cancel
Showing results for 
Search instead for 
Did you mean: 

NetBackup 5200/5220 Appliances 2.5.1 is now available!

CRZ
Level 6
Employee Accredited Certified

I'm very pleased to announce that the next Maintenance Release for NetBackup 7.5 and Release Update for NetBackup 52x0 Appliances is now available!

NetBackup 7.5.0.4 is the fourth maintenance release for NetBackup 7.5.  It is a cumulative release containing fixes and content from 7.5.0.1 through 7.5.0.3. In addition, this release contains an additional 315 fixes (bringing the total fixes in 7.5.0.4 to over 700!) including the most commonly downloaded EEBs, several customer escalations, and internal engineering defects.

The simultaneously released NetBackup 5200/5220 Appliances 2.5.1 Release Update is the equivalent Appliance patch release to NetBackup 7.5.0.4.

The significant content of NetBackup 7.5.0.4 is:  

  • No new features in 7.5.0.4.  However, some features that were delivered in previous maintenance releases are included:
    • Solaris 11 master server, Bare Metal Restore server and OpsCenter Managed Server support – delivered in 7.5.0.3
    • NetBackup for VMware on SLES – delivered in 7.5.0.2
    • PureDisk seeding utility – delivered in 7.5.0.2
    • Client Direct Whole Image Mode – delivered in 7.1.0.4
    • Multi-process nbostpxy for Client Dedupe – delivered in 7.1.0.4
  • Fixes included in EVERY publicly available 7.5.0.3 hotfix are included in 7.5.0.4
  • Other miscellaneous fixes across NetBackup which were not publicly released

Additional content applicable to Appliances 2.5.1 is:

  • Improved agent for proactive support
  • Support for OST 3rd party plug-ins

Information about 2.5.1 is available here:

NetBackup 5200/5220 2.5.1 Update
 http://symantec.com/docs/TECH197011

The 2.5.1 Update can only be applied to an Appliance already running at version 2.5.

To check to see if your particular Etrack is resolved in 2.5.1 (NetBackup 7.5.0.4), please refer to both sets of Release Notes:

NetBackup 7.5.0.4 Release Notes
 http://symantec.com/docs/DOC5514
 
NetBackup Appliance 2.5.1 Release Notes for NetBackup 52xx
 http://symantec.com/docs/DOC5797

46 REPLIES 46

Mark_Solutions
Level 6
Partner Accredited Certified

Anyone had issues with the download?

Downloaded and joined today but won't extract as it says it is corrupt / incomplete

(Doing this on Windows with the copy command - correctly I believe!)

 

SRP
Level 4

Mark

No problem downloading files,(had to download the 1st one a couple of times), produced the single file, checksum ok.

Windows environment.

 

Sean

Mark_Solutions
Level 6
Partner Accredited Certified

Thanks - re-downloaded and all OK now - have applied it - will be writing a blog about these upgrades when i get a chance as a LOT of things have happened due to the 2.5 upgrade which have needed a lot of fixing!

Stanleyj
Level 6

Mark,

What fixes have you noticed so far? 

I applied the update and something i noticed that needs fixing is the fact that you can't connect to the appliance using something like winscp anymore.  They recommended is openeing a log share now but windows explorer wont connect to that either.  Atleast for me.  This is causing me an issue becuase i cant send support logs to fix a spoold issue.

Do you know of another way to get access to the file system?

Mark_Solutions
Level 6
Partner Accredited Certified

What I did notice with the logs share - and no sure why i actually tired it - but to connect via Windows explorer I mapped a drive to the logs folder using the credential of admin and P@ssw0rd - which is clearly the wrong password for that account - and that works!

There is the fan speed issue - which i haven't had on this site ... if you open the web gui up for the appliance it says that a power supply has failed / is missing - this is actually just that the fan speed is set wrong - have noticed that there is a tech note on that one now.

I have also found that if you set SIZE_DATA_BUFFERS but not SIZE_DATA_BUFFERS_DISK it uses the same value for both so if you want the disk size to be different to tape you have to specify it using the DISK file!

Other than that my 7.5 license did not register accelerator so I had to have a new one generated.

All of my SLPs were corrupted after the 7.5 Master Server upgrade

Also, after an upgrade to 2.5.0 and to 3.5.1 the appliance was not usable for between 1 and 4 hours - all self test backups failed - I haven't had an answer to what exactly it is doing after an upgrade yet but if it fails with disk pool down followed by 84 errors after an upgrade then you just have to wait and fire off the self test occasionally until it starts to work again.

