NBU5:奇怪的96号错(已解决)

huguozhi

新加了全新的带,加到scratch pool
在数据库备份时开始mount该带,(后vmquery也看到转到database pool)
但还是报96错
(以前96号错这样处理都ok)

连加了几盒新带同样如此
用tpclean清洗了驱动器还是不行

bperror看到信息如下1，请教可能是什么问题导致
那个Error(0x10) MediaDev(0x84) bptm在troublehooting的对应什么代码?0x10 0x84?


bperror看到的该job信息如下:
===========================
03/13/2008 20:41:29 V1 S:gdclaf C:gdgrp01 J:484501 (U:484501,0)
  Info(0x4) Backup(0x4) bpsched
  added backup job (jobid=484501) for client gdgrp01, policy D_gdgrp01,
  schedule Default-Application-Backup part 0 to NetBackup scheduler work queue

03/13/2008 20:41:31 V1 S:gdclaf C:gdgrp01 J:484501 (U:484501,0)
  Info(0x4) Backup(0x4) bpsched
  started backup job for client gdgrp01, policy D_gdgrp01, schedule
  Default-Application-Backup on storage unit gdgrp01-hcart2-robot-tld-0

03/13/2008 20:41:34 V1 S:gdclaf C:gdgrp01 J:484501 (U:484501,0)
  Info(0x4) Backup(0x4) bpsched
  client gdgrp01 handling path /orange_tcp/100/4807

03/13/2008 20:53:47 V1 S:gdgrp01 C:gdgrp01 J:484501 (U:484501,0)
  Error(0x10) MediaDev(0x84) bptm
  FREEZING media id B002L2, it is unmountable and cannot be used for backups

03/13/2008 20:53:48 V1 S:gdgrp01 C:gdgrp01 J:484501 (U:484501,0)
  Error(0x10) MediaDev(0x84) bptm
  Media Manager volume pool database has no more unassigned media in robotic
  device TLD(0)

03/13/2008 20:53:52 V1 S:gdgrp01 C:gdgrp01 J:484501 (U:484501,0)
  Info(0x4) BackStat(0x44) bpsched
  CLIENT gdgrp01  POLICY D_gdgrp01  SCHED Default-Application-Backup  EXIT
  STATUS 96 (unable to allocate new media for backup, storage unit has none
  available)

[ 本帖最后由 huguozhi 于 2008-3-17 13:59 编辑 ]

huguozhi

谢谢两位。

请问什么叫barcode不匹配呢?
加了新带后做了Inventory 也认到了
以前96号带不够 都这么处理，没问题的

是不是通讯方面或驱动器问题？
主要是0x84代码在哪查?troubleshooting也看不到头绪

huguozhi

bperror看到的media信息:
#bperror -media看最近media情况如下(mediaid=B002L2)
03/13/2008 20:53:47 gdgrp01 gdgrp01  FREEZING media id B002L2, it is unmountable
                    and cannot be used for backups
03/13/2008 20:53:48 gdgrp01 gdgrp01  Media Manager volume pool database has no
                    more unassigned media in robotic device TLD(0)


在java界面
20:41:35  started procedd bpbrm (pid=909)
20:41:35  connecting
20:46:15 started procedd bptm(pid=919)
20:41:38 connected;connecttime 0:00:00
20:41:40 mounting B002L2
20:58:25 Error bptm(pid=919) FREEZING media id B002L2,it is unmountable and cannot be used for backups
20:58:25 Error bptm(pid=919) Media Manager volume pool databases has no more unassigned  media in robot device TLD(0)
20:53:51 end writing

unbale to allocate new media for backup,storage unit has none available(96)


96号错就是没带可用了，但mount了那么久，然后报没带
而且换了几盒新带均如此 郁闷阿

[ 本帖最后由 huguozhi 于 2008-3-14 10:40 编辑 ]

huguozhi

多谢大家 问题解决了

#ioscan -fnCtape的磁带机路径与
#tpconfig -l显示的磁带机路径不一致。

通过下面sh在MediaServer上运行解决:
#!/bin/sh
# adddrive.sh
# add NBU driver
export PATH=$PATH:/usr/openv/volmgr/bin

# delete old driver
tpconfig -l|grep drive| awk '{print "tpconfig -delete -drive " $3}'|sh

# add new driver beyond ioscan
ioscan -fnkCtape|awk 'BEGIN { iii=0} /CLAIMED.*Ultrium 2/ {iii=iii+1;getline;getline;printf( "tpconfig -add -drive -type hcart2 -path %s -asciiname HPUltrium2-SCSI%d -index %d -shared yes -drstatus UP -robot 0 -robtype tld -robdrnum %d;sleep 1\n", $4, iii,iii,iii)}'|sh

# reboot ltid
/usr/openv/volmgr/bin/stopltid
/usr/openv/volmgr/bin/ltid

huguozhi

虽然已解决  但奇怪:

MediaServer端tpconfig -l看到的与ioscan -fnCtape看到的磁带机路径不一致

数据库逻辑日志不能mount带，导致FREEZE带;
但为什么文件系统可以备份呢？