Lessons learnt from DR migration to new hardware platform
Not so much as a discussion more of a "this is what caught me out" when I migrated from 2008R2 DL380 G8 to 2012R2 G10 on NetBackup 8.1.2. I hope this helps someone in the same position that I was in. So it should be a pretty simple build a vanilla corporate build. Stop all you backups, carry out a full catalogue. Shut old server down ensuring that you have the DR files and know where you backed up the catalogue to. Rename and re-IP new server to old server configuration. Install NetBackup software with same role os old server. Using the DR wizard to recover the server configuration. All pretty simple except for just one thing. The DR process installs a new CA. This now means all your clients and media servers now can no longer communicate with the master and report the the error "vnetd proxy encountered an error". Genius Looking at the sheer weight of the issue made me shudder as we have hundreds of client an the thought of having to visit each and every one would have kept me busy till christmas. The fix was buried deep in a Veritas knowledge base Article: 100043536 The command to restore your old CA prior to carrying out any restore from the DR wizard is as follows excerp from article: nbhostidentity -import -infile M:\NB_DR_FILE\NBCAT_nbmaster1_1531458511_FULL.drpkg Now you can run your DR wizard and presto you now have all your clients happily communicating with the new master server once you DR restore completes. I think that the propellor heads at Veritas need to add a checkbox to the DR wizard to ask if you want to restore your old CA. Might ulimately be mute since 8.2 you can use your own internal CA. Or did I just imagine it. :) Full articles URL: https://www.veritas.com/support/en_US/article.100043536 Happy DR'ing503Views2likes1CommentNetBackup GRT Support for Exchange 2016 CU5
We recently issued EEB 3918337 for NetBackup 8.0 to support Exchange 2016 CU5. Database backups and restores are fine without the EEB, as are GRT backups. You need the EEB to browse and restore mailbox content from your backups. The EEB is an adaptation to changes in the Exchange database. Future NetBackup releases will include the adaptation. Earlier Exchange versions such as 2016 CU4 continue to be supported with or without the EEB.Solved3.3KViews2likes8CommentsMoving MSMQ on Windows Server 2012 for EV11
There is no way to use a GUI to move MSMQ on Windows Server 2012. You can move MSMQ using PowerShell. NOTE: In the example below I have created a new MSMQ folder on G:\. The path is G:\MSMQ. You do not need to create the MessageFiles, TransactionLogs or Storage folders as they will be created automatically. To move MSMQ perform the following actions:Netbackup - Transfer dedup storage to another storage media server
Hi All. I have a LUN from an HP P2000 storage presented to my media server. So it appears as a partition on my media server. This server is configured as a storage server in my Netbackup. and a "pure disk" disk pool and storage unit is created from it. So we are writing to this dedup disk (storage unit) since long time. Now we are bringing a new media server, and we want to transfer this dedup to it. off course we will present this LUN to the new server, so it will appear as a partition on it. But how to do this transfer between servers in Netbackup, so the current data in this dedup will remain readable and restorable in Netbackup. Any hints or Ideas ? Thanks,886Views1like6CommentsQueries regarding MS-SQL backups with Log Shipping
Hi Folks, I'm looking for some thoughts from those of you who have configured NetBackup to backup MS-SQL databases that are using Log Shipping. I'm aso thinking of raising a case with Veritas, but as I'll have to give the case a lower severity I'm thinking it may take a little while to get a response. My current environment is 2 x NetBackup Master Servers (7.7.3) running on Windows 2012 R2, with 2 x NetBackup 5230's (2.7.3) as Media Servers. SLP's and AIR are used to cross-pollinate the data between sites. I currently use a mix of script based SQL backups, Intelligent SQL Policies and VMware Intelligent policies using the SQL functionality (most of my SQL backups use the last option). I have a new environment coming onboard (HPE Content Manager) which will have two SQL servers (one at each site) with Log Shipping enabled. I've read the manual (in this case it is the Veritas NetBackup for Microsoft SQL Server Administrator's Guide for Windows Release 7.7.2, as it wasn't re-released for version 7.7.3), and I have a