[Veritas-bu] missing digests
I missed some digests. If there is another packrat such as I on the list who has volumes 78 through 80, I'd appreciate copies. Off-list, please. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] HP library and 5220 appliance - zoning
We're trying to configure some LTO5s, shared out from an HP ESL, on a 5220 at 2.0.3. Linux isn't creating device nodes, so none of the NBU methods of scanning are finding anything--no robotics, no tape drives. The FibreChannel Show appliance command shows the HBAs configured correctly, the qla2xxx driver loaded, the card as an 8Gb initiator, online, and the correct number of Remote Ports entries for the dries and robotics. But the Remote Port entry is all there is for each drive, and no drives get configured by Linux. Does anyone have an example of FC Show with some tape drives so I can see what right looks like? There is dmesg output (from rescan-scsi-bus.sh, perhaps?) with the appropriate number of lines like: 3:0:23:0: scsi scan: consider passing scsi_mod.dev_flags=HP:Ultrium 5-SCSI:0x240 or 0x1000240 so _something_ in the OS can see the vendor and ID fields, but no dev entries are being created. Before I get the fibre group more involved, can anyone comment on what the zoning should look like at the appliance end? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] competing schedules (was:RE: Veritas-bu Digest, Vol 73, Issue 19)
When it comes to overlapping schedules, set both schedules to run at the same time and the schedule with the longest retention will be the only one that runs. Actually that's the most common misconception in NetBackup. With frequency-based scheduling, more than one schedule due to backup, and identical window-open times, the schedule with the longest _frequency_ setting will run; retention does not factor into this. That is the elegant mechanism by which an infrequent schedule (say, a yearly backup) always wins when it's due, no matter how many monthly, weekly, daily... schedules may also be due. Those lesser schedules get their turn at the next window opening, and again, ties go to the due sched with the greatest frequency value. This is why daily/weekly/monthly/quarterly/annual/... setups work correctly without extra admin work. Simple example: test-policy daily-schedule, cinc, freq 1 day, -0400/7-days weekly-schedule, full, freq 1 week, -0400/7-days monthly-schedule, full, freq 1 month, -0400/7-days [add quarterly, annual, 5-year, ... scheds the same way] Retention does not matter. The daily above will run every day unless it is trumped by a weekly which hasn't run in a week or a monthly which hasn't run in a month. The weekly as well as the daily will be trumped whenever the monthly is due. And so on up the _frequency_ hierarchy. Demonstrations of this are in the list archives. You can create a test policy and test this for yourself. So the yearly schedule should have a longer retention, than the monthly or weekly so the others will not run. Hope this helps. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Archive list
Haven't seen this much interest in archives (in the NBU sense) in years. GOOD discussion. [confession:] Hi. My name's bob944 and I've used user archives. Wow. This is the most press UARC has received in ages. First, every person who commented on this made sense and appears to understand the way NBU archives work. Second, an issue: o Not telling the user community about archives is like not telling them the rm or del commands exist. Users--and admins--make mistakes, but my personal High Mighty Pedestal isn't quite high enough to decide what they can't know about. In the following, I'm assuming that o UBAKs and UARCs are understood by the admin and the owner of the data: user specifies the data; upon successful archive, the data are removed o I and the users decided where and for how long the data are be preserved Re: David Rock's follow-up: Backup admins are generally in the business of preserving data for recovery, not removal of files from systems. I agree, yet don't see a conflict here. IMO, the admin _does_ preserves the data--automatically at it's very last (that may be important) state--before the user blows it away. The user has made sure we backed it up before it's deleted. He could have just deleted it himself. Why is that not a win? If we back up /ora/db/our_2011_customer_masterfile.dbf, we do it so that we can recover it after some bonehead mistake OR DELIBERATE USER ACTION (the backup guys have backed up the DB; let's blow it away and go with the new schema) removes data that is needed today or a year from now. There's no intrinsic difference when we, and the user, set up an archive. The user can already 'rm our_2011_customer_masterfile.dbf' without any archive in scope, call up and need it restored. Same thing for a deliberate decision by the user that they no longer need it--but just in case, would like the last version backed up before their schema change. To Mr Rock's point, we _have_ preserved the data so that the user can remove it with assurance that it can be recovered within the parameters agreed to--the same as for backups. I see no difference between a UARC and a morning phone call of did you back up that old customer master file? | yes | Okay, we're blowing it away. Unless one has a) incorrectly instructed the user community on what a UARC does, b) set the UARC parameters to something inferior to what the user expects or c) treats UARCs as sacrificial data, I don't/haven't seen an issue: we're doing exactly what we do for scheduled backups--preserving the data to agreed-upon standards. Regarding the archive policy (I'll still call it that, assuming that the admin and the user know the user runs a user backup, specifying what gets backed up and, if it's an archive and successful, subsequently deleted), I've only set this up in production a couple of times in the last 15 years. My favorite was in the late '90s with NBU 3.1 or 3.2. A user group had a finance application in which month-end created an elaborate set of directories, subdirectories (thousands) and files (hundreds of thousands). The app left the files in place, leaving it to the user to decide when the data were of no value. None of the users (or perhaps, their management) was keen on using a command line or GUI to recursively remove thousands of directories and the files therein, and the space used was a big issue in those days. They used the app's data for a day or two, then their business process used a UARC schedule on (IIRC) a unique policy to back it up just in case and delete it. I've also set up UARC schedules for groups that were leaving the corporate HQ's backup umbrella so they could get tapes of their data to be recovered in their new home. [Rock:] Basically, what you are trying to do is something that NBU is not meant to do. The expected behavior for Archives is that they are managed by the client admin, not the backup admin. Archive jobs get a filelist from the user, run the archive, and [if and only if the backups were 100% successful,] delete the files when they are done, so the user would already know what files were archived because they would have supplied the list to the command in the first place. Whether the original poster understood this or not, IMO it's an excellent clarification of how UBAKs (without the deletion) and UARCs work. It's scary or unknown territory to many admins and most all users. From: Kevin Holtz All is true but I wouldn't say it's something NBU was not designed to do. It was made difficult for a reason rather than easy due to the nature of the outcome e.g. Archiving. This is very common practice for DBA's. Yes, this all needs to be scripted, blat etc. and you will need admins to help if you don't have access to the local systems. What he said, except the admin isn't helping me, I'm helping them by providing a user tool. Though I think Rock was commenting on the policy aspect to make sure the original
Re: [Veritas-bu] Issues with media server dedupe pools
[ restore uses copy from remote serer ] Thanks so much for your reply. So if I understand correctly, I add the following in the bp.conf on the master server FORCE_RESTORE_MEDIA_SERVER = nbumedia2 where nbumedia2 is the name of the media server I want to use for restores ? I haven't used these two in an OST environment, but there are two ways to specify the copy number to use for a restore: 1. bprestore -copy N, where N is the copy number you want to use as the source copy 2. /usr/open/netbackup/ALT_RESTORE_COPY_NUMBER, where N is the content of the file ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Exclude lists on Unix clients
Patrick Whelan said: How do most of create and populate exclude lists on Unix clients, if you don't have access to said client? As Scott Kendall mentioned, bp[gs]etconfig is the cleanest way to go, (technically and politically). That said, setting and maintaining *clude lists with a) alternate-client restore to force standard *clude lists and/or b) backup|script|restore to preserve and update existing lists can be a big win if the need is non-trivial and lines of authority allow it. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NetBackup OpsCenter vs. OpsCenter w/Analytics
From: Justin Piszcz jpis...@lucidpixels.com [...] However, I've just started looking into Analytics (w/trial key) and was curious if anyone had found any killer features it offers over the regular OpsCenter in terms of metrics/reporting or other things I may have missed? I'm not much for graphics, but the temperature charts for data like tape drive utilization were much more usable than any of my scripted analyses of bptm and other logs that I've done over the years have been. More info for 1/1000 of the work. I know you can run custom queries against the OpsCenter DB vs. the ~100 or so canned reports of the regular OpsCenter, are there customers out there who have taken advantage of this or any other feature I am not be aware of? If you're a SQL guy and can puzzle out enough of the schema (apparently, a VBR background and documentation is a big help with that), it's pretty easy to write custom SQL for admin or user reports. You can also add custom objects to the schema but a) the method seemed awkward and undocumented to me and b) it seems necessary to plan ahead to remove all customizations before a version upgrade, then reapply [and re-test, re-verify] afterwards. And that makes sense to me--if I were customizing the metadata files in the NetBackup catalog, I wouldn't expect the vendor's upgrades to function properly with arbitrary user changes to those files. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] massive catalog growth
Great, so it appears NetBackup is actually not
respecting my exlcude_list. I just ran a test job
and realize it is backing up DSSU's, the catalog
and a whole bunch of other things that should be
and are excluded.
And we all assume it's not _really_ named exlcude_list, right? :-)
Not that I've seen every NetBackup problem by any means, but since
NetBackup 3.1, I have never seen {ex|in}clude_list processing not work
correctly.
Maybe this is that first time, maybe not. Suggest you look at your
bpbkar log; it (at verbosity 1, IIRC) will show all of the *clude_list
processing, file by file. When I don't understand why something is
included or excluded, that has always shown me my error.
Also, related to your catalog growth. There's a technote or two
explaining the per-file catalog entry. Often, each file is
generalized to take up 100-130 bytes (less with binary catalogs).
That number is a good working average, but can be way off if the
pathnames--almost always the longest part of each entry--are long.
Backups of Windows clients with a lot of filer shares mounted often
have paths over 400 bytes long, IME. The size of the files file of a
backup of that sort of path could be five-or-more times larger than
the same number of files with normal-length paths. FWIW, the one
Isilon box I ever dealt with had very long paths if backed up from the
root.
bpflist is a cantankerous command, but its 17 fields will clearly show
how much of the differences in your older backups and newer ones are
due to path lengths. Even simpler, divide the size of the .f file by
the number of files backed up by that stream to compute your own
metrics on bytes-per-file-entry.
If you're still having difficulty, it could be helpful to post
platform and NetBackup versions of the master and clients involved,
our exclude list and bpbkar log entries that illustrate the problem.
