Hi,
I have same problem and I followed the procedure mentioned above, but the store goes into "error" state and I can't even start or stop it. Could you share any other methods or hints to solve this issue. Thanks in advance.
P.S. Sorry for my bad English.
Best regards,
I'm not sure how you verified the username and password were correct, but, what I usually recommend first is to make sure that your login string is correct outside of Data Protector
On the Oracle server, run
sqlplus /nolog
connect username/password@service as sysdba
Of course, you would use the Username and password that you are trying to connect with and 'service' can be found in the 'tnsnames.ora' file, although, more often than not, it is the name of the instance
If this doesn't connect, then the problem is on the Oracle side
If it does connect, I suggest that you redo the Integration, as the Configuration file may have become corrupt. This can be done either from the GUI in the Backup -> Oracle context, right-clicking on the Oracle server, and pick Configure, or, from the command prompt on the Oracle server
You didn't specify what version of DP, so I can't be more specific, but, the procedure for either method is in the Oracle Integration Guide
I don't understand what is meant by this:
"after that i tired to add driver. but counld not found driver"
Data Protector has nothing to do with Drivers. Could it be that you meant that you tried to add Drives?
Great, I am so glad to hear this... I have found the the '-detail' information helps me an awful lot when trying to do things from the command line
Thank you for the document.
My scenario was : Server1 - cell manager, Server2- HP storeeasy storage. I wanted to backup to disk over the network to Server2. One way was creating a iSCSI disk within Server2 and then mounting it to Server1. I thought that was cumbersome.
However, there was a line in the document that mentioned that you would need to install StoreOnce component on the Server where you would want to store the backup. So, I did that and that populated the Server2 as the deduplication system. I was able to give the local path of Server2 volume/lun after that.
I guess my second last question would be what gateway to use. There are two gateways for this device that I can use - Server1 and server2 gateways. If I choose Server1 gateway then Server1 MA will be writing to server2, correct? Meaning, there will be less load on server2 during backup operation?
A StoreOnce gateway is just a media agent. If the Media Agent (Gateway) is on the same server being backed up, I would definately be using that so that backups are local, and don't go over the LAN
If you are backing up a diskl agent that is not either of the Gateways, it really doesn't matter which one you use
I will be backing up various clients to Server2 (HP storeeasy storage). But the cell manager is Server1. Just so that I understand it clearly, I am to choose Server2 as gateway when setting up backups from Sever1 (cell manager)?
Also, Server2 has two teams of NIC. One for frontend VLAN and the other for backup (similar for Server1). Is there a way to specify that they use the backup VLAN for backups, I mean the traffice between Server1 and Server2, plus the clients that will be backed up to the storage on server2?
I appreciate your help!
"I am to choose Server2 as gateway when setting up backups from Sever1 (cell manager)?"
Not necessary, you can use either media agent as a Gateway
==============
The VLAN should have it own unique hostnames and IP addresses. When creating a device, use the VLAN hostname
HI,
Im about to make a Projekt with DP for my trainee finals.
Now i try to get skill with dp but nothing realy works.
i use VirtualBox an have 1 Centos VM an a Windows7 VM.
Both have a static IP and can ping eachother. I Install the Cell Manager on the CentOS Machine
./omnisetup.sh -CM
On the Windows Machine i Install the Client. Now i want to connect to the cellmanager from the Windows Client, but it dont works.
Check if the Cell Manager host is up and running. Check if it is accessible on the network (use ping command). If not, check the configuration of network services and routing (use nslookup command).
I start xinetd on the Cellmanager Host and set the 5555 Port. But i think the manager is not Running. I read a lot in this Forum but every command like omnisv -?? gave me the respond "command not found".
update:
When i try to start the omnisv.sh i get the error
1053 cannot start uiproxy
I stuck now for 2 Days, would be great if somone have a idea.
Prior to v8.x when you needed to inslall a linux/hpux agent ( Cell Manager is Windows), you needed to run the java gui from a linux IS server, and also have trusted SSH configured.
running xomni on Linux IS server gave a java based UI, from there you could install a linux/unix client.
Now with version 8.1 I cannot find the java GUI install for my LInux installation server.
