Having problems with your account or logging in?
A lot of changes are happening in the community right now. Some may affect you. READ MORE HERE
Highlighted
ianbutton1 Respected Contributor.
Respected Contributor.
270 views

Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

We upgraded DP from V9.09 to 10.30 on Friday. All Windows backups are working OK but only 1 Solaris client is backing up. All other Solaris backups fail with error message
[Critical] From: BSM@dpcm1.cmbc.internal "SolarisRoot" Time: 10/05/2019 21:00:02
[61:12500] Cannot connect to inet for getting filesystem list on host"deanheaddev.cmbc.internal".

The Solaris client that backs up OK is no longer in the CM’s config file – all others are still in config. I have not done anything different on this client.
In DP Manager GUI, on Clients screen, if I try to access Patch list for Solaris clients I get message “Could not load data from the client”.
In Backup screen, if I try to access the filelist for Solaris clients, I get “Error 12:3039 IPC Protocol negotiation error”
Solaris clients are all on DP V9.00. Before upgrade, Solaris clients did not have ECC configured. Windows clients had ECC enabled, but it was disabled for the upgrade and has not been enabled.
Telnet on port 5555 (old DP port) from CM to the “good” Solaris client is working OK, and it also works OK from CM to the Solaris clients that are not backing up. Telnet on port 5565 is not working.
Do I need to run the command “omnicc -secure_comm –configure_peer CMname –accept_host” on all Solaris clients, in order to issue/accept certificates? I have not run it on Windows clients, and everything is OK with them. Is this command option available on V9.00 clients, or do I need to upgrade the clients manually?

I have tried "omnicc -secure_comm -configure_exception clientname" & retried my test job, but no better.

I understand from the documentation that retaining port 5555 is OK, and that old 9.00 clients should work with DP 10.30 on the Cell Manager.

Any ideas?  I feel the problem must be either that CM is trying to use 5565, or secure communication isn't working owing to the old client software (but if either, why does 1 client work OK?).

TIA

0 Likes
1 Solution

Accepted Solutions
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

You should try to restart the Windows CM tomorrow to see if this clears the issue with the hanging omniinetpasswd. I have never seen this before. If it does not clear the issue try running omniinetpasswd -clean -debug 1-200 and share the last 50-100 lines it produces.

Have you considered building a UNIX Installation server? You can have it on CentOS/RHEL/OL 6/7 or SLES 11/12. It does not consume much resources and would make your life easier maintaining those Solaris and AIX clients.

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
17 Replies
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

Upgrading only the Cell Manager should add SSC (secure socket communication) exceptions to all clients not upgraded to 10.x. So you should see entries in the ssconfig file of the Cell Manager for the Solaris clients, too. If the 9.x client is no longer part of the cell (not in cell_info anymore) you can import it back using omnicc -import_openvms_host Hostname. This will add the SSC exception to ssconfig and import the client to cell_info. If the client is is cell_info but nothing in ssconfig you can run omnicc -secure_comm -configure_exception Hostname to bring it to the expected state.

All of this is using the same port on the Cell Manager. If it was 5555 before it is 5555 after the upgrade.

What Solaris release are you running. Is it supported for a upgrade to 10.30?

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Thanks @Sebastian.Koehler.  All Solaris clients are V11.3 - pretty recent so I assume supported.

cell_info file has all Windows & Solaris clients in it, so -import_openVMS_host doesn't add it again

ssconfig file has all Windows & Solaris clients in it, so -configure_exception ... -overwrite works but doesn't change anything.

Maybe I should edit these files to remove the Solaris entries & try re-adding them via the omnic commands.

Any other possibilities?

Thanks again.

 

 

0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

I have removed one of the Solaris clients from cell_info, config and ssconfig files, and tried the command omnicc -import_VMS_host clientname

But I get [12:1625] Import host failed.

I think something has gone wrong with the SSC part of the upgrade, perhapss after the one (working) Solaris client was done.

Any ideas, anyone?

I have opened a case, I just thought there might be a simple answer before I get debugs and upload them.

Thanks

0 Likes
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

Have you tried to upgrade one of the clients to A.10.30 to see if the issue is resolved by creating a full SSC trust with the client?