___
Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NUMBER data buffers
So I've read the tuning guide, I've played around with different options for SIZE and NUMBER of buffers and I understand the formula of SIZE * NUMBER * drives *MPX as it relates to shared memory. You're going to get a lot of replies. Everyone is a buffer tuning expert. :-) If all you really need is how many buffers, of what size, muxed how many ways to how many drives can I possibly use, skip everything after this paragraph. It was only six years ago that most NBU platforms were 32-bit, media servers might have only a few GB of core and have other limits on shared memory from the OS, so size size*number*mpx*drives was a more pressing concern. Even today, it takes a serious amount of buffering and streams to use up 32GB, probably more than is useful. With, say MPX 3 and 32 256KB buffers, that's over 1000 simultaneous streams. It would take one killer network|media-server|storage combo to keep up with that. The NUMBER of buffers and MPX level seem to be the two variables Makes more sense to me to think of SIZE and NUMBER as the two variables for a backup stream. Then think separately about MPX as a how-many-TunedWithBuffers-streams do I deliver to a tape drive to make it stream. (Recognizing that you may mux different classes and schedules in different ways for backup or restore performance considerations.) You didn't ask for tuning methodology, but since you're in the tuning guide already... One of the points you may have gleaned from the tuning guide is to look at the wait and delay counters, and whether you're measuring a data supplier or a data consumer. Understanding producer/consumer and wait/delay together gives you a sound basis for making changes. As does gaining the numbers to see if it's 300 seconds of delay on a 10-minute backup or 300 seconds on a two-hour backup. That's plan A. Plan B is empirical. (My and others' methods will be in many posts from the last ten years if you check the archives, so this'll be relatively brief.) Define which path, under what conditions, with what data, you want to investigate/optimize. Strongly recommend you work initially with one server, one invariant chunk of data, one stream, no conflicting activity to nail down the basic limits of that client/network/STU combination. Only after that is optimized would I throw multiplexing into the game. Make a test policy, say TUNE-win, full schedule, no scheduling windows, specifying the STU, client and path that will take a non-trivial amount of time to minimize variables yet be short enough that the testing doesn't take forever. Say, enough representative, unchanging data for 10-15 minutes elapsed write time. Record number, size of buffers, wait and delay values and write time. Double only one of the parameters. Retest/record. Do that until there is no gain and leave that value alone. Now repeat the cycle while changing the other parameter. Retest until no gain. Then go back to the first parameter and change it up and down (one doubling and one halving) to see if that needs to be revisited. BTW, NUMBER_DATA_BUFFERS is per backup stream. Just want to make sure you are clear on that--not sure from your note. You have a good start on setting size/number now. Extra credit for trying other clients/STUs and other variables, of course. Use those values and try controlled multiplexing (you've probably maxed out a client, so generally this will be with multiple clients. Net bandwidth might also rule here, of course.) Since 6.x NetBackup, you are unlikely to run out of buffers. Status 89 if you do. Regarding number versus size, there's no point in having a huge number of small buffers or a small number of huge buffers in any environment I've seen. The methodology above usually shows you an optimal combination that is somewhat balanced. Sooner or later, the allocating of a ton of little spaces, or of a few huge (contiguous) spaces will lead you to a reasonable balance. I probably never had speed improvements over 1024 buffers, and often much less. DON'T FORGET NET_BUFFER_SZ! Hugely important on both Windows (in the GUI) and *nix clients to get the data out of the client. Here's my question. Of the four parameters: MPX level # of drives (I have 12 drives) NUMBER of buffers SIZE of buffers (must be multiple of 1024 and can't exceed the block size supported by your tape or HBA) The NUMBER of buffers and MPX level seem to be the two variables here. I have MPX set pretty low (2 or 3) and NUMBER of buffers set to either 16 or 32. When I multiply it all out, I get a hit on my shared memory of less than a GB. My media servers are dedicated linux hosts that only function as media servers and that's it. Furthermore, they each have somewhere around 35 - 50 GB of memory a piece. With my current configuration, I'm not even scratching the surface of the amount of shared memory that's sitting idle in my system while my backups run at night. Is there any reason I
Re: [Veritas-bu] NBU LiveUpdate 7.1.0 - 7.1.0.2
Just set up LiveUpdate for the first time, and was trying to use it to update a 7.1.0 client to 7.1.0.2. It says that it found no updates to install...It is because you can't use LiveUpdate to make such a minor change. Would it only work, for example to go from 7.1 to 7.2? LiveUpdate for 7.1.0.2 is supported. From the Web site (http://www.symantec.com/business/support/index?page=contentid=TECH16 5302): The NetBackup 7.1.0.2 LiveUpdate containers include all the files distributed as part of NetBackup 7.1.0.2. and the four bundles (win ia64, win x86, win x86 and Unix) are available there. Unless LU has changed since 6.5, it's not the easiest thing to set up (understatement) and get working. Make sure you have the correct lU packages, they've visible on the specified URL, the client end of LU is set up, the LU policy is setup, and a bunch of other details. Sorry I'm not more specific; I gave up on it after a partially successful trial. 10/10/2011 12:25:01 - Info nbliveup (pid=14762) Updating LiveUpdate Server location to: http://msback02/LUServer/NBLU_7.1.0.2 [...] ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Warning Beta 7.5 OpsCenter
[ Dave points out that Beta upgrades aren't supported ... ] A support case is open at Symantec. So far they did not mention anything about upgrading is not supported. As that talldude suggested, beta releases should be clean installations. See this from the NB_75_Beta_Primer.pdf: Supported installation criteria The following list describes what this Beta program supports: │ Clean installations of NetBackup 7.5 are supported. │ Upgrades from a previous NetBackup Alpha, Beta, or an earlier version of NetBackup to this Beta release are not supported. │ This Beta release does not support mixed-server configurations. (The final release of NetBackup 7.5 provides the capability of maintaining a mixed master and media server environment.) They believe it is related to the fact of existing custom reports. Entirely possible. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Tapes mounting in ACSLS but not Netbackup
[...] When i initiate a backup i can see the job which sits in a mounting state in Netbackup, however, the tape itself is mounted in the drive in ACSLS. Running a vmoprcmd shows nothing in the drive. Although i'm sure at one stage i've seen it pick up the mount in vmoprcmd but not in the GUI. I guess this is a communications error but i'm stumped, can't see anything meaningful in logs. Sure sounds like incorrect drive mapping. Like Len, it's been a long time since I had an ACSLS environment, but remember that ACSLS is essentially a robotics controller and is not in the data path. The first step I'd take is mounting a tape in each drive in turn, and use appropriate OS commands like 'mt -f /dev/rmt/0cbn status' to establish which drive it actually loaded. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Open file backup error
I would agree! VSP is dire!! Ignore it. I can say that in all my 11+ years in NBU, I turned it off ! I have to throw the BS flag here. That's just FUD. Until Microsoft finally came up with VSS in w2k3, the OS vendor provided *zero* tools for open file backups. Veritas had been filling that void since... what, 3.2 c. 1999 or so? OTM (pre 4.5fp6) and the Veritas-written replacement, VSP, functioning in the horribly flaky world of Microsoft device drivers, do a pretty good job of getting a consistent version of write-locked files that the OS won't otherwise allow to be read. The admin had to do his part--configure VSP, have the snap space and not let his PC's antivirus program lock the snaps. And while I'm being bitchy, are there more than three people left on this mailing list WITH THE COURTESY TO EFFING TRIM THEIR 400-LINE REPLIES? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Open file backup error
Hey Bob, it's not FUD. VSP has been problematic since, I don't know, 4.5? Yessir, 4.5fp6 was the total replacement release where Columbia Data Products' OTM was replaced with VSP. It would fail to release the lock on the file and therefore ]...] Sure there were problems, the worst IMO being what I believe you're alluding to--where antivirus software that wasn't told to ignore the .vsp files and would keep them open so NetBackup couldn't delete the cache file, requiring a reboot to remove. And several other to be sure--but where I intended to go was that, properly configured, VSP generally did a good job of OFB for me--where _any_ OFB backup was a win since Microsoft had nothing to help their own OS. I agree with you that MS took too long to come out with VSS. But now that VSS is around and works very well (most of the time), there is no reason for VSP. Agree, and that's the reason this is worth a reply. I intended to say it explicitly in the first message but forgot: once MS offered OFB with w2k3, there was no reason to keep using a third-party installable device driver--configure the OS vendor's solution and move on. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NetBackup 7.1
How has everyone's 7.1 experience been? It runs filesystem backups well in the lab. Any gotchas or things to look out for? Things you would expect to be found in FA and to be in LBN before you discover it the hard way, including - upgrade of a solaris master with windows clients fails/takes days (look up bpplconvert in the archives; workaround available) - NDMP 3-way backups fail (workaround in Release Notes) - bandwidth directive for one client will apparently limit the entire domain's aggregate throughput to that limit (caution in Release Notes) If you've skipped 7.0[.1], be sure to read all those docs as well; important changes there are not repeated in the 7.1 docs. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Restoration Issue
R.I.P., Reading Comprehension. sheesh ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Restoration Issue
Got a File Server with a share that is approx 400GB and contains millions of files. San Media Win2k3 Sp2 NBU 7.0.1 talking to a 7.0.1 Master Problem: We have identified 40,000 files, of various types (ie: word, excel, jpeg, bmp, pdf) that are ZERO bytes. They appear to have been like this for 4 months [...] 1) Files are scattered across various folders / subfolders that are in use 2) I cant just restore everythink back to its original location, because there are files that have been recently used and amended Restoring from a file of filenames (as has been suggested), to a different directory, would be my preference. Then maybe script a comparison between the list of zero-byte files and the ones you have restored... If that is impractical and you have a backup of the current state of the server, zero-byte files and all, you could a) rename/move/delete the zero-byte files and b) restore the entire share to original location but do NOT check Overwrite existing files. You should get 40,000 restores and (millions - 40,000) file exists messages in the log. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Solaris master 7.1 upgrade hang
Wish I would have read this yesterday. I'm in the middle of an upgrade right now and all my window's policies are slowly being converted right now. I figure it should finish in 10 hours. I'm very lucky to have the ability to keep this env down this long. Little upset that this known issue wasn't published on the 7.1 late breaking news website. Did you call support? You don't have to wait it out. When I first called it in, I received the run-time workaround (my phrase) which is to do a ps, kill the bpdbm process shown, repeat until done. My understanding is that bpdbm is the process waiting five minutes to time out for each client; killing it over and over will get through that part as fast as you can repeat (or script) the ps/kill. When the upgrade the hardware definitions of the client part is done, either from it going through all your clients five minutes at a time, or by you killing them to speed the loop, the upgrade will finish as normal. At that point, run the bpplconvert manually, which will work normally and be done in a second or two. Whether you let it go 10 hours, timing out on each client, or use the ps/kill method, the upgrade is the same: successful except for those client hardware definition updates. Running the bpplconvert then corrects that in milliseconds. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
[Veritas-bu] Solaris master 7.1 upgrade hang
Well, _this_ was interesting. Simply, upgrading a Solaris master to 7.1 can/will hang if there are Windows clients with hardware types of x86, x64 or IA64. The good news is that it does no damage and that there is a simple workaround that, um, works. http://www.symantec.com/docs/TECH156810 ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] three copes with ost
I will start using OST for my primary site (NY) and secondary (NJ) with cross replication. I also want a third copy going to tape, but I want the tape to be created at NY (for all backups, NY and NJ). here is my question... With Storage Lifecycle, for my NJ backups, if I create a copy to tape, then it will read from my primary (NJ), so it will kill my NY- NJ link. Is there a way that I can tell it to use the second copy that is now in NY? Or can I make the NY primary? Two SLPs should be all you need to force dups to work the way you want. NY backup to NY disk duplicate to NJ disk [from NY disk] duplicate to NY tape [from NY disk] NJ backup to NJ disk duplicate to NY disk [from NJ disk] duplicate to NY tape [from NY disk] See Adding a hierarchical duplication destination in the NetBackup Admin Guide. The indentation is the visual indicator of which backup, copy or duplication will be the source. A duplication at the same indentation level as multiple backup copies will use the primary copy. There's a good diagram in the guide. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Size buffers / Performance Tune on Ex2k3 with LTO4
