求助：使用EMC阵列进行FlashBackup的问题，也希望无牙老大可以看一下

s_p_kvw3000

最近在NBU 7.1下使用EMC阵列的快照功能进行FlashBackup备份的配置，按照官方手册的校验都通过了，快照策略校验也没有问题，但是备份过程中出现错误，错误相关的日志如下：

1. 11/26/2011 09:02:44 - begin Flash Backup: Create Snapshot
   
2. 11/26/2011 09:02:44 - started process bpbrm (pid=29195)
   
3. 11/26/2011 09:02:59 - Critical bpbrm (pid=29195) from client redhat4: FTL - snapshot creation failed, status 156
   
4. 11/26/2011 09:02:59 - Warning bpbrm (pid=29195) from client redhat4: WRN - /dev/sdb1 is not frozen
   
5. 11/26/2011 09:02:59 - Info bpfis (pid=29198) done. status: 156
   
6. 11/26/2011 09:02:59 - end Flash Backup: Create Snapshot; elapsed time 0:00:15
   
7. 11/26/2011 09:02:59 - Info bpfis (pid=0) done. status: 156: snapshot error encountered
   
8. 11/26/2011 09:02:59 - end writing
   
9. Operation Status: 156
   
10. 11/26/2011 09:02:59 - end Flash Backup: Delete Snapshot On Exit; elapsed time 0:00:00
    
11. snapshot error encountered  (156)

复制代码

所使用操作系统是RedHat 4.8 64位，EMC阵列是CX4-240，使用的快照方法是EMC_CLARiiON_Snapview_Snapshot，备份目标是/dev/sdb1，也就是EMC分配过来的LUN的路径。
就在这个系统上，我如果用Standard+快照的话备份是可以成功的，官方的兼容性列表里也有对应的操作系统和阵列型号，希望大大们可以指教一下是哪里出现了问题？

再付上bpfis的一些日志：

1. 09:02:45.006 [29198] <4> bpfis: INF - FIS_ID=redhat4_1322269364
   
2. 09:02:45.007 [29198] <4> bpfis: INF - Preparing freeze of /dev/sdb1 using snapshot method EMC_CLARiiON_Snapview_Snapshot.
   
3. 09:02:59.284 [29198] <32> bpfis: FTL - VfMS error 9; see following messages:
   
4. 09:02:59.284 [29198] <32> bpfis: FTL - Operating system error was reported
   
5. 09:02:59.284 [29198] <32> bpfis: FTL -
   
6. 09:02:59.284 [29198] <32> bpfis: FTL - VfMS error 9; see following messages:
   
7. 09:02:59.284 [29198] <32> bpfis: FTL - Operating system error was reported
   
8. 09:02:59.284 [29198] <32> bpfis: FTL -
   
9. 09:02:59.284 [29198] <4> bpfis: INF - Thawing /dev/sdb1 using snapshot method EMC_CLARiiON_Snapview_Snapshot.
   
10. 09:02:59.285 [29198] <4> bpfis: INF - do_thaw return value: 11
    
11. 09:02:59.285 [29198] <32> bpfis: FTL - VfMS error 11; see following messages:
    
12. 09:02:59.285 [29198] <32> bpfis: FTL - Fatal method error was reported
    
13. 09:02:59.285 [29198] <32> bpfis: FTL - vfm_thaw: method: EMC_CLARiiON_Snapview_Snapshot, type: FIM, function: EMC_CLARiiON_Snapview_Snapshot_deport
    
14. 09:02:59.285 [29198] <32> bpfis: FTL - VfMS method error 3; see following message:
    
15. 09:02:59.285 [29198] <32> bpfis: FTL - snapshot services: snapshot deport failed: invalid argument(s).
    
16. 09:02:59.285 [29198] <16> bpfis: FTL - snapshot creation failed, status 156
    
17. 09:02:59.285 [29198] <4> bpfis: INF - Thawing /dev/sdb1 using snapshot method EMC_CLARiiON_Snapview_Snapshot.
    
18. 09:02:59.285 [29198] <8> bpfis: WRN - /dev/sdb1 is not frozen
    
19. 09:02:59.286 [29198] <8> bpfis: WRN - snapshot delete returned status 20
    
20. 09:02:59.287 [29198] <32> bpfis: FTL - VfMS error 11; see following messages:
    
21. 09:02:59.287 [29198] <32> bpfis: FTL - Fatal method error was reported
    
22. 09:02:59.288 [29198] <32> bpfis: FTL - vfm_fi_terminate: method: VxFI, type: FIM, function: VxFI_detach
    
23. 09:02:59.288 [29198] <32> bpfis: FTL - VfMS method error 0; see following message:
    
24. 09:02:59.288 [29198] <32> bpfis: FTL - snapshot services: snapshot detach failed 1 times: invalid argument(s).
    
25. 09:02:59.288 [29198] <4> bpfis: INF - EXIT STATUS 156: snapshot error encountered
    
26. 09:02:59.530 [29220] <2> logparams: bpfis delete -nbu -id redhat4_1322269364 -bpstart_to 300 -bpend_to 300 -clnt redhat4 -S redhat4
    
27. 09:02:59.532 [29220] <2> logconnections: BPRD CONNECT FROM 129.22.0.208.33732 TO 129.22.0.208.13720 fd = 4
    
28. 09:02:59.539 [29220] <4> bpfis: INF - BACKUP START 29220
    
29. 09:02:59.539 [29220] <32> bpfis: FTL - cannot open /usr/openv/netbackup/online_util/fi_cntl/bpfis.fim.redhat4_1322269364.0
    
30. 09:02:59.539 [29220] <4> bpfis: INF - EXIT STATUS 0: the requested operation was successfully completed

复制代码

C字裤男爵

snapshot services: snapshot deport failed: invalid argument(s).