THere is no java component to be installed. ( thought it was replaced by "cc" ). looking in /opt/omni/bin, there is no xomni or anything that looks like a UI.
omnisetup.sh [-source directory] [-server name] [-install component-list ] [-IS] [-CM] [-bundleadd bundlenumber] [-bundlerem bundlenumber]
component-list: cc da ndmp ma sap sapdb saphana emc oracle8 sybase ssea informix lotus db2 smisa vmware vmwaregre_agent vepa StoreOnceSoftware autodr docs jpn_ls fra_ls chs_ls
Anyone have any insight to this ?
I spend hours today troubleshooting and found a solution. I wanted to share this for anyone else that may get snagged on this.
Scenario:
DP v8.1
Windows Cell Mananager
Linu base Installation server
HPUX 11iv3 agent with DA and MA
I was running into issues upgrading the HPUX agent from DP v7.01 to v8.1 due to continuous time-out errors
-Pushing the agent though the IS server would install the software but fail during the import
-manual install of the agent worked with no issues but could not import into the CM
-/var/opt/omni/log/inet.log would report success for all steps except at the end a time-out during the import
-I manually added the hpux server info into the cell_info file on the CM but when attempting to configure a test backup of the HPUX client, it just sat and spun during the filesystem browse.
During the sit and spin session I looked into the /var/opt/omni/log/debug.log file and found a number of ipw_getaddrinfo(.....) Network performance warning messages.
found this post http://h30507.www3.hp.com/t5/Technical-Support-Services-Blog/DPTIPS-Data-Protector-IPv6-and-the-Unintended-Consequence/ba-p/109643
the cause... looks like DP8.1 wants to do ipv6. My server has no ipv6 entries in the nssswitch.conf file so it was going after DNS for ipv6, there is no ipv6 in my environment( nor most other people's for that matter), so it was timing out..
the solution... (read the blog entry), add an "ipnodes" entry to the nsswitch.conf file and problem solved ( after 3 hours of troubleshooting ).
Hope this helps someone else if they run into the same issue.
Cheers
I just can't work out the best way to offload backup from our D2D to tape.
Setup:
The tapedrive is an MSL2024 with 2 x LTO-5 SAS drives connected via an SA P411 ctrl. on a DL380 G7 server.
I've changed the IP of the server withing same range, failed
I've done clean reinstallation of DP, failed
I've verified DNS correct, still failing
I've tried manually import from DP, again failed
Servers from the same range, can telnet to the Failed server on 5555
Though from Cell server it's always failing to telnet to the Failed server on 5555
I'm not sure where I am heading now..
Anyone can advice?
Hi,
tape-offload in Data Protector is always done using a copy job. But in your case you can create a backup specification using "network share backup" to backup the files from CIFS share. I recommend to use the VTL mode, where you create some tape drives/library to backup the files to.
Best regards
Daniel
Hi,
with DP 8.XX the Java-GUI became obsolete and is no longer delivered. So in your case use the Windows GUI instead. The push installation will work the same for you Linux based clients. Just import the installation server, enable ssh in omnirc on your Linux installation server and exchange the keys between your IS and clients. After that you are able to diustribute the Linux based clients the same way, as it was with the Java GUI.
Best regards
Daniel
Hi,
as far as I know CentOS is not supported as Cell Manager and Disk Agent only. However, when omnisv gives unknown command, it seems your search pathes are not expanded. Add /opt/omni/bin, /opt/omni/sbin and /opt/omni/lbin to global path, restart your terminal session and try again.
Best regards
Daniel
Hi,
the error you posted is not a Data Protector error, it is the result given by RMAN. So, with the credentials configured in Data Protector, RMAN sees a wrong username or password. Try to reconfigure and make sure the sqlplus stuff Bob told is working.
Best regards
Daniel
First of all, Thank you for your reply.
i'm sorry my English expression.
"after that i tired to add driver. but counld not found driver."
=> please refer to attached a figure.
that mean is no data devices detected.
please help me.
I am able to boot from ISO image.
While following the DR wizard I get below error:
Cannot copy files from X:\Omniback\Temp to X:\Windows\system32\bin\omni.
Not enough space on destination folder!