Just a quick question, but I used some default NetBackup settings to tune Exchange backups and restores. If I move to a different library with LTO4 (rather than LTO3), do I need to reconfigure the settings? Windows 2003 SAN Media Exchange, with Master Win2k3 SP2 Need? No, unless there were tape driver (not NetBackup) settings changes--which I haven't seen in years. For NBU, since the maximum transfer rate of the drive is now faster (and LTO4s vary transport speed to make up for slower transfer rates rather than stop/back up/resume when the host can't keep up), you may want to revisit the bptm logs to see if buffer tuning is indicated. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Media Servers Per Tape Drive Ratio
How do you determine how many media servers per tape drive in your environments? I have 10 LTO3 tape drives and have been using 2 media servers per tape drive. Thank you. I read this question differently than others who responded. In case the question related to SSO, recent NetBackup versions (6.5 and higher IIRC) reduced the traffic among media servers sharing a drive, making it practical to have more media servers sharing more drives reliably. A dozen media servers sharing four LTO4s certainly works; I don't know what the practical maximum would be. Note: the more of your media servers that are Windows, the more problems you will have (RSM, HBA drivers, network comms for SSO, ...). ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] KMS encryption
I just went to my library and turned on Application Managed Encryption Interesting. Since NetBackup doesn't need anything but bptm and an LTO4 to do KMS, do you know if your library somehow blocks that process if that app managed crypto isn't turned on? It's my understanding that the keytags and keys are handled in the SCSI command stream between the media server and the drive. Now I'm wondering if there are libraries that snoop on the data stream and detect/block those commands. In all this, I'm assuming your library and drives aren't set up with the little add-on Ethernet cards that deliver the crypto from other apps. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Fw: KMS encryption
Today i tried configuring the KMS on my master server(running on AIX). It worked perfectly fine, i took help from veritas support and according to them we can only keep one key in the key database, it will always use the same key for encrypting the data. Every time we need to change the encryption key , we need to define the new key and deactivate the one that is activated. Either they were wrong or you misunderstood. You can have ten (from memory--it's in the book) keys in a keygroup. Only one key in each keygroup can be in the Active state, which is the key used for writing. The rest of the keys in a keybroup can be in the other states (pre-live, inactive, deprecated and terminated). All active AND inactive keys are available for decrypting; NetBackup matches the key-tag, which you can see in your database and in a NetBackup image list. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NBU 7.0 media server upgrade problem
I performed an upgrade from 6.5 to 7.0 on my master server all went well, I performed an upgrade on 1 media server all went well. Before I could perform this on the next media server SCACIFS01.domain.com I started getting NetBackup TLD Control Daemon - Invalid magic number from client SCACIFS01.domain.com in the event log on the master server. I performed the upgrade on this media server SCACIFS01.domain.com it still has the issue. The only time I've seen invalid magic number messages is when two mismatched components were trying to communicate. Once was some upgrade where I'd upgraded, say, (say == I don't remember) the client but not an agent. That's pretty vague, but if you're still fighting it, it might be worth scripting a comparison of the size (or, better, the sum) of each NetBackup binary between systems that works and the one that doesn't. Uninstall/reinstall might be faster. There's an FT mention, AVR, drives that appear configured ... I'd first blow away the device configuration and rediscover. And whatever FT configuration there is--I've only done FT once and don't remember it fondly. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu Digest, Vol 49, Issue 20
I want to add that netbackup 7 clients does not have VSP. Because--and I'm sure you know this--now that Microsoft no longer supports its server OSes which don't have VSS, neither can Veritas (no Win2000 in NetBackup 7)--hence no longer a need to provide OTM/VSP. And I don't remember anyone complaining so much about OTM/VSP for all the years when Veritas provided an open-file backup solution because the clueless OS *vendor* did not. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NDMP backups
I am trying to run NDMP backups with certain paths and I am getting error 99 in the GUI, and the error ndmp_data_start_backup failed, status = 9 (NDMP_ILLEGAL_ARGS_ERR). [...] Backup Selections: [...] /vol/vol2/data/[A-M]* [...] Unfortunately (to my knowledge), wildcards are not supported with NDMP backups. You?ll have to explicitly supply the paths. What Thanks Jonathan. That is fantastic news? Anyone know if that will change or has changed with version 7? This should be an NDMP/filer issue, not anything to do with the backup vendor. NetBackup (or any NDMP DMA) doesn't produce that error; it only passes it back from the NDMP server. NDMP_ILLEGAL_ARGS_ERR is part of the NDMP spec (more than you want to know at www.ndmp.org) and it is returned (for many things, but in this case:) if the NDMP implementation doesn't accept the backup type or variables in an NDMP_DATA_START_BACKUP request. BlueArc filers accept asterisk wildcards; some filers accept them in certain places, other NDMP implementations don't accept them at all. Same thing for accepting directory versus file paths. Isolon accepts all sorts of wildcards and environment variables and a bunch of environment variables Did you RTFM? There's a NetBackup Administrator's Guide for NDMP for every release which tells you how to configure a policy, and has a link to the vendor-specific NDMP guide (http://entsupport.symantec.com/docs/267773). ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Resetting the archive bit on windows backups
This occurred to me on the way to work. Does the NB client for windows reset all the archive bits on files after the backup's completion? That's the only thing that makes sense as I see it. If it reset the bit after each file, then a last-minute failure of the backup would not let the next incremental backup what failed the previous run. This has always been documented--but never well--in the Admin Guide. See the NetBackup Administrator's Guide, Volume 1.(I looked in the 7.0 manual in case it's more clear than earlier versions--but it's still ambiguous on the obvious questions) Configuring hosts Configuring host properties Client settings (Windows) properties Wait time before clearing archive bit Incrementals based on timestamp Incrementals based on archive bit (fulls clear it, diffs clear if successful) There's a new (to me) technote on this that fills in a few blanks: http://seer.entsupport.symantec.com/docs/239393.htm ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] open files to be backed up
[re: the usual Windows SQL Server MDF and LDF files in use] Can't these filed be backed up via open file file backup (VSS/VSP) Not sure you can back them up even _with_ Windows open file backup methods--try backing up the perflib stuff. Microsoft seems to lock these files with an exclusive write lock that even the volume snap methods don't get past. Even if you could, to what point? The data are effectively useless--they are, at best, crash-consistent. No DBA in his right mind would think that was an acceptable backup strategy. Like perfdata* files, I exclude SQL Server (and any other database) files unless the database will be down and backed up as flat files. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] TapeAlert Code: 0x1e, Type: Critical
We think we have found the cause, a netapp tape zone in one of our fabrics which had an unattended server in it I hope it wasn't a Windows box. Maybe 6-7 years ago, an admin zoned a Win 2000 system to tape for some reason and also didn't shut off the RSM service. RSM seems to rewind drives at random, apparently assuming that it should be managing the drives. That trashed a lot of backups--both in-progress and, IIRC, overwriting some. There should still be technotes on RSM. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] KMS Key Rotation
Once you have setup the KMS and assuming you want to restore them. What is the necessary info required to restore. Pool Name ?? Key Name = ?? Key Tag ?? etc Phase-1 and Phase-2 don't show this info. From where we will get this info for the restore. Why are you importing the tapes? If you're restoring to the same master which created them that's unnecessary. But whether you've imported the images or the images are still on their original server, the key tag is what you need, and that shows up in the GUI (it's in the manual) for each image and, IIRC, in bpimagelist. That key tag is what NetBackup matches against keys in Active and Inactive status; if found, that key is used for decryption. If there is no matching key tag, you must restore/import/re-create it from your documentation and/or the keystore backups you have maintained. Example management of keys/changes/records has been supplied earlier, notably by Hinchcliffe. FYI, I have been told, but have not tested, that _all_ keys in the keystore, regardless of keygroup, are tested when looking for a decryption key. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Migrate master from HPUX to Solaris
IMO, those characterizations of the content of the cited technotes can be improved: This technote says that you can /must not do it http://seer.entsupport.symantec.com/docs/267137.htm No, that two-year-old technote that says support will 1) help you recover if you try it yourself and 2) support you after a migration/merge/clustering/move/rename if PS performs it. But this technote gives you instruction how to do it http://seer.entsupport.symantec.com/docs/337970.htm And this three-week-old technote says, for Unix/Linux platform changes, that you can do it yourself with catalog backup, catalog restore to new system, done. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Files to exclude for hot catalog backups
You know, I've been scratching my head over this one too but I can't remember what the logic was or where I got the info from (or even when). Whatever it was, it made sense to me at the time... I had a case open with support and I threw this question out and I got the following recommendation (though no technote to back it up): Exclude from regular backups on the master: /usr/openv/db/data /usr/openv/netbackup/db The backup selection in a catalog backup has the following paths: /usr/openv/netbackup/db /usr/openv/var/ Relational Database path (/usr/openv/db/data) Neil, I think we're talking about two different things. The recommendation above is don't back up the relational (db/data) or flat-file (netbackup/db) databases with a regular filesystem backup of the master--there's no point since the relational will be inconsistent and a normal restore can restore any of the flat files--which wasn't the case with the cold-catalog-backup format. What I'm talking about below is the original poster excluding any paths from a _hot_ catalog backup: Directories to exclude from hot catalog backup policy (6.0 master) /usr/openv/netbackup/db/images/master I can't think of any reason to do that. Well, none that aren't _really_ convoluted. And I don't know of a way to alter the selection list of the NBU_Catalog policy type--though you could use an exclude_list to accomplish that. Skipping the imageDB in a _cold_ catalog backup used to be the norm when your imageDB got too big to back up in the available time, or when the catalog backup wouldn't fit on one tape. The manual documents the two-phase [cold] catalog backup setup for those situations. But a hot runs concurrently with other backups, can span tapes and can do incrementals so there should be no need to improve the hot catalog backup by excluding things from it. On 1/23/10 1:28 PM, bob944 bob...@attglobal.net wrote: Directories to exclude from hot catalog backup policy (6.0 master) /usr/openv/netbackup/db/images/master Curious about why one would want to do that in a hot catalog backup. Is that a recommendation in a manual that I've missed? (Just checked my 6.0 Admin I and II and didn't see it.) Is it possible you have that as a legacy of a two-phase cold catalog backup implementation? Personally, I don't exclude _anything_ from the NBU-Catalog policy--including being careful not to have a general exclude_list on the master. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Backup Policy Selection for Windows
