johnsot
ADSM.ORG Member
Hello All,
We just went through our yearly DR testing and it was a Disaster. The funny part was we did the test a year ago and it went just fine. I had not changed anything from the year before except that I upgraded us from version 6.2.1 to 6.2.3.0.
When we went to restore machines for some reason the restores were looking for their data in the Primary pools only. My understanding was that if they looked to the PrimaryPools first they should see the PrimaryPools as Destroyed and then go to the Copy Storage Pools for their data. However, they never looked to their CopyPools for their data. It was like their data never made it to the CopyPools. I did check to make sure that DRM did mark the PrimaryPools at DR as Destroyed and mark the CopyPools as Read-Only. Below Is my Daily Maintenance routine. Please take a look at it and let me know if you see anything that could of caused this problem in our DR testing.
1.Migration (I have auto copy setup on my DiskPools) 03:00
-----------------------------------------------
Parallel
migrate stg diskpool lo=0
migrate stg diskpool2 lo=0
migrate stg diskpool_db2logs lo=0
2.Backing Up Primary TapePools to CopyPools 06:00
------------------------------------------------
Parallel
ba stg "Onsite Primary Tape Pools to Copy Storage Pools" w=y (we have 8 Primary Tape Pools that we backup to Copy Storage Pools here)
Serial
ba db t=dbs devc=lto3 w=y
ba volhist
ba devconfig
q drm wherest=mo s=dbs f=cmd cmd="tape out &VOL" cmdf="/tsm/tapes.out"
move drm * wherest=mo tost=cour s=dbs rem=bulk toloc="Ready to go to Vault" w=y
move drm * wherest=vaultr tost=courierr s=dbs w=y
q drm wherest=courierr s=dbs f=cmd cmd="Please return &VOL" cmdf="/tsm/return.out"
q drm wherest=cour s=dbs f=cmd cmd="Please Vault &VOL" cmdf="/tsm/vault.out"
ba db dev=DBBackup type=f w=y
prepare s=dbs w=y
del volhist tod=-3 type=dbb (Here we are just deleteing any volhist out of the DBBackups that we keep on our TSM server just in case we need to do a DB restore. This way we do not need a tape we do the restore Disk to Disk.)
Please feel free to let me know of anything that you might see that could have caused the problems that I saw at our DR test. I appreciate any help you might give me. Feel free to ask me for any other information that you might need to help me figure out what went wrong.
We just went through our yearly DR testing and it was a Disaster. The funny part was we did the test a year ago and it went just fine. I had not changed anything from the year before except that I upgraded us from version 6.2.1 to 6.2.3.0.
When we went to restore machines for some reason the restores were looking for their data in the Primary pools only. My understanding was that if they looked to the PrimaryPools first they should see the PrimaryPools as Destroyed and then go to the Copy Storage Pools for their data. However, they never looked to their CopyPools for their data. It was like their data never made it to the CopyPools. I did check to make sure that DRM did mark the PrimaryPools at DR as Destroyed and mark the CopyPools as Read-Only. Below Is my Daily Maintenance routine. Please take a look at it and let me know if you see anything that could of caused this problem in our DR testing.
1.Migration (I have auto copy setup on my DiskPools) 03:00
-----------------------------------------------
Parallel
migrate stg diskpool lo=0
migrate stg diskpool2 lo=0
migrate stg diskpool_db2logs lo=0
2.Backing Up Primary TapePools to CopyPools 06:00
------------------------------------------------
Parallel
ba stg "Onsite Primary Tape Pools to Copy Storage Pools" w=y (we have 8 Primary Tape Pools that we backup to Copy Storage Pools here)
Serial
ba db t=dbs devc=lto3 w=y
ba volhist
ba devconfig
q drm wherest=mo s=dbs f=cmd cmd="tape out &VOL" cmdf="/tsm/tapes.out"
move drm * wherest=mo tost=cour s=dbs rem=bulk toloc="Ready to go to Vault" w=y
move drm * wherest=vaultr tost=courierr s=dbs w=y
q drm wherest=courierr s=dbs f=cmd cmd="Please return &VOL" cmdf="/tsm/return.out"
q drm wherest=cour s=dbs f=cmd cmd="Please Vault &VOL" cmdf="/tsm/vault.out"
ba db dev=DBBackup type=f w=y
prepare s=dbs w=y
del volhist tod=-3 type=dbb (Here we are just deleteing any volhist out of the DBBackups that we keep on our TSM server just in case we need to do a DB restore. This way we do not need a tape we do the restore Disk to Disk.)
Please feel free to let me know of anything that you might see that could have caused the problems that I saw at our DR test. I appreciate any help you might give me. Feel free to ask me for any other information that you might need to help me figure out what went wrong.