I have a couple of outstanding issues still but Support are looking into those

Hope this helps

Stanleyj
Level 6

Yeah that about sums it up.  I have expeirenced almost all of those issues.  I didn't have a corrupt slp but i did have two jobs that strangly enough just started failing about 2 weeks after the upgrade and when i recreated them they havent failed since.  Support thought that was odd also but since you had corrupted slps it makes sense to now.

The share situation has me confused.  I tried every login combination imaginable last week and never could get connected.  Then after a few days of trying and nothing working it just up and stopped prompting me for credentials and i can get to it???  So now that there is a log share you still have to know the root access credentials and be fluent in linux (which i am not the greatest) because you have to copy the logs there to get access to them.  Not sure the purpose  in the share but oh well. 

I'm still having trouble keeping the spoold service running becuase it appears that since the upgrade the appliance cant process the transaction logs fast enough.  Im working with support now. 

I upgraded two appliances with this update and so far the second one has not had any issues (keeping fingers crossed).  But it is only used for AIR imports because its offsite and since my primary media server is not working properly i cant confirm if the issue with transaction logs is effecting this one also.

Mark_Solutions
Level 6
Partner Accredited Certified

So are you seeing performance issue since the upgrade?

I am having major problems at the moment - especially around duplication to tape on one appliance which is terribly slow but some backups are also slow

It really doesn't look like the queue is being processed correctly as it runs but does not clear very much up

I have, when i have time, been kicking it off manually but have to run it several times to clear the queue down

Stanleyj
Level 6

I guess i am seeing performance issues but i dont know to what extent yet because i cant keep all of my jobs running.

performance to tape seems to be ok for me because at the moment its the only backups that are sucessfull.  I had to change all of my jobs to go directly to tape because dedup disk is failing once the t-logs get to a certian point.

To keep all of my accelerated jobs functioning I am having to do the same thing.  Manually kick off the processqueue job.  Your right it will run for a little but and then stop and never fully clears out the logs.

The other day when we found this problem my tlogs were over 2tb and it took about 10hours to get them down so that i could get spoold to stay started. 

Just something i have been noticed and wanted to see if its the same for you.  But i really dont start seeing the issues with the logs until sql jobs kick off.  Most of my sql jobs are towards the end of the backup window so that could be why they are the ones with the most trouble but i have 146 accelerated backups jobs that run without issues just as long as i keep the sql jobs backing up to tape.

 

Stanleyj
Level 6

Mark,

Just wanted to let you know that it does appear i am seeing major performance issues after the 2.5.1 update.  I have said we are having trouble with the transaction logs not clearing and that moved everything to backup to tape and all has been well ever since.  Well i started slowly moving everything back to disk and all appeared to be well until i added a single sql policy lastnight.

My prod jobs including exchange and 160 remote offices was taking about 2-21/2 hours to complete daily incrementals.  I added the sql job in there lastnight and jobs were still attempting to run 10hrs later.  Exchange which is probably one of the bigger backups was prior to adding the sql jobs averaged around 45mins to complete the incrementals.  Lastnight it showed 7 hours for the elapse time. 

I dont know for sure that sql is the culprit but for me it is the one that i keep having trouble with.  I remove the jobs all is well.  i add them back and things start slowing down and failing.

fcurrie
Level 3
Employee

Mark and Stanley,

If you could let me know you case numbers I will chase them up internally.  Email me at frank_currie@symantec.com.

Thanks,

frank.

Frank Currie | Director of Technical Product Management | NetBackup Appliances

E - frank_currie@symantec.com

Mark_Solutions
Level 6
Partner Accredited Certified

Frank ....All case numbers .. at least so far ... sent via email

Stanley ... is your a N5200 or a N5220?

MatsHolm
Level 4
Partner Accredited

Hi!

Have you had any luck with this?

Upgraded two 5220 appliances this week and one of them is alright and working fine (better then before upgrade) but the other is kepp getting spood going down after about 10 miutes of running.

This is our secondary appliance and we have all backups to the working one but we're not getitng any secondary copies.

Haven't been lookning into all logfiles yet but interested to see if you and support have found anything before I start a case?