Sorry I confused... No, you're not confused. You are configuring NetBackup in what seems the obvious way from your background (we _all_ do this, whether it's backup or programming or word processing). The question is I have about 10 windows clients and each client has different mount points to be back-up. C:\cygwin C:\Inetpub C:\net-snmp C:\Inetpub C:\net-snmp C:\PHP D:\Program Files D:\user_files D:\Program Files D:\Webroot N:\Documents and Settings N:\notesdata N:\oracle N:\Program Files E:\Backup E:\Inetpub\visa E:\Inetpub\wwwroot G:\CheetahMailCLA G:\CheetahMailExtraction G:\CheetahMailUnsubUpdate V2 I have got a new request to backup new clients with whole C:\ and D:\ drive. If I use C:\ or D:\ then rest of the clients will also be backed up, which I don't want. Is it I have to create a new policy for specific requirements. And what I'm telling you is that you _do_ want to just back up all the C: and D: drives. One policy, with all the clients, with one selection list. Trading backing up more than the minimum on those drives is cheaper than managing a half dozen policies and their different results, day in an day out. Your brain should be occupied in something more productive than trying to get backups onto one less tape. So, again, make one policy, put all the clients in it, put all the drives (or all_local_drives) in the selection list and go on to other things. If you don't like that advice, then, yes, you need multiple poliies any time that the Attributes, Schedules, Clients and Selections _have_ to be different. And, yes, there _are_ reasons to do that but you haven't put forward any except new requirement.A requirement to back up x and y on clientA and z on clientB can be satisfied most easily with one policy that backs up x, y and z on both clientA and clientB. The selections need not exist on all clients, as long as one selection (within each stream--should you define them) exists. [...] I am new for Windows backup. Need your help to explain on below Here is the requirement. Windows Clients List ... CTLBIS1 CTLFS1 CTL-JAM [...] Backup Selection Example. CTLDS04 Full C: drive NS Full C: drive Full D: drive NS0 C:\WINNT\System32\dns I can understand if we have absolute path selection. But how about C:\ or D:\ or E:\ etc... how system will check from the above clients list which server's C:\ or D:\ or E:\ need to backup and other clients just absolute path. If you have any background in NetBackup (that's not clear from your note), it's no different than backing up any other system, like Unix boxes with a Standard class. A selection list entry, such as c:\ /c/program files/a* c:\windows\system32\drivers\etc /usr/openwin/bin/xclock tells NetBackup to back up that selection (and its expansions if wildcards are used, and everthing under it if it's a directory). If the schedule is a full, it's all backed up. If an incremental (differential or cumulative), only the changed files per the rules for diffs and cincs are backed up. If the schedule has a window and a frequency (or equivalent calendar-based automatic scheuling), invoke it automatically accoring to those rules. You tell NetBackup what to back up (Selection List), on what clients (Clients), when and how (Scheules) by the class (policy) settings. NetBackup will follow the rules you have written in the class definition. If any of this is new to you, you must read at _least_ the Veritas NetBackup Administrator's Guide, Volume 1. For now, create one Win-FS policy, put ALL_LOCAL_DRIVES in the selection list, put all the windows clients in the client list, put in a full and a differential scheule and run them manually Add automatic scheuling and other attributes as your knowledge grows. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Files to exclude for hot catalog backups
Directories to exclude from hot catalog backup policy (6.0 master) /usr/openv/netbackup/db/images/master Curious about why one would want to do that in a hot catalog backup. Is that a recommendation in a manual that I've missed? (Just checked my 6.0 Admin I and II and didn't see it.) Is it possible you have that as a legacy of a two-phase cold catalog backup implementation? Personally, I don't exclude _anything_ from the NBU-Catalog policy--including being careful not to have a general exclude_list on the master. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Backup Policy Selection for Windows
I am new for Windows backup. Need your help to explain on below Here is the requirement. Windows Clients List ... CTLBIS1 CTLFS1 CTL-JAM [...] Backup Selection Example. CTLDS04 Full C: drive NS Full C: drive Full D: drive NS0 C:\WINNT\System32\dns I can understand if we have absolute path selection. But how about C:\ or D:\ or E:\ etc... how system will check from the above clients list which server's C:\ or D:\ or E:\ need to backup and other clients just absolute path. If you have any background in NetBackup (that's not clear from your note), it's no different than backing up any other system, like Unix boxes with a Standard class. A selection list entry, such as c:\ /c/program files/a* c:\windows\system32\drivers\etc /usr/openwin/bin/xclock tells NetBackup to back up that selection (and its expansions if wildcards are used, and everthing under it if it's a directory). If the schedule is a full, it's all backed up. If an incremental (differential or cumulative), only the changed files per the rules for diffs and cincs are backed up. If the schedule has a window and a frequency (or equivalent calendar-based automatic scheuling), invoke it automatically accoring to those rules. You tell NetBackup what to back up (Selection List), on what clients (Clients), when and how (Scheules) by the class (policy) settings. NetBackup will follow the rules you have written in the class definition. If any of this is new to you, you must read at _least_ the Veritas NetBackup Administrator's Guide, Volume 1. For now, create one Win-FS policy, put ALL_LOCAL_DRIVES in the selection list, put all the windows clients in the client list, put in a full and a differential scheule and run them manually Add automatic scheuling and other attributes as your knowledge grows. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] How to replace robot on Netbackup 6.5
I replaced the robot, and changed the magazines with all the tapes over into the new robot, but when i try to inventory, it will not allow me to do so: Insert media failed: barcode not unique in database (36). [...] Two things: 1) have you let NetBackup know these are the same media, moved to a different library? 2) is the new library reporting barcodes the same way the old one did? Did you move all the media to standalone (non-robotic) first? If NetBackup still has the tapes in robot X, and you try to inventory robot Y with those tapes in it, they will be duplicate media, not moved media. Look in the GUI at your old robotic volume group. Select all of the tapes. Move them to standalone (unchecked volume is in a robotic library and --- in volume group name). Inventory the new library. If the barcodes are being returned differently from the new library than they were from the old, you must change the reporting of barcodes in the new library to match the old. If you had mediaID generation rules for the old library, set them up the same way for the new library. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu Digest, Vol 44, Issue 19
so I have a problem with backing up MSCS clusters with ALL_LOCAL_DRIVES backup selections. The cluster nodes not holding the volume attempt to backup the volumes and fail. This was easy enough to fix with policy specific client exclusions, and these exclusions for the clustered volumes get skipped correctly when backing up with Full, or Differential-Incremental schedules, and images for those volumes don't get created. The Synthetic Full schedule however errors out with the cluster volumes with an exist status 671. I don't know the answer but have a suggestion. If you include a known-good file with each stream (however you have them broken up), does the synthetic work? Example: NEW_STREAM f:\ c:\windows\system32\drivers\etc NEW_STREAM g:\ c:\windows\system32\drivers\etc I do this (/etc on *nix and the above on windows, with also winnt vs windows and with the likely system-drive-letters--c: and d: usually) anywhere that I back up an optional mount point which may or may not have anything mounted to it and I want to be alerted by a stat 71. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu Digest, Vol 44, Issue 7
We already are running with abort on error (WOFB_error 0) Do you have a link for the etrack as I can't seem to find it on the symantec site In case a dozen others haven't sent this already... :-) : From a Symantec Technical Advisory of 7/2/2009: - A potential for System State data loss has been discovered in NetBackup Server / Enterprise Server 6.5.4. This occur when the System State is selected for backup and snapshot error control is changed from the default setting to: Disable snapshot and Continue. This issue only affects Windows clients. http://entsupport.symantec.com/docs/327105 - ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Netbackup 6; tuning with [mpx/streams]
clarify a few things. I'm trying to get my throughput up to an appropriate speed to avoid shoe shining and also cut down my backup windows. We have 2 LTO3 drives (with LTO3 tapes, obviously). There are 2 specific objectives I'm trying to achieve: - send multiple jobs from a single policy to a single drive concurrently. I'm thinking 3 concurrent jobs should be fine initially. Once one of the clients finishes, another is added to the job. - send multiple streams (e.g. 3 streams) from a SAN client to a single drive. Once one stream finishes, another is added to the job. I've already got the Global Attribute of 'number of jobs per client' at 99 (I think it was set like this for MS-SQL backups) so this doesn't need to be configured. 'Maximum concurrent write drives' is set to 2 for storage units. For the first objective, I think I need to change a few settings; 'Maximum streams per drive' on appropriate storage units (set to 3), 'Media multiplexing' (set to 3) on schedule in policy. Does this sound about right? Do I need to worry about 'Limit jobs per policy' setting? The second objective is a bit more confusing. Is it simply a case of setting the same settings as the first objective (but on appropriate storage unit and policy), setting the streams in the backup selection tab on the policy and the 'Allow multiple data streams' setting on the policy. I know to get the most out of this I'll need to tweak a few things to get ideal numbers for the settings but just trying to understand exactly which setting I need to configure for the things I'm trying to do. You may be overthinking this. Generate a bunch of streams and send them to storage units with multiplexing set high enough to keep the drives spinning. o don't worry about multiple jobs from a single policy to a single drive, or the like, if you really mean to control it like that (perhaps it is just your phrasing, not your intent). o generate as many streams as you productively can o set STU max concurrent drives to as many as the STU has (generally) o set max mpx for a STU to the value you guesstimate/testimate the drive-type can handle productively o set the mpx in schedules as if that many streams, considering the policy type, its clients and the schedule type (full should be sending a more robust stream of data than an incremental, for instance) were going to one of the STU's drives o remember that the more pools you use, the less sharing of streams to drives will happen; same thing with retention periods o verify your intent is being realized--look at the active jobs and see if you're getting N streams all going to the same mediaID, that all the drives you want to use in a STU are being used... o remember that a given client will be able to send data only so fast; more simultaneous clients is often more effective than more streams from fewer clients... more clients may saturate the network segment... maybe everything goes through one switch or one link to the media server... all the normal throughput items that have nothing to do with STUs and multiplexing o IMO, there is no better investment than time (or dollars if you contract it out) spent in a rigorous tuning exercise to optimize throughput. Taking wild guesses hey, I have LTO3s; what should I set my buffer size to may be better than no change but is unlikely to be optimal. size/number data buffers, representative data, net buffer size, communications buffer size, understanding your network config and hardware... o assess throughput with real or at least realistic backup load and adjust fire. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] I am trying to figure how or best way to do
What I want to achieve is that if backup fails, business owners want information on failed clients i.e. dates of last good Inc, Full back to disk and tape. At the moment am doing this manually and is not fun, have over 300 clients and sometimes we get 20-30 failures. How can I automate this task? This isn't what you asked, but: Rather than work in the mode of failure, go back and figure out what happened in the past, I would take the approach of sending the backup notifications to the business owners for each backup--then there's no digging to be done. Regardless of your notification approach, 7-10% failure is a pretty bad backup day; perhaps the group can help address the causes of those failures. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Notify batch file - End of Job
I've looked again in goodies and admincmd for a script or batch file that can monitor by policy (or by server name) a backup's completion and then send an e-mail to notify a non-administrative user. Are you Reading The Fine Manual? The notification scripts can be suffixed the same way that include/exclude lists can, to be specific to a class or a class.schedule. See the Veritas NetBackup Administrator's Guide, Vol II, NetBackup Notify Scripts. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Backup policy Clarification
Day PolicySL No Sat Full backup 1 Mon Incremental 2 Tue Incremental 3 Wed Cumulative4 Thu Incr?mental 5 Fri Incr?mental 6 Sat Full backup 7 Mon Incremental 8 Tue Incremental 9 Wed Cumulative10 Thu Incr?mental 11 Fri Incr?mental 12 This is a mess. o Terminology is incorrect and ambiguous--NetBackup does not have an Incremental; more precisely, there are two. We can guess that you meant diffs only because your example shows cincs. o If you have diff and cinc schedules and the policy type is Windows-NT, you must specify whether the incrementals are based on archive bit or ctime. o You did not specify whether the schedules are frequency- or calendar-based; It can make a difference, and if both are used, is incorrect usage. o If frequency-based, you did not specify the frequency; the answer is different depending on the frequency values. o You did not specify retention; the answer is different depending on the retention value. Assuming you have reasonable values (frequency, windows, ctime, retention) Due to some policy configuration issue (Sat Full backup 7) did not happened. [...] 1. Will Mon Incremental (8) will happen as per last full backup ie (Wed Cumulative (4)) or with respect to Sat Full backup (1)? Neither. Read the NetBackup Administrator's Guide, Volume 1 | Policies | Schedule attributes tab | Types of backup | Differential incremental backup 2. Will Wed Cumulative (10) will happen with respect to Sat Full backup (1). Yes. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] ALT_RESTORE_COPY_NUMBER file
Our last DR exercise, we tried using this file to switch quickly to our duplicate copies. What we found out is that it works perfectly for Windows clients, but doesn't work for UNIX clients unless bprestore is used from the command line. We were using 6.0 MP5 last year, this year we are using 6.5.3.1. Our master server is on AIX 5.3. Does anybody know if this was addressed in newer versions or is there a way for UI initiated restores to catch this? ALT_RESTORE_COPY_NUMBER in NetBackup 6.5.4 worked as advertised with Solaris servers and both Windows and Unix clients, GUI or command-line restores. We noted that the restore Preview display in the Windows GUI was incorrect (displaying the DSU where the primary copy resided rather than the tapes where the A_R_C_N copy lived) but the restore used the correct copy. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Sharing a library
