HP-UX磁带备份错误收集

磁带备份命令：

make_tape_recovery -Av  默认备份至/dev/rmt/0mn。

如果有多个磁带机，那么需要使用下面命令

make_tape_recovery -Av -a /dev/rmt/1mn

lsdb#[/tmp]ioscan  -fnC tape  ----扫描磁带机

lsdb#[/tmp]make_tape_recovery -Av -a /dev/rmt/2mn   -这个通过ioscan -fnC 查看磁带机的路径

       * Creating local directories for configuration files and archive.

=======  12/21/12 18:07:42 EAT  Started make_tape_recovery. (Fri Dec 21 18:07:42 EAT 2012)

         @(#)Ignite-UX Revision C.7.13.240

         @(#)ignite/net_recovery (opt) Revision: /branches/IUX_RA1109/ignite/src@83765 Last Modified: 2011-06-14 10:28:20 +0530 (Tue, 14 Jun 2011)

       * Recovery Archive Description = Recovery Archive

         

       * Recovery Archive Location    = /dev/rmt/2mn

 ........

错误实例一：

备份最后报错：（PA的CPU，安腾的就没有这个错）

ERROR:   System Error, /opt/ignite/bin/make_medialif failed creating boot LIF file.

ERROR:   Failed to generate LIF on tape.

解决办法：

solution:In /opt/ignite/bin/make_medialif

change LIFVolSize='500000000' to LIFVolSize='750000000'.

there is obviously not enough space in the defined LIF area to hold all the files

 

错误实例二：多个的备份进程在执行，备份时速度很慢的，需要耐心等待，不能马上就KILL掉，容易出现这种问题。

ERROR:   The make_sys_image command failed. The system recovery archive will 

         not be created.

       * Creating local directories for configuration files and archive.

=======  09/17/13 16:39:18 EAT  Started make_tape_recovery. (Tue Sep 17 

         16:39:18 EAT 2013)

         @(#) Ignite-UX Revision C.6.0.115

         @(#) net_recovery (opt) $Revision: 10.655 $

NOTE:    Another make_sys_image process appears to be running.

         Please wait for its completion or remove lockfile(s):

         /var/tmp/ign_configure/make_sys_image_flag16491

NOTE:    The other make_sys_image process still has not stopped.

NOTE:    The other make_sys_image process still has not stopped.

解决办法：

#ps -ef |grep make

Kill any make_sys_image processes you find.

Kill -9 if you need to. (sometimes you need to)

Remove files from /var/tmp/ign_configure.

Try it again. 

查看/var/tmp/ign_configure

把下面的文件删除，或者MV掉，重新执行make_tape_recovery -Av 

http://h30499.www3.hp.com/t5/Ignite-UX/ERROR-previous-make-sys-image-process-hung/td-p/4057549

http://h30499.www3.hp.com/t5/Ignite-UX/The-other-make-sys-image-process-still-has-not-stopped/td-p/3243696

 

错误实例三 http://www.itpub.net/thread-529815-1-1.html

make_recovery -ACv后出现下列问题

make_recovery(466): Call to /opt/ignite/bin/make_medialif failed.

       Check the /var/opt/ignite/logs/makrec.log1 file.

        Cleanup

察看文件

********

Started - Wed Apr 19 03:34:01 2006

********

make_recovery -A -C -v

/opt/ignite/lbin/list_expander -d

/opt/ignite/lbin/list_expander -d

/opt/ignite/lbin/list_expander -d -v

"/opt/ignite/bin/save_config" -f "/var/opt/ignite/recovery/config.recover" "vg00

"  >> "/var/opt/ignite/logs/makrec.log1" 2>&1

/usr/sbin/vgcfgbackup "vg00" >> /var/opt/ignite/logs/makrec.log1 2>&1

Volume Group configuration for /dev/vg00 has been saved in /etc/lvmconf/vg00.con

f

/usr/sbin/swlist -X /opt/ignite/lib/sd_opts -l product > "/var/tmp/makrec.lasttm

p"