//Mats

Mark_Solutions
Level 6
Partner Accredited Certified

Mats

Ours is still in progress with little or no improvement so far

Could you do me a favour - on each appliance go into its CLISH and go to Support - Test Hardware

Out of this you will see the BIOS version / firmware versions (MegaRAID, BIOS and QLE)

Let me know what versions your appliances are at and if they are different which one is the slow one

Stanleyj
Level 6

Mark and Mat,

Appliance is a 5200 and here is the Firmware Information incase your interested.


+-----------------------------------+
|ID|        Name                        | Version       |
|--+--------------------+-----------|
|1 |MegaRAID SAS 8708EM2   | 11.0.1-0046 |
|--+--------------------+-----------|
|2 |BIOS                                 | 2.08.010      |
|--+--------------------+-----------|
|3 |QLE2462 Slot3                   | v5.03.01      |

 

The issue i was having with spoold crashing similar to mats for some strange reason has stopped happening for now.  I worked with support yesterday and we decided to move my sql jobs to a window that did not interfier with other jobs and the performance hit i was seeing didn't happen for lastnights jobs.  Everything finished like it was suppose to. 

But..  (There always is one) Now that i have moved my sql jobs back to disk backups my error 6's and 1's have appeared again.  So i really have no idea whats going on there.

I'm sorry fellas but i have no answers for you at the moment.

Are either of you having issues getting into the web interface?  Mine is a little slugish.

Also what does your cpu voltage say?

I'm showing two different voltages for the CPU.  It may have always been this way and i have just never payed attention but that seems a little odd to me.

Mark_Solutions
Level 6
Partner Accredited Certified

Ok thanks - my slow one has a BIOS version the same as yours - my faster one has a BIOS version of 2.09.010 - makes you wonder!

The web gui is slow on the slow one - just makes you wonder what it is doing under the hood - a TOP usually shows spoold very busy but it varies a lot.

Voltages are 1.1 and 0.96 but the same on both appliances

The slow one is using 24 watts more power than the other (384 vs 360)

Stanleyj
Level 6

Yes it does make you wonder.

Have you made any headway on what could be going on? 

We rearranged some of my jobs which appeared to help but when monthly fulls kicked on this past saturday all was looking good until the end of the of my backups and then the dreaded 2074 happened.  Ever since then i have been having real trouble keeping the disk pool up. I start it a few jobs rerun and then fail.  I finally just canceled the whole SLP.  Now things seem to back going. 

Yesterday a 160 remote servers ran accelerated monthy fulls and all duplicated just fine. i can not pinpoint where the issue is.   

Stanleyj
Level 6

Mark/Matt,

when your disk pool is down what are the results when you run this command

/usr/openv/pdde/pdcr/bin # ./crcontrol --getmode
 

Mode : GET=Yes PUT=No DEREF=Yes SYSTEM=Yes STORAGED=Yes REROUTE=No COMPACTD=Yes RECOVERCRDB=No

Everytime i start having troubles i notice the PUT is = to NO and when all is running well it shows YES. 

Check out this article: http://www.symantec.com/business/support/index?page=content&id=TECH183707

This points back to those transaction logs not being processed correctly.  I may be barking up the wrong tree but until i get the put=yes i recieve error 84s on all backups which if to many are running then the 2074 appears and i have to cancel them all.

SRP
Level 4

If I have cause to reboot both my 5220s, then I also get the same problem - I put it down to "disk scanning" - one of the appliances is 2tb short of being FULL, the other had 20tb free.  The near capacity one would take up to 2 hours before "Put" went to "Yes", the other appliance would take a lot less time. 

Don't know whether this is a 2.5.1 problem (didn't happen prior to upgrade though...), but I've noticed that since applying the patch, that if there are a large number of concurrent jobs running, that I'd get 2074 errors.  This then prevented me from cancelling the "running" jobs.  Had to go in and manually kill the processes of a large number of the jobs.  This then brought "Put" back into "Yes" state and the remaining jobs that were "running" continued to run.  I also deactivated all policies, so that no further jobs would run until things calmed down. 

The backup windows were as I created them when I commissioned the 1st appliance, so nothing had changed.

 

Sean

Kenneth_Hansen
Level 6
Partner

Appliance 5220 upgrade 2.5.-> 2.5.1

I see same error as Stanleyj on 1 out of 10 upgaded appliancen.
Spoold terminates unexpectedly. Have rased a support case on it.
Not good.....

Also:
/usr/openv/pdde/pdcr/bin # ./crcontrol --getmode

Gives me error message. Seems crcontrol is not able to talk to spoold process..