We have an SL8500 that will be sharing applications (NetBackup and Quantum Storage Manager). The Quantum will be using 9840D tape drives and NetBackup will use LTO4, and we are not hard partitioning the library. The tapes for the different apps have different barcode schemes. The NBU tapes all start with BKP***. shudder I agree with Mr Dyck; sharing an unpartitioned robot between two applications that are not aware of each other, relying on barcodes is not the way I'd go. Not to sound snarky, but NetBackup has an inventory robot command, not an inventory part of a robot command (vmphyinv aside). In my limited experience with partitioned libraries, I'd recommend revisiting that; it's what partitioning was designed to do, IME it's bulletproof, and you don't have to jury-rig procedures for two applications. Is there a way to tell NBU to only add tapes that start with BKP to its volume database? If you were determined to do this with barcodes, you could use a barcode rule to recognize that BKF (you'll never have more than 1000 BKF tapes?) for your production pool and everything else goes to the do not use these tapes pool. An better way would be to define either the 9840 or the LTO drives and media as something other than hcart. Say, manually change the LTOs to anything other than hcart and set up a barcode rule to make BKP media that same type. I have scanned through the manuals and done some volume previews with different options and nothing has worked so far. ACSLS reports both 9840D and LTO tapes as HCART so I cant specify that way... ACSLS would handle this situation (ACS pools and matching volmgr directives in NetBackup), though if you get the INVENTORY_FILTER directive wrong you'll be back in the same boat of inventorying tapes you don't want to see. (Caveat: haven't had an ACSLS setup in ten years so this may have changed.) ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Make drives read only in NBU
Does anyone know if you can make a subset of drives in a library read only in NBU . i.e. I want a specific 10 out of 50 drives in a robot to be read only for duplication purposes? Something else I haven't had the need to do, but if you configure them in a separate storage unit(s) and don't share the drives with a normal STU, that should do what you want. Think of it as hooking up a little 10-drive dedicated robot. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] bpduplicate help
If they're run too closely together, it, for some reason, picks a new tape - it must be a way to delays caused by mounting positioning. I've seen this in backups, too. On my third backup of the set, it picked the first tape. Haven't seen this myself, but perhaps this is affected by how a site has nbrb configured (and the load, of course). OP, have you looked at the nbrb.conf options documented in http://support.veritas.com/docs/300442 ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] SL500 Tape Library Issue
Host OS : Solaris 9 NBU : 6.5.3 NO it gives that logical unit is in the process of getting ready Dude. How long are you going to keep trying to fix this tape library in a software mailing list? You need to get a tech working on your library. Period. By the way, I have seen one bad drive in that same type of library make it fail the same way. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] SL500 Tape Library Issue
Um, the one whose text I quoted? The one on the cc list? The one whose library is down? The one who's posted about it for a week? :-) -Original Message- From: WEAVER, Simon (external) [mailto:simon.wea...@astrium.eads.net] Sent: Thursday, September 17, 2009 11:14 AM To: bob...@attglobal.net; veritas-bu@mailman.eng.auburn.edu Cc: qureshiu...@rediffmail.com Subject: RE: [Veritas-bu] SL500 Tape Library Issue Bob who you talking to ? Me or gureshiumar? Simon -Original Message- From: veritas-bu-boun...@mailman.eng.auburn.edu [mailto:veritas-bu-boun...@mailman.eng.auburn.edu] On Behalf Of bob944 Sent: Thursday, September 17, 2009 1:32 PM To: veritas-bu@mailman.eng.auburn.edu Cc: qureshiu...@rediffmail.com Subject: Re: [Veritas-bu] SL500 Tape Library Issue Host OS : Solaris 9 NBU : 6.5.3 NO it gives that logical unit is in the process of getting ready Dude. How long are you going to keep trying to fix this tape library in a software mailing list? You need to get a tech working on your library. Period. By the way, I have seen one bad drive in that same type of library make it fail the same way. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frozen Tapes
Just out of curiosity, for all those stating to check the logs, to which logs are you referring? Depends on the error, the source of it, the platform and the NetBackup version. I've talked offline to half a dozen people about this in the last week; everybody has their own special places to look. For I/O errors in general: Most Unices put device errors in the messages file, and maybe syslog if configured that way. The Windows equivalent seems to be the Event Viewer. This has to happen since the operating system and its driver facilities are what talks to the drive; NetBackup is an app talking to the OS's facilities. The bptm log. NetBackup's report (say, status 84, write error) is a restatement of what the OS told NetBackup, condensed into an 83 (open), 84 (write), 85 (read), 86 (position) or 87 (close). grepping the output of bperror -all, -media or -problems is a lot easier than looking through the GUI Reports (see below) if you already know what you're looking for. Whereever you find them, TapeAlert messages are worth paying attention to, especially LTOs and the DIRECTORY-type errors. I also habitually peruse the Problems (and sometimes the All Log Entries) report for the previous 24 hours, sorting it by description and skimming it to look for the unusual. My memory tells me I used to find FREEZ[ing it] and DOWN[ing it] there but I don't have any examples. Maybe these changed between 5.x and 6.x. 6.0 or 6.5 introduced new report compilations in the GUI: Check Media Logs, Tape Logs and Tape Summary Here's a technote that explains NetBackup's basic logic on making the call of whether it's a media or drive problem: http://support.veritas.com/docs/317050 ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frozen Tapes
[...] but I don't concur with simply destroying any tape that ever gets an error. Nobody advocated that; toss-versus-try-it-again was after addressing the OP's other possibilities and after checking the logs to see what really happened. 10 minutes of troubleshooting isn't going to kill you Ten minutes of troubleshooting, after one has already expended more than that to determine that it is a media error, that it was more than one on the same tape, that NetBackup's simple is it the tape or the drives logic has decided it was the tape, and in which the testing will either confirm the tape is bad or, worse, not fail because it won't write similar bits to the same spots on the same tracks as the original errors, is not only ten minutes of even a junior admin's time wasted, it is dangerous. So you ran a test and it worked, you re-use the tape and it craps out a long backup the next night. How good a decision was that? Worse, it fails on a restore two years from now when you Really Need That Data. Saving fifty bucks on a new tape was worth losing data? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frozen Tapes
1) NetBackup detects Non-NetBackup data format [...] 2) NetBackup detects that [it is a catalog tape...] 3) NetBackup tried to read/write to the tape and [got write or positioning errors...] ... if the barcode and recorded mediaID don't match... if the tape winds up in the wrong drive, ... Assuming I'm correct so far, then is the proper method of troubleshooting Frozen media to: 1) Ensure there isn't some catalog data on the tape. 2) Ensure that the tapes aren't from some other commercial backup product environment's tape pool (for those of you running multiple commercial backup applications at a single site). 3) Make sure your tape drives have been cleaned recently. No matter what the reason, it should be in the logs; IMO, that should always be your first troubleshooting step: find out why it was frozen and go from there. Special mention to: 4) Use bpmedia -m media id -unfreeze to unfreeze the tape(s), make a note of the tape you're unfreezing, and leave it in the scratch pool to see if it gets used for tonight's backups. No. Either toss it immediately, or, if you _must_ try to re-use it or do root-cause, put it in the None pool until you can thoroughly test it end-to-end error-free. But even if it passes, how much of your time does it take to exceed the cost of a replacement tape? How much time/money will you spend rerunning a backup that fails on that tape again? How much time/money/resume' will you spend if you cannot recover a backup from that tape when you need it. (I see Simon has commented on this and I concur.) Now for my question: Assuming I was correct on my selection criteria and my troubleshooting steps, am I correct in saying that if I came in tomorrow and that media from step 4 was frozen a second time, that it indicates that the media is more than likely defective? Is there any other troubleshooting steps anyone would care to add? Kudos for doing the research you show above. But why did you list all those causes but not look in the logs to see which one caused the error and address it directly? If it's a media-overwrite that you haven't allowed, there's no point in re-running; it'll still be ANSI or whatever. Then, why do you have tapes in inventory that you must preserve but rely on a method that's only a mouse-click away from causing someone a disaster becomes a critical question. If it was media errors, NetBackup already made the educated guess of whether it was drive or media (see the manual), and that'll show up in the logs. If it was a cold-catalog-backup tape, that's in the logs but why/how did it get put into a scratch or data pool? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu] run a script on the Master any time a backup[...]
Need to run a script on the Master server when each backup job starts. I tried parent_start_notify and found that it only runs for policies that have Allow Multiple Data Streams checked. It seems to me that there should be some script that can be run reliably on the master every time a backup runs... Any suggestions? Are you looking at the Veritas NetBackup Administrator's Guide, Vol II, NetBackup notify scripts chapter? backup_[exit_]notify runs on the media server for each backup. Thnaks. I'm looking in NBU 6.5 Administrators Guide for Unix Volume 2: *The backup_exit_notify script runs on the master server. It is called to perform site-specific processing when an individual backup completes.* I confused the issue by trying to show the complementary backup_notify and backup_exit_notify in the same sentence. Look again at 6.5 Admin II on page 173; backup_notify runs on the media server and backup_exit_notify on the master. So, yes: But what I need is a backup_start_notify* *script that runs on the Master when my backup job starts. As far as I can tell, such a script does not exist. Is backup_notify on the media server not helpful for your requirement? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Generate list of file systems in a Policy[...]
On the client, I have a need to generate a list of file systems to be backed up by a particular Policy. There does not seem to be a way to do this from the client. Anyone know better? It's not clear to me what you want. If Ed Wilts' interpretation--run some command on a client that gives the selection list of a policy--his answer is all you need: the question makes no sense in NetBackup. If the client is a media server (or one on which you've installed the Windows Admin Client, which is effectively a full media server installation), you have the tools and permissions to do commands like bppllist to find your answer. Just *why* you'd want to do this, I don't know. Still working from the client end, there is a ton of info in: o the client job tracker probably tells the client something (I've never used it) that is being backed up in realtime o the BAR GUI tells the client everything that has been backed up, how and when--but not by policy o the ~/veritas/netbackup/logs directory has the VxUL logs by default and the bpbkar log if you create the directory; I'm not used to looking in VxUL logs, but the bpbkar log has the policy information as well as everything that got backed up o bpbkar logs also have the selection list (the log entries that determine what FS/volume a selection occupies and the CreateSnapshot entries (client-dependent, I'm sure) o bplist o all the above, obviously, tell you what has happened, not what will happen in the future o client-notification email o bpstart scripts And the way I first read your question, something like how do I make sure I back up all the filesystems on a client without knowing what they are in advance, the answer is you put ALL_LOCAL_DRIVES in the policy's selection list. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Transition from LTO1 to LTO4
We're considering switching from our old ADIC Scalar 1000 (with LTO1 drives) to an HP MSL8096 (4 drives). The logical step will be to go the LTO3 route so we can read our old media but ... considering we do backups 99.5% of the time and restores 0.5% of the time...I'm considering getting three LTO4 drives and the fourth one: an LTO3 drive (just for backward readability). On Netbackup... Is it just a matter of labeling the drives and corresponding media with the same HCART# so they don't get mixed up? Yes, if you're going to mix them. LTO4s are automatically detected as hcart and so are LTO1s. So when you add the LTO4 drives, change them to anything else--LTO2... 8MM, whatever, and set up barcode rules so that all the LTO4 tapes you'll ever add get the same type. Later responses showed that you didn't relish keeping the old library forever. Maybe it's time for a real technology refresh. Have you looked at your data to see if it's feasible to keep your old library long enough to duplicate your long-term retention data to LTO4 and then get rid of the LTO1s when you've duped everything that hasn't already expired? Much cleaner way out, and you can then toss the old tapes and eliminate the storage costs. You could also do this with your other approach--one LTO1 in the new library--until done, then put the fourth LTO4 back in. This may well reopen up another favorite topic: retention and the costs/logic/risks in keeping and trusting the usefullness and recoverability of old data. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] run a script on the Master any time a backup[...]
