Error 83 : Open Media Error

[lanira_1509]

Hi Everybody,

I reiterate my issue concerning the title.
Last time, I did not sent any log because I expected that our partner would have solved the issue. Well it is not.

Before I introduce the configuration of the NB :

Master Server = Win 2008 R2 NB 7.0.1

Media Server (where the issue is) = Win 2003 R2

I attached a file wich contains 3 sheet (3 client with the 83 status error code)

We escalated it to the support but it proves to be that they had to escalate to the maker.

The maker feeback's is that all the client return the following error = "image open failed: error 2060017:system call failed"

Thus, they recommend to create a touch file on the Media Server named SIZE_DATA_BUFFERS_DISK and to set up the value to 327680

We did it BUT, the problem is still here....

Anybody knows how to resolve this problem permanently ?

Thank you for your help !

Lala.

[Michael_G_Ander]

If you are running backup to a dedup pool, the problem can be the number of streams going in and out of the pool at the same time. I have solved status 83 on dedup pools by limiting the IO streams

As your media server is Windows 2003 you also need to look at the memory to see if there is anything getting exhausted e.g. pool, non-pool, heap

[RamNagalla]

I am seeing the backup destination as F:\Backup for 2 clients and E:\Backup for one client.

is there any other clients going to this same destinations...? if yes are they getting successfulll or failures?

[lanira_1509]

Hi Michael,

Thank you very much for your prompt reply.

You know I am a beginner in this environment and I apoligize in advance.

Concretely, when you say limiting the IO streams, what should I do ?

As your media server is Windows 2003 you also need to look at the memory to see if there is anything getting exhausted e.g. pool, non-pool, heap

==> concretely, what do you mean ?

Thank you very much in advance !

[lanira_1509]

To Nagalla,

Yes there are other clients and the result is successfull.

[RamNagalla]

try to run the backup for only these 3 clients to this storage unit.. and make sure no other jobs are running on this srorage unit , for testing purpose.

only 3 clients jobs active for testing.. it will make sure if this is due to the load on the disk or not... 

and what type of storage unit it is.. does it basic disk or advanced disk , msdp?

[lanira_1509]

@Nagalla,

OK, I will do it tomorrow.

The type of storage unit is basic disk.
 

[lanira_1509]

@Nagalla,

Hello,

Today, (very weird) there was no failed jobs...
I will keep this discussion and see on Monday.

I will keep you informed.

Regards,

Lala

[lanira_1509]

@Nagalla,

Today, I tried to rerun the job, all 3, which are on the same storage disk, failed with the same status 83.

I do not know what to do...

[lanira_1509]

Please find a example (same for the 2) of the log.

[RamNagalla]

does this basic disk is from the NFS mount? if yes please have a look into below T/N

http://www.symantec.com/business/support/index?page=content&pmv=print&impressions=&viewlocale=&id=S:TECH54767

to make sure this this is not specific to storage unit issue... pleaset try to take the backup of these 3 clients to another storage unit DIsk or tape and see how its working..

if the clients are getting successfull in another storage unit..

provide the output of  failed storage unit

1) bpstulist -label <storageunit> -U

2) also enable the bptm log in media server 

--> make sure bptm directory is created in /usr/openv/netbackup/logs/bptm (unix media server)

installpath/veritas/netbackup/logs/bptm (windows media server)

and keep the verbose to 5 and test the backups again.

collect the bptm log and attach it to this post...

[lanira_1509]

Hi Nagalla,

Below my responses :

does this basic disk is from the NFS mount? if yes please have a look into below T/N

http://www.symantec.com/business/support/index?pag...
==> As I said, and I apologize for this, I am beginer . How can I check if this basic disk is from the NFS mount ?

to make sure this this is not specific to storage unit issue... pleaset try to take the backup of these 3 clients to another storage unit DIsk or tape and see how its working..
==> I already tried to take backup to another storage unit Disk and it works.

if the clients are getting successfull in another storage unit..

provide the output of  failed storage unit

1) bpstulist -label <storageunit> -U
==> See Attached file

2) also enable the bptm log in media server 

--> make sure bptm directory is created in /usr/openv/netbackup/logs/bptm (unix media server)

installpath/veritas/netbackup/logs/bptm (windows media server)

and keep the verbose to 5 and test the backups again.

collect the bptm log and attach it to this post...
==> for these log, I have to wait tomorrow because other jobs are running. And the only period I can do that is between 11:00 AM and 14:00.

[RamNagalla]

Path:                 "F:\Backup"

it looks like its not the NFS/CIFS share... 

keep the Verbose 5 and provide the bptm logs by running the job again.. and also attach the detail status of the failed job that you are going to trigger tomorrow.

[lanira_1509]

it looks like its not the NFS/CIFS share...
==> If it was, which pattern should we got ?

keep the Verbose 5 and provide the bptm logs by running the job again.. and also attach the detail status of the failed job that you are going to trigger tomorrow.
==> Yes, I will keep you informed about the progress.

Thank you very much for your help.

[Marianne]

If the storage unit was a CIFS share, it would have a path like this:

\\servername\path 

Please have a look at suggestions from Mark Solutions in this post:

https://www-secure.symantec.com/connect/forums/timeout-netbackup-job-manager-service 

Although the topic is different, we often see errors like yours on a 'very busy server'.

Also see Mark's recommendation for Windows servers in this post:

https://www-secure.symantec.com/connect/forums/sometimes-jobs-fail-status-84-83#comment-6300681

[Mark_Solutions]

Thanks Marianne for the reccomendations here!!

I am thinking something is going quite wrong here (maybe obvious!) so could you tell me a few more things about the Media Server please:

1. What is the disk size and free space on the F drive

2. How much memory does the server have (RAM)

3. Can you export the System and Application event logs for the server (as evtx if possible) and post them on here for me to look at.

4. What is the network speed of the server

I think either something is crashing or the server runs out of resources - just need to identify which and then what needs to be doen to correct it.

Thanks

[lanira_1509]

@Marianne,

Thank you very much for your reply.
I will take a good look on Mark's documentation and give him the information.

[lanira_1509]

Hi Mark,

Thank you very much for your reply.

Just to say that today, there is no error.. very strange. It is very uncertain...

Below my answer to your reponse :

1. What is the disk size and free space on the F drive
==> The disk size is 1.95TB and the free space is at this time 1.08TB

2. How much memory does the server have (RAM)
==>The RAM of the Media Server is 8GB

3. Can you export the System and Application event logs for the server (as evtx if possible) and post them on here for me to look at.
==> See the attached files. It is not evtx but evt. (Runinng on 2003 Server)

4. What is the network speed of the server

==> the NW speed is 4.0Gbps

I think either something is crashing or the server runs out of resources - just need to identify which and then what needs to be doen to correct it.

Thank you very much.

Lala/

[lanira_1509]

@Nagalla,

Sorry, Yesterday, I could not get the job running.
Another maintenance was scheduled...

And today, there was no error on the jobs....

[RamNagalla]

 today there is no failures..so its an intermittent failures...

as i suggested in my initial posts... did you try running only those 3 client jobs to the Media server STU.. no other jobs..?

check the possiblity to schedule the jobs in differnt time when it has less jobs moving... if you see the failures again...