660 /dev/raw/raw1

此时恢复可以成功（服务器没有变化，是同一台服务器）

四，恢复测试
上面的恢复步骤有点担心和服务器没有重新安装有关系，把这台服务器重新安装，然后重新做恢复。
1，scanner重建磁带媒体和索引信息
[root@test06 installfiles]# scanner -i /dev/nst0
scanner: scanning 4mm 20GB tape test.02 on /dev/nst0
scanner: adding 4mm 20GB tape test.02 to pool DBMIData
scanner: ssid 4293417871: scan complete
scanner: ssid 4293417871: 63 KB, 1 file(s)
scanner: ssid 4276789414: scan complete
scanner: ssid 4276789414: 5023 KB, 1 file(s)
scanner: ssid 4260012200: scan complete
scanner: ssid 4260012200: 23 KB, 1 file(s)
scanner: ssid 4243235126: scan complete
scanner: ssid 4243235126: 5023 KB, 1 file(s)
scanner: ssid 4226457912: scan complete
scanner: ssid 4226457912: 17 KB, 1 file(s)
scanner: ssid 4209680848: scan complete
scanner: ssid 4209680848: 5023 KB, 1 file(s)
scanner: ssid 4192903634: scan complete
scanner: ssid 4192903634: 17 KB, 1 file(s)
scanner: ssid 4176126419: scan complete
scanner: ssid 4176126419: 104 KB, 30 file(s)
scanner: ssid 4159349203: scan complete
scanner: ssid 4159349203: 20 KB, 7 file(s)
scanner: ssid 4142572070: scan complete
scanner: ssid 4142572070: 5023 KB, 1 file(s)
scanner: ssid 4125794856: scan complete
scanner: ssid 4125794856: 17 KB, 1 file(s)
scanner: ssid 4109017641: scan complete
scanner: ssid 4109017641: 153 KB, 141 file(s)
scanner: ssid 4092240425: scan complete
scanner: ssid 4092240425: 20 KB, 7 file(s)
scanner: done with 4mm 20GB tape test.02

[root@test06 installfiles]#

2，加载磁带
[root@test06 installfiles]# nsrmm -m
4mm 20GB tape test.02 mounted on /dev/nst0, write protected
[root@test06 installfiles]# watch
volume written (%) expires read mounts capacity
test.02 21 MB 0.1% 03/28/09 0 KB 1 20 GB
Current tape test.02 MB is used!
[root@test06 installfiles]#

3，用recover恢复备份的配置文件
[root@test06 installfiles]# recover /home/informix/etc
recover: Current working directory is /home/informix/etc/
recover> lf
ixbar.0 oncfg_ontest.0 test.cfg
recover> add ixbar.0
1 file(s) marked for recovery
recover> add oncfg_ontest.0
2 file(s) marked for recovery
recover> add test.cfg
3 file(s) marked for recovery
recover> recover
recover: Total estimated disk space needed for recover is 24 KB.
Recovering 3 files into their original locations
Volumes needed (all on-line):
test.02 at /dev/nst0
Requesting 3 file(s), this may take a while...
./ixbar.0
./oncfg_ontest.0
./test.cfg
Received 3 file(s) from NSR server test06'
Recover completion time: Thu Mar 27 06:33:44 2008
recover> quit
[root@test06 installfiles]#

4，
[informix@test06 informix]$ onbar -r
[informix@test06 informix]$ echo $?
131
[informix@test06 informix]$

2008-03-27 06:47:05 15306 15304 /home/informix/bin/onbar_d -r
2008-03-27 06:47:05 15306 15304 Successfully connected to Storage Manager.
2008-03-27 06:47:33 15306 15304 Begin cold level 0 restore rootdbs (Storage Manager copy ID: 1206559269 1206559270).
2008-03-27 06:47:33 15306 15304 Unable to write storage space restore data to the database server: Error opening file /informix.links/rootdbs.
errno=13.
2008-03-27 06:47:33 15306 15304 (-43391) Skipped backup/restore of space 'rootdbs'.
2008-03-27 06:47:38 15306 15304 (-43140) Due to the previous error, logical restore will not be attempted.
2008-03-27 06:47:38 15306 15304 /home/informix/bin/onbar_d complete, returning 131 (0x83)

当时的raw文件权限不对，重新onbar -r 恢复也是可以成功的！