Need to run a script on the Master server when each backup job starts. I tried parent_start_notify and found that it only runs for policies that have Allow Multiple Data Streams checked. It seems to me that there should be some script that can be run reliably on the master every time a backup runs... Any suggestions? Are you looking at the Veritas NetBackup Administrator's Guide, Vol II, NetBackup notify scripts chapter? backup_[exit_]notify runs on the media server for each backup. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Media Server offline for disk (now DSSU stuff)
The high watermark serves two purposes - (1) It prevents any new backup jobs from being sent to the STU, and (2) It triggers image expirations on the STU. The function #1 is not applicable to Disk Staging STU; however, the functionality #2 should work for it. So, if you lower HWM for the Disk Staging STU, it would trigger image expirations earlier so that when the backup jobs are sent to the STU, NetBackup does not have to expire images then (potentially saving time for the backup). Hope this helps. I agree, the documentation is little unclear on this. That's an understatement! 6.5 DSSU behavior is re-described in technote 316679. HWM, LWM and DS[S]U behavior have never, to my knowledge, been correctly described in one place, and the incorrect descriptions have been most unhelpful. Let's hope this technote is a step in the right direction. As long as this has morphed to DSSUs, I had to look up how to display actual available space on my 6.5 Basic Disk DSSU setup now that the .ds files are gone. In case it helps anyone else, it's something like ~/admincmd/nbdevquery -listdv -stype BasicDisk -dp DSSU-a -D See the 6.5 NetBackup Administrator's Guide, Volume 1, p 242 or so. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas MSEO product
We are considering using the Veritas MSEO (Media Server Encryption Option) to encrypt our tapes for offsite storage. Does anyone have any experience with this product? Yes, in multiple-server/multiple-agent configuration which was very easy to set up and customize, and not too bad to maintain. Can't comment on throughput impact as there was no before data. rcarli...@serverwarecorp.com: I am a big fan of the product Same here. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Unsupported file system type in BackupExec
I am getting the following: 18:03:22 INF - Skipping Backup Exec backup having backup set number 180. Unsupported file system type :47 18:03:22 INF - Skipping Backup Exec backup having backup set number 181. Unsupported file system type :59 18:03:32 INF - Skipping Backup Exec backup having backup set number 182. Unsupported file system type :70 18:04:56 INF - Skipping Backup Exec backup having backup set number 183. Unsupported file system type :70 18:04:58 INF - Skipping Backup Exec backup having backup set number 184. Unsupported file system type :70 18:06:36 INF - Skipping Backup Exec backup having backup set number 185. Unsupported file system type :47 when I import some BackupExec tapes to Netbackup 6.5.3 and wonder what I'm not getting in. Can anybody shed some light on this as the Symantec webiste wasn't very informative Technote http://support.veritas.com/docs/295433 is the one that rules--I assume you found that? A BE admin was able to tell me what was on tapes I was trying to import, by which found that type 47 is a Volume Shadow Copy type.and I had other type numbers that corresponded to other Microsoftisms, like System State. None of the unknowns were filesystems or SQL Server, which is what I was after. The technote has been updated with 6.5.4 info; you should now be able to import BE VSS and System State, and w2k8 and compressed in case your type 59 and 70 are one of them. Worth a tech support call? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Synthetic Backups
There is a case open with Symantec about this issue, and I am told an EEB will be made available soon. You are not supposed to have to keep a full around, No. You _must_ have a full plus subsequent TIR and diffs or cincs to synthesize a new full, but it does not matter how the full was generated. (That was for clarity--I assume you meant to keep a traditional full around.) but once the full expires a new full is taken automagically. I've used synthetic fulls and cincs for years, demonstrated them in this mailing list and found the documentation to be pretty darned clear. IME, they always work. What's been missing from this discussion (apologies if I've missed it) is detail about what setup you or the original poster have. What release, what platform, what is the policy config, what images and TIR are available when it doesn't work for you, what's the bpdbjobs output (easy way to track the progress of a synth and the components it is using)? I've confirmed it is working properly in 6.5.3.1 Solaris: no natural full exists, nor is one created--just the current synthfull + cinc = new synthful. Is this (your experience, and the support case and EEB) a 6.5.4 issue? QuoteSo it appears I need to keep the Full backup around even though the Synthetic Full is suppose to be its equivalent. The Full backup runs on Day 5 even though I don't have it scheduled. In Fact Full are manual. Synthetics are scheduled. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frequency Based Schedules
[bob944] and the longest frequency will win--the monthly will run. [Rusty] Bob, that's what usually happens, but that's not why. The trump card is the retention level, which usually is matched to a longer frequency, but may not always be the case. [bob944] I'm always glad to learn new things, but I'm not sure this is one of them. To me, the logic _has_ to be that longer frequency wins, else a longer-frequency schedule would never win out--and this has always been the behavior that I've observed and relied upon. But since I have never set up, say, a weekly full with a retention shorter than its matching daily diff, maybe it really works as you say, but I'll need to see it to believe it. Any interest on ... oh, I dunno.. how about a bottle of single-malt Scotch to the winner? ;-) One test is worth a thousand expert opinions. I win. When multiple schedules are due; the schedule with the longer frequency interval runs; retention has nothing to do with it. Details on request, but here's the edited summary of the class and bpdbjobs: # /usr/openv/netbackup/bin/admincmd/bppllist TEST-std-freq -L Policy Name: TEST-std-freq Policy Type: Standard (0) Residence: TEST-dsu Client/HW/OS/Pri: mouse Solaris Solaris9 2068001349 0 0 0 ? Include: /etc/hosts Schedule: full-f2h-r1d [goofy name guide: freq:2hours-ret:1day Type:FULL (0) Frequency: 0+ day(s) (7200 seconds) Retention Level: 22 (1 day) Day Open Close W-Open W-Close [Sun-Sat] 020:00:00 028:00:00 020:00:00 028:00:00 Schedule: diff-f1h-r1w [freq:1hour-ret:1week] Type:INCR (1) Frequency: 0+ day(s) (3600 seconds) Retention Level: 0 (1 week) [Sun-Sat] 020:00:00 028:00:00 020:00:00 028:00:00 # date Wed Jun 3 19:58:01 EDT 2009 full has a frequency of two hours and a retention of one day. diff has a frequency of one hour and a retention of one week. # /usr/openv/netbackup/bin/admincmd/bpdbjobs|head -20 Time Type State StatusPolicy Schedule 01:00Backup Done 0 TEST-std-freq diff-f1h-r1w 00:00Backup Done 0 TEST-std-freq full-f2h-r1d 23:00Backup Done 0 TEST-std-freq diff-f1h-r1w 22:00Backup Done 0 TEST-std-freq full-f2h-r1d 21:00Backup Done 0 TEST-std-freq diff-f1h-r1w 20:00Backup Done 0 TEST-std-freq full-f2h-r1d Summary: despite its shorter retention, the schedule with the longer frequency interval always wins out. H. Macallen is a little pricey; Glenlivet will do nicely. :-) ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frequency Based Schedules
and the longest frequency will win--the monthly will run. Bob, that's what usually happens, but that's not why. The trump card is the retention level, which usually is matched to a longer frequency, but may not always be the case. I'm always glad to learn new things, but I'm not sure this is one of them. To me, the logic _has_ to be that longer frequency wins, else a longer-frequency schedule would never win out--and this has always been the behavior that I've observed and relied upon. But since I have never set up, say, a weekly full with a retention shorter than its matching daily diff, maybe it really works as you say, but I'll need to see it to believe it. Any interest on ... oh, I dunno.. how about a bottle of single-malt Scotch to the winner? ;-) One test is worth a thousand expert opinions. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Frequency Based Schedules
Hello Guru's ... We are struggling with this scenario. We are a Netbackup 6.5.3.1 shop and recently discovered that using Frequency Based Schedules that a Monthly Full backup and a Weekly Full backup for the same client will run on the same weekend. What are we doing Wrong?? You misunderstand how the (frequency-based) scheduler works. If the monthly and weekly schedules start at the same time, it will work as you expect. Assuming that you do both on, say, Friday night (I let mine float but that's another issue), you might set up: diffSun-Sat, 2100-0600, freq 1 day full-wk Fri 2100-0600, freq 1 weeks full-mo Fri 2100-0600, freq 4 weeks The scheduler will find, obviously, only the diff to run at 9pm every day but Friday. On Friday night, the weekly will run because it is due (freq requirement has been met) and will trump the diff (yes, you could just skip putting the diff in but then you have two changes to make if you move the full to another day). Every four weeks, the scheduler will evaluate a due diff, a due weekly, and a due monthly, and the longest frequency will win--the monthly will run. It sounds as if you might have set your schedules up the way I do (on purpose), such as: diffSun-Sat, 2100-0600, freq 1 day full-wk Sun-Sat, 2100-0600, freq 1 weeks full-mo Sun-Sat, 2100-0600, freq 4 weeks beause I don't care that fulls run on any day (there are enough to balance out through the week naturally), I don't care that on the first night the monthly runs, the next night the weekly runs, then six nights of diffs and a weekly..., that there will be a weekly within the same week (though not on the same night) as the monthly. If I run a manual weekly for some reason, it won't run again for a week after that. This logic is simple and, um, logical to me; calendar-based scheduling is _far_ too much work for me to manage... the midnight-split on calendar was so amateurishly implemented in 4.5... calendar-based just doesn't make sense to me except for the (IMO) *very* few requirements I have to back up X based on a business process that produces data in X on a certain date and has to be captured immediately and handled specially. This isn't meant to be an anti-calendar rant, just an illustration of the different ways of thinking about scheduling. I know if we used Calendar based Schedules for the Monthlys and excluded these dates from the Weeklys it would solve the problem but surely there must be a better way. Thanks in advance... Don't mix calendar and frequency-based in the same class, though. I think there's still a statement to that effect in the admin guide vol 1--originally put there because, as I understand it, the freq and cal schedulers were (still are?) completely different processes and didn't share. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Troubleshooting Network Issues
I need a way to call a script when a backup first gets kicked off on the master. Bpstart_notify appears to work on the client, but I'm having these strange network error related to when the backup starts - before bpbkar32 is initialized on the client. Is there another script on the master / media that responds to jobs firing similar to bpstart_notify? They're all in the NetBackup Administrator's Guide, Volume II. Look for NetBackup Notify Scripts. Depending on how weird a problem you're working on, one of session, parent or backup should be useful. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NB 6.5.3 slow response for administrative tasks
What I'm seeing is really poor response when adding or changing backup policy specifications. Sometimes to update the backup windows via the Admin console, it takes longer than 2 minutes - even when there is no other activity. I've also seen significant pauses when activating/deactivating policies via the command line. This sounds suspiciously like things that happen with name-resolution issues (as noted by Brown), especially storage unit hostnames (a bogus STU would add 300 seconds to the start of any policy, for instance). You might mention which of NetBackup's dozen incoherently/inconsistently named consoles you are using, where the console is being run, and whether the command-line equivalent of all the oddities exhibit the same delays. vxlog output going back the two minutes or so it takes to run one of the operations might be useful if all the Troubleshooting Guide name resolution tests are successful. And pbx, ports, firewalls, the goofy Linux inetd and patches, ... ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu] Netbackup, VCB, vRanger Policy Setup
