Unsolved

This post is more than 5 years old

46 Posts

5832

April 24th, 2008 04:00

"client name" is not a registered client

Hi Guys,

I am experiencing some sporadic issues with my Legato NetWorker environment.

We are currently running Legato 7.1.4 build 577 Network Edition on a Win 2003 SP1 enterprise edition OS.

The backups work OK most of the time, however there will be instances where we will see failures accross the datazone with the following error message:

"client name" is not a registered client

At first I thought it was a problem closely related to DNS, however the sporadic failures still occur if hosts tables are updated with the correct information on both the clients and the NetWorker server.

The unusual this is that the NetWorker server will not back itself up either, failing on "client name" is not a registered client..

The issue is usually solved by a restart of the NetWorker services on the server.

I am now leaning towards the theory that a 3rd party application maybe locking the /res files causing NetWorker to not recognise any of the clients, however I cannot quite pin it down.

Has anyone experienced any similar issues?

Thanks in adavnce
dd1980

46 Posts

April 24th, 2008 08:00

I am just in the process of finding out now, however I would assume that all server and clients are on static IPs.

It is also worth mentioning the fact that if I am working on the GUI on the NetWorker server, after around 5-10 minutes the GUI will disconnect from the server itself, as if the connection is dropping. However nothing is logged in the event viewer.

NetWorker 7.1.4 was also unistalled and re-installed just in case the application had become corrupt.

198 Posts

April 24th, 2008 08:00

Hi,


Sounds more like something is "messing" with the client host name / IP address .. are these DHCP allocated details ?

IKP

46 Posts

April 24th, 2008 08:00

That is the plan for next week, we are going to go to version 7.4.1

253 Posts

April 24th, 2008 08:00

Sorry I haven't experienced any similar issue but I think it would be a good idea updating the NetWorker Server version to check if the same problem persists.

198 Posts

April 24th, 2008 09:00

Hi,


Have you tried running a constant ping for and see what happens ... (on Windows use ping -t) and just seeing if anything happens.

Also, what else is running on your Network ... could someone else be playing with firewalls etc ?

Have a look at the Windows device drivers ... are you using some form on teaming / bonding driver which could be tripping between multiple nics - seen this before where the driver is unable to handle volumes of network traffic associated with backup ops... this casues the driver to "re-locate" the IP address to the oterh card and fail the backup... but this was sometimee ago with suspect device driver.




IKP

Message was edited by:
IKP

Message was edited by:
IKP

46 Posts

April 24th, 2008 09:00

Thanks very much for the response...

The server does have multiple Network cards and teaming is involved, I am in the process and looking at the device driver for it. I did suspect the fact that NIC teaming could be responsible for the backup failures.

The clients that fail are both behind and within the firewall.

The thing that I find strange is that 4 days of the week the backups will succeed whilst the other 3 they will fail.

Like I said I will look into NIC teaming and see if I can spot any abnormalities.

455 Posts

April 24th, 2008 09:00

In this case, have your DNS administrator check the DNS entry for the server and clients. If you have multiple NICs ensure that just one NIC has DNS auto registering/update set. Update you host file with the correct IP address for your server and clients.

I think you may have some kind of connectivity issue.

46 Posts

April 25th, 2008 04:00

Thanks for all your help guys.

We have discovered that the drivers we use on our NICs have a history of dropping packets when failing over.

What we are going to try is to run on a single NIC for 1 week and see what results we get, after that we will upgrade the drivers and see how we get on.

Thanks
dd1980

20 Posts

April 25th, 2008 10:00

On my experience - communication errors (like DNS resolution, network connectivity etc) produce logs like:
save: RPC error: Unknown host
save: Cannot open save session with XXXXX

Your log appears to have something to do to NW unable to read clients' database. We are running 7.1.1 (on Solaris though) and I've never seen a problem like yours. I would first try to upgrade and see if it changes anything.

Do you have a definitive pattern for the errors or they occur randomly? In the moment of failed backup - can you access any info about that particular client, like client setup, indexes?

46 Posts

April 28th, 2008 09:00

Hey Guys,

the saga continues...

We ran Monthly Full backups at the week-end and we received the same failures. Going through the daemon.log I am seeing the follwoing errors:

Lost connection to Media DatabaseRPC Failure: Unable to receive; errno = Connection reset by peer

savegrp: Failed to update server, aborting Savegroup.

savegrp: Cannot query the pool resources. Unable to verify the save sets on the media.

savegrp: lost 'NSR group' resource

nsrindexd: RPC error: Unable to send

device resource lookup fails

save: File index error: `xxxxxxxxxxxxxx' is not a registered client

This to me looks like the server is loosing sight of its databases therefore loosing all of its "knowledge" of what it has to do next.

Looking through the Event viewer nothing stands out that could lead to an installation or OS problem.

Anyone experienced any of those errors before?

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

April 28th, 2008 12:00

Not really...

46 Posts

April 28th, 2008 12:00

I thought that may be the case, but then surely something would be flagged in the event viewer about the nsrd "hanging".

6 Operator

 • 

14.4K Posts

 • 

56.2K Points

April 28th, 2008 12:00

Perhaps your server is a bit overloaded for all operations running at that time?

46 Posts

April 29th, 2008 02:00

Ah OK,

The thing is we have 2 sites, 1 in the US and 1 in the UK. The 1 in the UK works without any problems at all, the 1 in the US is having the issues. Both servers are purely dedicated to NetWorker and they are both identical. The only difference is that the one in the US has more RAM.

The interesting this is that the US data zone has less data to handle than the UK.

I am going to upgrade the site 7.4.1 next week and then see how we go from there.

198 Posts

April 29th, 2008 05:00

If you are going to upgrade to 7.4 then go to 7.4SP2 and NOT SP1.


IKP
No Events found!

Top