/opt/Navisphere/bin/naviseccli –h CLARiiON_hostname -user array_admin_username -password array_admin_password -scope 0 getagent
看一下。

再确认一下cx4-240兼容不？
http://www.symantec.com/docs/TECH126901

这个有操作系统日志，查看一下。
再看一下EMC的日志。

尝试使用EMC navisphere CLI 做一下snapshot看可以成功不。

s_p_kvw3000

/opt/Navisphere/bin/naviseccli –h CLARiiON_hostname -user array_admin_username -password array_admin_password -scope 0 getagent
这个命令可以得到阵列的信息，没有问题

阵列FLARE版本是04.28，使用的host agent和cli版本是6.28
而且，如果不是用flashbackup策略的话，standard策略里面调用快照是可以成功的

无牙

你设置reserved LUN pool 了吗？

s_p_kvw3000

阵列上已经为每个SP都设置了reserved LUN pool

而且现在我又发现一个很奇怪的现象。如果将卷mount上，则会在创建快照的时候报错，错误类似于/dev/sdb1 is not frozen；如果将卷umount掉，则创建快照可以成功，阵列上可以看到新的快照被分配到客户端，操作系统上也可以看到新增了一个硬盘sdc，但是备份会卡在begin writing上长时间（已经5个小时了）没有动静。

无牙

你先升级到7.1.0.2试一试。

s_p_kvw3000

EMC CX4-240没有折腾出来，把环境换成了HP EVA 4400和SUSE 11 x86_64。查过最新的NetBackup 7 Snapshot Client Compatibility List，阵列型号和操作系统都在列表里面，文件系统也是ext3，按照文档配置了flashbackup策略，结果这次连策略校验都通不过了。

这次bpfis.log的错误是这样的：

1. 09:09:52.963 [21043] <4> bpfis: INF - FIS_ID=1323306592
   
2. 09:09:52.963 [21043] <4> bpfis: INF - Testing freeze of /dev/sdb using snapshot method HP_EVA_Snapshot.
   
3. 09:09:54.370 [21043] <32> bpfis: FTL - VfMS error 11; see following messages:
   
4. 09:09:54.418 [21043] <32> bpfis: FTL - Fatal method error was reported
   
5. 09:09:54.418 [21043] <32> bpfis: FTL - vfm_freeze: method: HP_EVA_Snapshot, type: FIM, function: HP_EVA_Snapshot_vfm_freeze_commit
   
6. 09:09:54.418 [21043] <32> bpfis: FTL - VfMS method error 1060; see following message:
   
7. 09:09:54.418 [21043] <32> bpfis: FTL - snapshot services: snapshot method analysis failed: no combos generated: stack elements not capable of any split or quiesce.
   
8. 09:09:54.418 [21043] <32> bpfis: FTL - VfMS error 11; see following messages:
   
9. 09:09:54.418 [21043] <32> bpfis: FTL - Fatal method error was reported
   
10. 09:09:54.418 [21043] <32> bpfis: FTL - vfm_freeze: method: HP_EVA_Snapshot, type: FIM, function: HP_EVA_Snapshot_vfm_freeze_commit
    
11. 09:09:54.418 [21043] <32> bpfis: FTL - VfMS method error 1060; see following message:
    
12. 09:09:54.418 [21043] <32> bpfis: FTL - snapshot services: snapshot method analysis failed: no combos generated: stack elements not capable of any split or quiesce.
    
13. 09:09:54.418 [21043] <16> bpfis: FTL - snapshot verify failed, status 156
    
14. 09:09:54.419 [21043] <4> bpfis: INF - EXIT STATUS 156: snapshot error encountered

复制代码

按照官方文档的说法，这个错误是由于在策略中指定的源设备或目标设备有误或者所选磁盘阵列方法的类型有误。但是EVA的快照配置并不需要指定目标设备，阵列型号也确认没有问题。

不知道有没有人使用过NBU基于磁盘阵列的快照功能啊，什么阵列的都行，有没有什么经验可以传授一下

无牙

晕了

maoj2008

EMC的测试过DMX的盘阵，CLARiiON的没有试过。不过HP EVA的应该比较好操作，你在bpfis log中搜索一下sssu，看一下这个有没有相关报错，安装的位置是否正确。如果sssu没有问题的话，Array不用额外的配置。

s_p_kvw3000

回复 9# maoj2008


可以帮忙看看是不是我sssu安装有问题。我从官网下载的sssu 4.0、sssu 5.0以及随机光盘上的sssu 9.3都只有一个可执行文件，这个文件点开后直接就进入了sssu的命令行，也没有安装的选项。我就按照《Symantec NetBackup ™ Snapshot Client 管理指南》里测试客户端端到HP阵列连通性里写的路径，将这个可执行文件放到了/opt/hp/sssu/目录下，并改名成了sssu_sunos。我也发现了，在NBU备份过程中，根本就没有调用到/opt/hp/sssu/sssu_sunos这个程序，bpfis的日志里也没有出现sssu，bpfis日志目录下的hpevafi.log里也是空的