All, Just trying to get a policy within NBU setup, and having some difficulties getting it setup. Currently I have the following setup: Policy = XX_VMTest01 Policy Type = MS-Windows-NT Schedule = Full Client is the NBU Media Server (also the master) Backup Selection - ?? I need help here on this part I have also created a bpstart_notify.XX_VMTest01.bat and it is placed in the Veritas\Netbackup\bin with the following contents: If %2 == XX_VMtest01 goto backup_test Goto end :backup_test CALL E:\vizioncore\esxRanger Professional\esxRangerProCli.exe -virtualcenter vc2://Folder=group-v10270 -copylocal E:\mnt - drives:db -zipname [config]_[weeknum] -onlyon -vmnotes -noquiesce - totalasync 10 -hostasync 2 -lunasync 3 -vcb -failbacktolan -totalvcbasync 2 -diffratio 50 -maxfullage 1 -retendays 1 Goto end :end vRanger works independently from NBU VCB, if anyone could give some insight it would be greatly appreciated. Perhaps I missed something earlier, but why are you using two completely different backup products to do this? I know _of_ VRanger, but haven't used it. OTOH, I _have_ used NetBackup and VMware's VCB and that's all that's necessary (or just install a normal NetBackup client inside a VM if you need to do database quiescing or whatnot). o set up VirtualCenter and/or ESX credentials once in NetBackup o set up a proxy server once o configure VMware policy in NetBackup per the 6.5.2 or 6.5.3 update guide, option 3 for full VM plus incrementals, put in a full (and an incremental if you like) schedule, discover/add in all the VMs you want, ALL_LOCAL_DRIVES, Done. Take the rest of the day off. It's just too easy. No VRanger, no kludgy windows batch files, no synchronization needed. Do you have some special requirements that are filled by the multi-vendor, multi-phase backup and restore arrangement you're trying to convigure? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] MSEO - How to understand the images are
We are trying MSEO option. Is anyone know that how we can undaertand the images are Encrypted by MSEO. or Is there any commad to list/show images which are backed up by MSEO? In an evaluation, I customized the audit log (that's in the manual) and then used the audit log information in the Security Server's /var/log/meso.log to produce reporting which included the mediaID, backupID, keygroup and key for this purpose. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] New Library Configuration Parameters
Hello, my department just bought a new Spectra Logic tape library and I'm tasked with installing it. Where do I get the information for the configuration parameters like robot type, drive type, etc? Is the provided by Spectra Logic or Symantec? You shouldn't need to. The device configuration wizard does a very good job of figuring this out. Read the manual or just click your way through the startup screen steps. Patch after installation and before doing any setup. This includes the Mappings patch. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] bpimport - Phase I Issue
[expired phase 1 imports] Is this image really expired? I seem to remember something about Phase I imports only lasting 7 days, but I thought setting all the retention periods to infinity would help with that. My next step is to re- import the same media and see if I can restore then. Anyone run into this before? I've run the same command against media that I've scanned just a few days ago and I've got the same issue. I'm not really sure where to go with this one. We cover this every year or two on the list. You have one week from the time of the phase 1 to do the phase 2. This used to be in the Admin Guide, but I didn't see it on a quick spot-check. You don't need the manual to remember it once you've done a phase 1 on hundreds of tapes, then found out--as you have--that you have to do it all over again because of the one-week limit. DAMHIKT. I wrote something before on a process that worked best for me, but briefly: if you have many days of phase 2 imports to do, start managing it by the phase 1 requirements: make sure you have done a phase 1 on all tapes for the image(s) in question--you'll have to start over if you've only phase-1-imported, say, three of four required tapes before trying to run phase 2. Creating the NOexpire file will probably keep phase 1s from expiring but I've never tested this. You can also futz with the expiration data in the phase 1 metadata file or in a hundred other ways copy/move/modify/backup/restore the metadata. But the easiest way is to manage the process with the one-week clock in mind. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] question about checkpoint
I have only a question with the checkpoint in policies : I have test a backup with use checkpoint positionning all the 15 minutes and after about 20 minutes i cancel this job, but it'is not possible to resume job with checkpoint position! is it normal that the job can't restart with the last checkpoint ? or exist a another soluce to do it ? In addition to what Ed said, you may not have completed a checkpoint. The checkpoint is made when a client reaches the end of the file being backed up when the time value is reached. If a large or slow file started at, say, 12 minutes past the start or previous checkpoint, and took 20 minutes to back up, the checkpoint would be at the 32-minute mark. If, somehow, it were critical to know this, look at the FRAGMENT lines in that backup's metadata file. One that is not that storage unit's Fragment Size probably signals that a checkpoint.was made. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Veritas-bu] Unable to find image in Restore/Catalog
Yes, correct i am looking out for specific date. But when i went in the directory. There are files for that particular date. # pwd /opt/openv/netbackup/db/images/EKMS/123600 # cd .. # ls -ltr total 16 drwxr-xr-x 2 root other512 Mar 6 13:41 123000 drwxr-xr-x 2 root other 1024 Mar 9 13:48 123300 drw-r-xr-x 4 root other 1024 Mar 13 04:44 123600 drwxr-xr-x 2 root other 1024 Mar 22 04:31 123400 drw-r-xr-x 4 root other 1024 Mar 25 02:16 123700 drwxr-xr-x 4 root other 1024 Mar 26 04:34 123500 drwxr-xr-x 2 root other512 Mar 27 16:35 123200 drw-r-xr-x 4 root other512 Mar 28 15:13 123800 # cd 123400 # ls -ltr total 18 -rw--- 1 root other 4040 Feb 14 18:03 EKMS_1234601889_FULL.f -rw-r--r-- 1 root other 4429 Mar 17 16:27 EKMS_1234601889_FULL # cd .. / 123700 # ls -ltr total 16 drwxr-xr-x 2 root other512 Mar 6 13:41 123000 drwxr-xr-x 2 root other 1024 Mar 9 13:48 123300 drw-r-xr-x 4 root other 1024 Mar 13 04:44 123600 drwxr-xr-x 2 root other 1024 Mar 22 04:31 123400 drw-r-xr-x 4 root other 1024 Mar 25 02:16 123700 drwxr-xr-x 4 root other 1024 Mar 26 04:34 123500 drwxr-xr-x 2 root other512 Mar 27 16:35 123200 drw-r-xr-x 4 root other512 Mar 28 15:13 123800 # - I am looking out for 22nd March data. # /usr/openv/netbackup/bin/bpdbm -ctime 123700 123700 = Fri Mar 13 23:06:40 2009 # /usr/openv/netbackup/bin/bpdbm -ctime 123800 123800 = Wed Mar 25 12:53:20 2009 # Any March 22 backup will be in the 123700 directory. If you intended to list that directory with the cd .. / 123700, you didn't. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NetBackup 5.1 and SQL Online Agent - Some
Situation: Win2k3 SP2 Master and Mutliple SAN Media Servers. I have many VM Machines that are doing backups via the lan at this stage. VMCB will be looked into later this year. Problem I am seeing: In particular, 2 clients have 500 SQL DB's on them and the behaviour of NetBackup seems to be: 1) A Schedule kicks off for SQL 2) Another job kicks in for the Default Application Backup and it backs up each single DB You do realize, right, that the full/incr schedule doesn't actually back anything up--it just tells the client to run the client script specified in the selection list? And that the file should be SQL Server backup commands that are executed on the client (and can be initiated just as easily from the client by any other means that suit your needs)? And that the client initiates a user backup request to the master which validates the client against allowed/applicable user backup schedules--_that_ is what initiates the D_A_B job(s)? So the fix you seek is at the client end. Break out a Microsoft SQL Server book and an editor, the Veritas NetBackup for Microsoft SQL Server Administraotr's Guide (or use the too-simple-to-screw-up NetBackup client GUI) and change or add a script that does what you want: stripes, multi-streaming, blocksize, shared buffers, ... Assuming you set the client and policy job limits per the Administrator's Guide, of course. Or is this behaviour I am seeing normal. It's normal if the script uses the default installation values. A backup or SQLServer admin who knows his stuff can make SQL Server backups rock if the box is capable of it. As I see others have said, open the admin guide. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] How to properly calculate the Catalog
That formula in the manual is completely worthless. I can't
believe they still publish it. The SIZE of the data you're
backing up has NOTHING to do with the size of the index. What
matters is the number of files or objects.
[...]
I could backup a 200 TB database with a smaller NBU catalog than
[snipping the obvious (though perhaps not to NetBackup beginners):
since 99% of the catalog is a list of paths and attributes of files
backed up, a list of a million tiny files and a list of a million
giant files are going to occupy about the same catalog size.]
To get the real size of the index:
1.Calculate number of files/objects [...]
(I say 200 bytes or so. The actual number is based on the
average length of your files' path names. 200 is actually
large and should over-estimate.)
Um, to quote some guy...
That formula [...] is completely worthless.
Just kidding. Files-in-the-catalog times 200 is very old-school.
And right out of the older manuals which used 150, IIRC.
There are a couple of things to take into account here.which made me
move away from files*150--aside from the drudgery of figuring out
file-count stats per client per policy per schedule per retention.
1. smaller sizes using the binary catalog introduced in 4.5. No
idea what the file formats are, but in perusing various backups,
there appears to be a lot of deduplication of directory and file
names happening.
2. catalog compression, which may or not be important to the
calculations. Using compression, IME, reduces catalog size by
two-thirds on average, thus tripling catalog capacity for users with
longer retentions.
3. Full backups versus incrementals. The *imgRecord0 file is
usually the largest binary-catalog file for a backup; in an
incremental it is not appreciably smaller than in a full. So, in
the event that an incremental finds only, say, 10 changed files in a
100,000-file selection, the size of the catalog entry for that
incremental is nowhere near what one would expect from a small
backup--it's much closer to a full.
Though this is little predictive help to a new NetBackup
installation, getting a handle on catalog sizing for existing
systems is too easy: the number of files backed up and the size of
the files file are each lines in the metadata file. Dividing size
by files doesn't _really_ give you the number of bytes per file
entry, but it yields a great planning metric. This script:
#!/bin/sh
cd /usr/openv/netbackup/db/images
find . -name \*[LR] | \
while read metaname
do
if [ -f ${metaname}.f.Z ]
thenCOMPRESSED=C
elseCOMPRESSED=
fi
awk '
/^NUM_FILES/ { num_files = $2 }
/^FILES_FILE_SIZE/ { files_file_size = $2 }
END { if ( num_files 2 files_file_size 2 ) {
printf %4d (%s %11d / %11d ) %s\n, \
files_file_size / num_files, \
compressed, \
files_file_size, num_files, FILENAME
}
}
' compressed=$COMPRESSED $metaname
done
can be used to get a handle on catalog sizing. Sample output:
(first column is files_file_size divided by files in the backup; C
is for a compressed catalog entry, followed by the files-file size,
number of files and the pseudo-backupID)
33 (C 331651 /9884 )
./u2/123500/prod-std_1235118647_FULL
36 (C 1654789 / 45203 )
./u2/123500/prod-std_1235119960_FULL
33 (C 331497 /9884 )
./u2/123500/prod-std_1235202798_FULL
36 (C 1655827 / 45223 )
./u2/123500/prod-std_1235203103_FULL
33 (C 74293 /2236 )
./u2/123500/prod-std_1235286142_INCR
35 (C 79497 /2212 )
./u2/123500/prod-std_1235286246_INCR
33 (C 332661 /9884 )
./u2/123500/prod-std_1235808812_FULL
36 (C 1657187 / 45245 )
./u2/123500/prod-std_1235810235_FULL
32 (C 73757 /2236 )
./u2/123500/prod-std_1235890933_INCR
35 (C 79389 /2212 )
./u2/123500/prod-std_1235891054_INCR
101 ( 1001512 /9884 )
./u2/123600/prod-std_1236498790_FULL
102 ( 4644469 / 45185 )
./u2/123600/prod-std_1236498992_FULL
446 ( 1001548 /2243 )
./u2/123600/prod-std_1236664989_INCR
2092 ( 4646723 /2221 )
./u2/123600/prod-std_1236665069_INCR
Notice the last and third-last lines. They are a full and a diff of
the same filesystem. imgRecord0 makes up 3.25MB of the 4.64
files_file_size whether it's a full (45,185 files) or an incremental
(2221 files).
To loop back to the middle of this, I find that 100 bytes/file
uncompressed (35 compressed) is a good planning value for fulls on
most systems; the exceptions tend to be systems where the apps use
pathnames longer than any human would want to type. (The imageDB
part of hot catalog backups produces a files_file_size / files
metric more like 170.)
More than most people wanted to know.