"/opt/ignite/bin/instl_adm" -T -f "/var/opt/ignite/recovery/config.recover" >> /

var/opt/ignite/logs/makrec.log1 2>&1

"/opt/ignite/bin/make_medialif" -v -a  -f "/var/opt/ignite/recovery/config.recov

er" -l "/var/tmp/uxinstlf.recovery" -C "HP-UX System Recovery" >> /var/opt/ignit

e/logs/makrec.log1 2>&1

lifcp: Cannot create SYSCMDS   ; no space on the LIF volume

make_medialif: error - lifcp SYSCMDS failed, probable insufficient size for /var

/tmp/uxinstlf.recovery

磁带是好的，磁带机是测试的

tar没有问题

stm没有问题

单单make_recovery有问题 

/dev/vg00/lvol6    2072576  258304 1800320   13% /var 

回答一：

这个问题我以前也碰到过，修改/opt/iignite/bin/make_medialif 

修改这一段：

# Create lif volume using ipl_options

把  $ipl_options -v72000000  修改成 $ipl_options -v102000000  （把参数适当调大一点）

回答二：

是ignite的问题

 

 

错误实例四：
ERROR:   Not Enough Disk Space. At least 33556500 bytes is needed for boot LIF
         file.
       * Creating local directories for configuration files and archive.

解决办法：
方法一：指定uxinstlf.recovery 生成目录
To resolve this problem, it is necessary to increase the disk space in
/var. In the meantime, use the -B option to specify a location other
than /var/tmp/uxinstlf.recovery where more space exists.
For example, suppose /opt has lots of space, then use:
# make_tape_recovery -Av -B /opt/uxinstlf.recovery

方法二：增加/var目录大小
因为默认路径为：The default file is /var/tmp/uxinstlf.recovery.

 

错误实例五：http://www.ixpub.net/thread-1204390-1-5.html
WARNING: Skipping symlink's target ('/var/adm/wtmp') because it is larger than
         2GB, which is not supported.  This file should be backed-up with your
         normal system backup -- Value too large to be stored in data type
pax: home/coder/ctei\webapps\oa\doc\mailhome/coder/ctei\webapps\oa\doc\mail\1250589128032 : No such file or directory  ---删除

问题解决：把那反斜杠文件目录删了
另外每个一文件有3G多，超过限制的容量2G也不能备份

错误案例六
报错如下：
       * Creating Control Configuration.
       * Creating Archive File List
       * Creating Archive Configuration


       * /opt/ignite/lbin/make_arch_config -c 
         /var/opt/ignite/recovery/2014-01-17,08:49/archive_cfg -g 
         /var/opt/ignite/recovery/2014-01-17,08:49/flist -n 2014-01-17,08:49 -r
         ipf -b 64 -d Recovery\ Archive -t 22 -i 1 -m t
       * Saving the information about archive to 
         /var/opt/ignite/recovery/previews
       * Creating The Tape Archive
       * Checking configuration file for Syntax
NOTE:    File: /var/tmp/tmp_cfg_file is not world-readable, this may cause 
         clients to fail to read it during an installation.
NOTE:    File: /tmp/cfg is not world-readable, this may cause clients to fail 
         to read it during an installation.
NOTE:    File: /var/tmp/IINSTALLFS is not world-readable, this may cause 
         clients to fail to read it during an installation.
       * Complete writing boot LIF onto tape.
       * /opt/ignite/data/scripts/make_sys_image -c n -d /dev/rmt/0mn -t n -s 
         local -n 2014-01-17,08:49 -m t -w 
         /var/opt/ignite/recovery/2014-01-17,08:49/recovery.log -u -R -g 
         /var/opt/ignite/recovery/2014-01-17,08:49/flist -a 67676280
         
       * Preparing to create a system archive.


       * Archiving contents of pms_db2 via tar image to local device /dev/rmt/0mn.
pax: write : I/O error
pax: A write error has occurred.
WARNING: The pax command returned a non-zero exit status (exit status 1).
ERROR:   The find_files command failed (exit status 141).
=======  01/17/14 10:21:50 EAT  make_tape_recovery completed unsuccessfully


查看磁带机状态：
[root@pms_db2] /tmp> mt -f /dev/rmt/0mn status   ----能备份，但到一半就报错
Drive:  HP C7438A
Format: 
Status: [81114700] EOF online compression immediate-report-mode   --磁带到了最后或者中间，需要反转
File:   23
Block:  -1
正常备份完状态为：
[root@pms_db1] /tmp>  mt -f /dev/rmt/36mn status    ----状态为正常，磁带处于开始状态
Drive:  HP C7438A
Format: 
Status: [41114700] BOT online compression immediate-report-mode 
File:   0

Block:  0
IO错误表明是物理设备存在问题，换了一盒磁带就备份成功了。可能是磁带不能读的问题吧。
同样的错误案例：http://h30499.www3.hp.com/t5/Ignite-UX/Ignite-backup-failed/m-p/4777397#M18221