couple of queries as some of the info is not quite clear to me. Point 2 on page 150 (Under the heading "To configure NetBackup to support database log-shipping) states the following: "Any policy that is used to back up the primary should also specify the host that contains the secondary database. See “Backing up SQL Server in an environment with log shipping” on page 150." On the same page under the heading "Backing up SQL Server in an environment with log shipping" states that: However, a backup must not be performed on a secondary (or standby) server. Databases must always be backed up on the primary server and restored on the primary server. This requirement is based on the Microsoft SQL Server restriction that is outlined in Microsoft knowledge base article 311115. If you try to perform a backup on the secondary server, you see a message in the dbclient log similar to the following: 16:33:26 [1208,2348] <16> CODBCaccess::LogODBCerr: DBMS MSG - ODBC message. ODBC return code <-1>, SQL State <37000>, Message Text <[Microsoft][ODBC SQL Server Driver][SQL Server]Database 'Mumbo' is in warm-standby state (set by executing RESTORE WITH STANDBY) and cannot be backed up until the entire load sequence is completed.> My query with these two points is that if we also have to add the secondary SQL server to the policy that backs up the primary, when the policy runs I would think it will also backup the secondary. That sort of seems to contradict the manual. Or does it not back it up even though it is in the same policy because it knows that it is in the wrong mode to do a backup? Hope that doesn't cause incomplete backups :) My other query is to do with log truncation. As I read it from the manual, it would appear that you still do transaction log backups, but with no truncation - correct? Based on what I've read, I think I'll have quite a bit of reconfiguration to do. VMware Intelligent policies using the SQL functionality won't be useful any more as there are other DB's on the same servers that do need log truncation. I'm thinking I'll have to do VM Intelligent policies without the SQL options to give me a backup of the machine as a whole, and then several script based SQL backups or Intelligent SQL Policies to backup DB's with and without log truncation as required. Messy and more policies, but I don't know I'll have a choice. Thanks in advance, SteveSolved2.8KViews1like5CommentsMSDP Storage server config error - credentials already exist
Lab environment with Windows Master and Media, NBU8. Adding media server as MSDP storage server. Goofed the first attempt (no directories added after drive letters), went back in wizard and fixed that problem, but ever since have received "Credentials alredy exist(99)" failure. I've restarted services on master and media, media server as MSDP storage server does NOT appear as a storage server in either the GUI or with nbdevquery -liststs (media server only shows as basicdisk and as an advancedisk that I configured earlier) Running PDDE_deleteConfig.bat on the media server fails because the EtcPath variable is not being set by the script. Any guidance?Solved7.4KViews1like4CommentsJob priorities are not granular enough
This is a feature request. As verified with support, job priorities are all linked at the server level, which is definitely not granular enough for my needs. For example,if you have multiple stages or sub-jobsfor eachserver, such as backup to disk full/incremental based on daily/weekly/monthly schedule, duplicate to disk/tape, or separate verify jobs, you cannot prioritize anything beyond the main server job over other entire server jobs. Theproblem becomes much more apparentwhen you have multi-terabyte jobs that take most of the week to complete. Add on another dozen or so servers that need backed up (some of those comprised of multiple terabytes of data as well) and you have an extremelyfinite backup window and this is where better job priorities would play a huge role. For instance, it would be nice to be able to prioritizelarge verify jobs lower than thesmaller ones since they take much less time to complete, but still have the actual backup to disk job be a higher priority level. Or to be able to prioritize certain full backup to disk jobs in one order, but have a linked duplicate to tape job be a different priority level entirely. I don't think I need to try and recreate all of the different scenarios of why this feature would be beneficial. It seems like this should already be part of the software, but it isn't for some reason.604Views1like0Comments