Version: 06.01.00
Vendor: IBM
CNXNAS*> storage show tape
No tapes found.
If port is changed, alias name also gets changed
storage unalias -a
storage alias mc0 WWN[xx:xxx:xxxxxx:xxxxxx][Lx]
storage alias st0 WWN[yy:yyy:yyyyyy:yyyyyy][Ly]
and to cause the filer to create the aliases via the "source" command
> source /vol/vol0/etc/tape_alias
FC card for tape and FC card for drives ( don’t confuse )
Tape for target
slot x: Fibre Channel Target Host Adapter 11a
(Dual-channel, QLogic 2312 (2342) rev. 2, 64-bit, <OFFLINED BY USER/SYSTEM>)
Firmware rev: 3.3.10
Host Port Addr: 000000
Cacheline size: 8
SRAM parity: Yes
FC Nodename: 50:0a:09:80:83:f1:45:b6 (500a098083f145b6)
FC Portname: 50:0a:09:81:83:f1:45:b6 (500a098183f145b6)
Connection: No link
I/O base 0xde00, size 0x100
memory mapped I/O base 0xa0400000, size 0x1000
Drives
slot x: FC Host Adapter 3a (Dual-channel, QLogic 2312 (2342) rev. 2, 64-bit, L-port, <OFFLINE (hard)>)
Firmware rev: 3.3.142
Host Loop Id: 0 FC Node Name: 2:000:00e08b:1c780b
Cacheline size: 16 FC Packet size: 2048
SRAM parity: Yes External GBIC: No
Link Data Rate: 1 Gbit
I/O base 0x9e00, size 0x100
memory mapped I/O base 0xa0c00000, size 0x1000
Time Not synchronizing +5 min shewed ahead
Options timed
Timed.enable on
Timed.servers ntp2.usno.navy.mil:<ip address>
Rdate <host>
Out of inodes
> df –i /vol/vol0
OR
>maxfiles <vol> - will display the number
To change
>maxfiles <vol> <max>
CPU 100 percent
*>ps –h –l ( eelll)
*> sysstat –x 1
Network traffic analysis
Download sas file and run it
http://www.apparentnetworks.com/sas/
sas -t 10.1.1.1 -s 500
NDMP
Ndmp debug 70 - this will start writing to log
/etc/log/ndmp<date> ------------ log files
Options ndmpd
Levels 0,1,2
Filer > ndmpcopy –l 0 ( minus eel )
> ndmpcopy –l 1
> ndmpcopy –l 2
0 is baseline copy
1,2 are incremental copy ( cannot go beyond 2 though )
NDMP copy from Vol to Vol ( /etc/hosts.eqiv file must have two filers information – their entries )
( best solution for data migration ; snapmirror or vol copy will cause fragmentation - filer will retain ACLs )
1. Ndmpcopy source:path_to_vol destination:path_to_volume
-level 0 –dpass
For data changed since level1
2. ndmpcopy source:path_to_vol destination:path_to_vol –level1 –dpass
Finally turn off cifs and nfs : for final incremental backup
3. ndmpcopy source:path_to_vol destination:path_to_vol –level9 -dpass
( After this level 0 is done, a level 1 ndmpcopy may be done to copy the data has changed since level 1 )
Private Note : My wife is being diagnosed as having stage III overian cancer. Please join and help if you can : http://www.facebook.com/group.php?gid=141171588041&ref=nf
Data Migration
Ndmpcopy /vol/vol0trad /vol/vol0
Vol options /vol/vol0 root
( this will also automatically set the aggregate option to root upon next reboot )
Tip:
If wrongly copied to vol – sometimes, we see vol inside vol0 and vol cannot be deleted. When accessed by \\filer\C$ - we see vol and that cannot be deleted. It says folders lost or not found. In that case, the folder can be deleted. Renaming possible, rename it and delete it.
Sync core
when cannot access by telnet, console, rsh, fier view
press reset button at back
(while physically connected console)
Ok> sync
Ok>bye
Filer reboot. Once filer comes up get core file from
/etc/crash…..
Unix commands
# tip hardwire - ---- direct access from unix/linux to filer
# cat messages | grep shelf
Snap Drive and Snap Manager for Exchange
1. Error Code : 9035 : An attempt to resize lun ‘/vol/vtape/nvlun0/lun0.un’ on filer ’10.40.3.2’failed. desription New size exceeds this lun&app geometry
Sol : size was more than 10%. Lun cannot increase more than 10% of initial size. Like if initial is 130GB then 1300 GB is the max possible.
Exchange data base restore to different location
Snap drive is used to restore the snapshot and hence database files if different location is desired
Snap manager for exchange error 249
249 : unable to delete snapshot due to busy snapshot state
So 124 -> 134-> 249-> is created.
Due to Overlapping of SME backup and snapshot
Scheduling timings. In SME , sme verification process (
7or 8’th process also fails) kb2370
Exchange Restore
1. Up to the minute – what ever have since last crash, will replay the log files automatically so database is up-to the minutes. Test back and restore, will have fundamentally no effect. If backup is done and mail is deleted, and restored instantly, the intermittent log files will be deleted and hence no effect. Basically it was within the database, so system take as it already exists and ignore, but will have all latest mail
2. Point in time – till that time ; all the backup after that date and time is not usable. Cannot get mails after that ; logs are deleted.
While restoring to Recovery Storage Group
1. Mount the previous SME shapshots – both database and log files
2. copy those to recovery storage group directory
3. if tried to restore – will get exchange error : C1041739 error
4. Copy eseutil.exe files to that directory – files are
Eseutil.exe
Ese.dll
Exchmem.dll
Exosal.dll
Jcb.dll
5. run eseutil /mh pirv1.edb
Eseutil /p pirv1.edb
6. Restore, during that time system ask to overwrite the database files, it is the option at the bottom of database properties, choose that option and restore
7. System is monted.
When IS is created in log volume
When IS is created in log volume, if it was in separate before, SME fails and reports
Evnet ID 111 and Event ID 131
( Event ID 131 is basically for cannot delete the snapshot )
VSS_E_PROVIDER_VETO