Can you please confirm that the name resolution works as expected on the Solaris clients? Check the Cell Manager and Clients debug.log and inet.log file for related errors.

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

1.PNG2.PNG

 

Hi @Sebastian.Koehler 

We're trying this - but the tar files in the 2019.02 .gz download files for Unix & Linux don't seem to have any software for Sun Solaris Sparc processor.  Our machines are Solaris V11 (supported for Disk Agent).

Have we missed something? 

Thanks

Ian

 

0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Ah, the Unix tar seemed to work OK.

omnisetup.sh on client has worked, but omnicc -secure_comm -configure_peer cmname -accept_host has failed with "Connection refused".

It seems that my colleague let the Solaris install use the new default port 5565.

How can we change it to 5555?  Where is the file containing the port number?

Nearly there . . .

Thanks

Ian

 

0 Likes
Super Contributor.. Aleksandr_Sviridov Super Contributor..
Super Contributor..

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

>How can we change it to 5555?  Where is the file containing the port number?

Run this command:

/opt/omni/bin/omnicc -update_local_port 5555
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

@Aleksandr_Sviridov
Большое спасибо - Thank you very much!

10.30 installation now complete, client imported and test backup is working.

We won't worry about why all the other clients (except 1) didn't get upgraded automatically - we'll do the manual upgrade.

Thanks Sebastian & Aleksandr for your advice!

Ian

0 Likes
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

For the local setup you can instruct omnisetup.sh to use the INET port you need with the -inetport parameter. Here is an example.

./omnisetup.sh -install da,ma -inetport 5555

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
0 Likes
Super Contributor.. Aleksandr_Sviridov Super Contributor..
Super Contributor..

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi Sebastian!

I have another question: what distributive should I use for local install for solaris\AIX?

0 Likes
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

According to the docs you can use the Linux or HP-UX media kit.

https://docs.microfocus.com/itom/Data_Protector:2019.02/LocalInstallUNIXMacOSX

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hmmm - it gets more interesting.

The omniinetpasswd database seems to have got lost/corrupted - when I try omniinetpasswd -list on the CM, the task just hangs.

I wonder if that explains why client upgrades stopped after all Windows and the first Solaris client.  And I have some GUI clients to upgrade for IT colleagues but GUI upgrade fails "could not connect to Service Control Manager"

There is a -clean option for omniinetpasswd, but if I use it, will backups fail?  I don't know what users/passwords I need to add back in.  Was there a file somewhere? (I have a backup of all the old OmniBack folder on the CM.)  Or do I need to add the logon accounts used for the DP services on the CM, and the accounts used for running the scheduled backup jobs. And anything else?

The Users page in the GUI is OK, and backup jobs are working OK.

Suggestions please . . .

TIA

0 Likes
Knowledge Partner Knowledge Partner
Knowledge Partner

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @ianbutton1,

You can clean it and add whatever is required. Backups usually not depend on this with the exception of the IDB backup on a Windows Cell Manager. The values are stored in the Windows registry on the Cell Manager/or client only. Usually there is only one or two entries inside with one marked for Installation Server. Make sure to run the omniinetpasswd command from an elevated command prompt or you will get inconsistent results.

You know that you can remotely upgrade Windows and UNIX clients if you have upgraded the installation servers?

https://docs.microfocus.com/itom/Data_Protector:2019.02/Upgradeto1000

Regards,
Sebastian Koehler

---
Please use the Like button below, if you find this post useful.
0 Likes
ianbutton1 Respected Contributor.
Respected Contributor.

Re: Solaris V9.00 clients backup fail after CM upgrade to 10.30

Jump to solution

Hi @Sebastian.Koehler 

No, -clean option freezes the task as well.  I killed it with Ctl-C and tried -add but that froze the task too - Ctl-C again

I ran a small test backup and that is still working, so hopefully all jobs will work this weekend and tonight.

I am running the command from an elevated command prompt.

The old case about Solaris clients is still open, so I have added this problem onto it, and we will see what support suggest.

The IS is the same as the CM (a Windows machine) so it didn't have the Unix depot files for remote install.  Local installs on Solaris clients have worked now, though.

I think the failed upgrade and omniinetpwd problem may be connected (corrupted install_admin credentials).

Thanks for your advice anyway.

 

0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.