Re: [Veritas-bu] SQL issue
Update from my DBA - these are transaction log backups, and he added a new database - since the TL will fail if there is not a backup, that is what caused the issue. Why would NetBackup not see this? He claims his SQL scripts ran through and finished, but the parent jobs never acknowledged this. I have received a recommendation to get a patch for the bpbrm, but that was based on 5.1 mp6. Shouldn't 6.5.2A have any patches from 5.1? NetBackup support response is that this is not their problem. Any ideas? Add a database; do a full. Is there a problem after that? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Change in NET_BUFFER_SZ
Will this take effect w/out stopping/starting the primary bprd on the Master? I'd imagine no need to start/stop anything on clients or media servers, but don't know about the master. NBU 6.5.3 This should be a bptm thing, set up per-backup using the setting in the file at the time. grep for recvbuf: 01:54:38.953 [8964.6088] 2 io_set_recvbuf: setting receive network buffer to 263168 bytes ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Schedule SQL Log backup in Netbackup
Actually we are implementing the Netbackup SQL Agent for all our SQL databases. So iam finding the best method to implement the Full, Diff tlog backup. Full - Needs to run weekends Diff. - Needs to Run week days TLog - Needs to run every 15 minutes This is the requirement. And, IMO, an exceedingly bad idea. If you have a NetBackup instance solely to be a little log-backup utility for the database, fine. Otherwise, this a) holds your enterprise backup solution hostage to a database, b) if backups every 15 minutes are a requirement, you'd better have a _very_ high-availability backup solution, c) the database should be writing logs to some decent storage that is not in the same place as the database, d) NetBackup isn't a real-time app--you couldn't afford to use NetBackup in any meaningful way, in the way it was designed to work, because your/their scheme assumes that NetBackup can reliably start the transaction log job every 15 minutes--the load will be low enough, there will always be free tapes and drives, e) why 15 minutes?--surely the database is really busy at some point of the day or night and really idle at others, so this sounds like somebody's half-baked guess, ... Suggest a better course is advising the DBAs to do their job and write logs to some reliable storage, not where the database lives, and have enough space that the logs will not fill it in the longest backup outage you plan for. Then give them a user schedule with appropriate windows and other parameters in a SQL Server policy so that they can initiate the log backups when needed. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Status = 12 (file open failed) in bpbackup
So anything other than a status 0 abbends our job. What's Is that the behaviour you want? You didn't ask, but that's probably not what I'd want my scheduler process to do. I can't find anywhere in any log which file bpbackup is actually choking on. Sometimes the status 12 happens before any files are backed up, sometimes it's in the middle of the backup. The reference you need is the Veritas NetBackup Troubleshooting Guide. For the specific problem of what file, create a bpbkar log directory on the client(s) and set the VERBOSE level to greater than zero; that will cause bpbkar to record the file names in the log as it processes them. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Netbackup 6.5 and Quantum Scalar 50
I have a server running Netbackup 6.5 on Windows Server 2003 x64, and a Quantum Scalar 50. I have the robot and both drives in the Scalar visible in Netbackup, I can move tapes around with the Scalar's web interface, and if I manually move a tape into the Passthrough ports on the Scalar, I can right click on the robot, select Inventory Drive, and those tapes will be added to netbackup. However, I can't see a way to move tapes that I've previously scanned from the Load Ports (the magazines on the side) into the drives. Right now, it seems like if I want to do a backup job, I have to manually move tapes from the load ports into the two passthrough slots, and only then will the robot find the tape and load it into the drives. Is this normal behavior? What happens if I have a job that takes 3 tapes? Will it stall out? I don't know your hardware, but are you selecting Empty media sccess port prior to updating in the inventory GUI? If the Scalar 50 is like some other small scalar that I saw in a demo, when you put tapes in the slid-out magazine, a little dialog comes up on the control panel asking you to choose between X and Y. One of those is system, IIRC. And I think the one you want is the pushbutton on the right side. Without doing that, tapes stay in the magazine. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Getting Netbackup 6.5.x job information
argos2-cls:/#vxlogview -X jobid=9109V-1-1-12 There are no records to be displayed.argos2-cls:/# What does that means ? What else can I do to obtain information for such job? How to obtain activity monitor as well as verbose activity info via command line ? Read the man pages (in the Veritas NetBackup Commands manual) for bperror and bpdbjobs; they are the workhorses of what happened from an admin-console sort of view. Consult the NetBackup Troubleshooting Guide for troubleshooting guidance and a lot of information about reports and logs. The normal logs are in per-process directories that you create under ~/netbackup/logs and are usually examined with an editor or text tools. vxlogview was added in NetBackup 6 to process a new log format for all the new processes that didn't exist in NetBackup before (the unified logs). It's a clever way to make you study the manuals to craft a horribly inconsistent, unintuitive command line, the output of which, if ever you get all the options right, will not answer whatever question you had in mind. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Backup not listed in history in BAR
While trying to restore backup for a client I could not find all the backups taken (the client has a weekly backup schedule). There were no failures for any of the backups. But Do you know that for a fact? How? Many operators/admins/managers destroy information by deleting failures from at least the activity monitor as a normal (bad) part of operations. Can you see them in the activity monitor or bpdbjobs? I am unable to see all the backups for the client. For example, if I want to restore a backup taken three months back, it is not listed in the history of backups in BAR. Do I need to import the backup (or catalog) from the media? How do I know which media is having a backup of a particular date, Since you mentioned BAR, I'd suggest the GUI: Reports | Tape Reports | Tapes Written. That's in newer versions of NetBackup; since you didn't provide any useful information, I won't try to guess what your customer is running. and then import that backup? RTFM and learn about importing. Surely you have someone who knows NetBackup if you have a *customer* for whom you're providing NetBackup support? The backup(s) might never have run: ... NetBackup or the master was not running for some reason ... Some NetBackup bug like the daylight-savings one missed the backup(s) ... The policy isn't set up correctly or ... The operator disabled the policy/policies, or cleared the window, or deleted the client, or changed the frequency, or changed the expiration or advanced the system time, ... perhaps a dozen other human things that could have kept a client backup from running when you think it should have The backup ran and somebody expired it. The backup failed and somebody deleted the job from activity monitor or job DB. You have not correctly configured BAR (server client, backup type, date range. Forget the GUI: change directory to images/name-of-client/time-directory-of-interest and list the files for the period you care about (use bpdbm -ctime to figure out the times or use the timestamps on the metadata and files files). If you have logs for the time period, you can theoretically reconstruct the backup time in detail, from scheduling through expiration. But that's a big if and a lot of work. Bottom line: for there to be an image to restore, all these things must be true: it must have run, it must have been successful (0 or 1), it must not have been expired by the retention parameters or manually/programatically, and you must be looking for it properly. This should spawn another discussion about operations checking to see if what was intended to run, actually did. And documentation. And logs. And permissions. And tracking. And security. And training. And SLAs. And quality. And responsibility. And auditing. And support. And competence. And drive-by webbies on this mailing list. And ... ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] error
Dear All i get the following error [ERROR] v-125-331 can't create sol 10 shared resources Tree on sol 5.10 boot servers can any one answer 1. I don't know the answer to your problem. 2. You have several BMR problems so far. What does support say about them? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Buffer settings LTO II vs LTO IV
In an environment with HP LTO II's, I've been using for some time and with good results: NET_BUFFER_SZ (65536) SIZE_DATA_BUFFERS (262144) NUMBER_DATA_BUFFERS (32) My question is has anyone seen the need, or any performance improvement by changing these setting after implementing a system with LTO IV's drives? The drives in question are HP, but I'll look forward in hearing about any experience that resulted in a performance fix or increase. This isn't quite the question you asked, but IMO, the _only_ way to know optimal settings for a given environment is to test. a) find the best your media server(s) can do with realistic data under ideal conditions (vary number, size buffers and mpx level to find best setting for each) b) test with repeatable data and conditions that replicate the environment for which you are tuning, varying primarily network/comm buffers and job load to try to reach levels found in a That said, in my limited LTO4 experience (though not with HP's LTO4s) 32x256K is not enough to run LTO4s at native speed without relatively incompressible data, a killer media server and a fair amount of multiplexing. In testing and real life, let the bptm logs guide. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NBU 6.5.3 on Windows and VMWARE Backups
I am configuring VCB backups for NBU 6.5.3, for FlashBackup-Windows policy, there are 4 options for backup types under Snapshot Client Options. These are 0 - File 1 - FullVM 2 - Mapped FullVM 3 - FullVM and File for Incrementals From reading the docs, Option 3 - FullVM and File for Incrementals, looks like best of both world, it will let me restore single files and restore full guest VM. It also allows incremental level backups. Is there any disadvantage when using it compraed to other types. I am trying to avoid running 2 or more backups to completely protect a VM. 3 is the correct choice; it does exactly what the manual says it does: a snapshot backup of the vmdk for a full (and from which you can restore individual files or the whole VM), and a normal, file-based backup for incrementals. Set up the schedules in the same policy, throw windows and clients at it and it will work like you expect NetBackup to work. It worked as advertised in my test environment, though my restore testing was minimal. An aside: Restoring files to VMs is, to somebody not used to VMware, kludgy, so do read up on how to restore and test that part before you need it. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NBU 6.5.3 Windows Java GUI for the Admin Console
Mr Major said: Bob, I have to disagree with you, for once. It's okay; I'm not invested in it. :-) And thanks for the implied kudo. You can only use the java version with that same major revision; and The java client is forward and backwards compatible within the same major revision - for the most part, there are things that may not work/populate if the Master is at a higher level than the Java client. I'm fine with that. That in my experience it seems backwards-compatible within all levels of 5.1, within all levels of 5.0, within all levels of 6.5 is what I at least _meant_ to convey with IME, all patch/MP versions are backwards compatible within that major version (5.1, 6.0, 6.5). it is not backwards compatible to other revision levels (6.5 is not backwards compatible to 6.0 or lower, etc.). And that's the part I didn't know about (never tried it) but thought mightt work: It's possible, even likely, that each major version understands the previous versions Not as likely as I'd thought, apparently. Now we know that it's possible, but not true. :-) Thanks for clearing that up, Rusty. -Original Message- From: rusty.ma...@sungard.com [mailto:rusty.ma...@sungard.com] Sent: Tuesday, January 06, 2009 6:00 PM To: bob...@attglobal.net Cc: veritas-bu@mailman.eng.auburn.edu; veritas-bu-boun...@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] NBU 6.5.3 Windows Java GUI for the Admin Console Bob, I have to disagree with you, for once. You can only use the java version with that same major revision; it is not backwards compatible to other revision levels (6.5 is not backwards compatible to 6.0 or lower, etc.). The java client is forward and backwards compatible within the same major revision - for the most part, there are things that may not work/populate if the Master is at a higher level than the Java client. I made this suggestion for backwards compatibility at the users forum, but it didn't sound like there would be much work put into it as the new user interface will take care of this. As stated, installing them to separate directories does the trick when multiple java versions/patch levels are needed on the same workstation. If you are not able to change the install directory, edit this key in the registry to contain the path that you want to install to: HKLM\Software\VERITAS\NetBackup - java (NB-Java)\InstallPath Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪ 281-584-4693 Keeping People and Information Connected® ▪ http://availability.sungard.com/ P Think before you print CONFIDENTIALITY: This e-mail (including any attachments) may contain confidential, proprietary and privileged information, and unauthorized disclosure or use is prohibited. If you received this e-mail in error, please notify the sender and delete this e-mail from your system. bob944 bob...@attglobal.net Sent by: veritas-bu-boun...@mailman.eng.auburn.edu 01/01/2009 05:30 PM Please respond to bob...@attglobal.net Toveritas-bu@mailman.eng.auburn.edu cc SubjectRe: [Veritas-bu] NBU 6.5.3 Windows Java GUI for the Admin Console Thanks, but I need to be able to support multiple versions...like 6.0, 6.5, 6.5.2, and 6.5.3 as well as IME, all patch/MP versions are backwards compatible within that major version (5.1, 6.0, 6.5). b so that you could use, say, a 6.5 jdc to admin your 6.5 master and a 6.0 and a 5.x media server. Install the 6.5.3 patch as recommended and tell us. When in doubt, you can always install to separate directories c:\program files\veritas\jav51 c:\program files\veritas\jav60 c:\program files\veritas\jav65 Note: it used to be that you couldn't choose the windows installation option to change the java admin console installation path if a version was already installed to the default (c:\program files\veritas?); you'd have to uninstall that one first. Since then, I install each major version to its own directory as above. Legato 7.x and Backup Express 2.3x and 3.x Can't help you with that. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] SuSE 10sp2
From: nitabillls netbackup-fo...@backupcentral.com Message-ID: 1230982128.m2f.295...@www.backupcentral.com Nothing - it never comes back. I get: Trying 127.0.0.1... Escape character is '^]'. But then it never returns anything. It doesn't refuse connection. Since you used a Web interface and didn't supply any context, I don't know what you were trying to do. I'll guess telnet to the bpcd port. What makes you think nothing happened? *If* your telnet client doesn't issue Connected to messages, that's exactly what I'd expect from a successful connection to bpcd. # telnet 127.0.0.1 13782 Trying 127.0.0.1... Connected to 127.0.0.1. Escape character is '^]'. [ the connection is made and bpcd will wait here for commands until timeout ] [ run a netstat from another window and you'd see: ] # netstat -an|grep 127.0.0.1.13782 127.0.0.1.38058 127.0.0.1.13782 49152 0 49152 0 ESTABLISHED 127.0.0.1.13782 127.0.0.1.38058 49152 0 49152 0 ESTABLISHED # [ bpcd drops the connection on receipt of a carriage return ] Connection to 127.0.0.1 closed by foreign host. # [ run another netstat and see: ] # netstat -an|grep 127.0.0.1.13782 127.0.0.1.13782 127.0.0.1.38058 49152 0 49152 0 TIME_WAIT # [ run another netstat after the time_wait interval and see: ] # netstat -an|grep 127.0.0.1.13782 # You _have_ already followed the procedures in the Veritas NetBackup Troubleshooting Guide | Troubleshooting procedures | Troubleshooting installation and configuration problems and General test and troubleshooting procedures, specifically Resolving network configuration problems, right? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] NBU 6.5.3 Windows Java GUI for the Admin Console
Thanks, but I need to be able to support multiple versions...like 6.0, 6.5, 6.5.2, and 6.5.3 as well as IME, all patch/MP versions are backwards compatible within that major version (5.1, 6.0, 6.5). It's possible, even likely, that each major version understands the previous versions so that you could use, say, a 6.5 jdc to admin your 6.5 master and a 6.0 and a 5.x media server. Install the 6.5.3 patch as recommended and tell us. When in doubt, you can always install to separate directories c:\program files\veritas\jav51 c:\program files\veritas\jav60 c:\program files\veritas\jav65 Note: it used to be that you couldn't choose the windows installation option to change the java admin console installation path if a version was already installed to the default (c:\program files\veritas?); you'd have to uninstall that one first. Since then, I install each major version to its own directory as above. Legato 7.x and Backup Express 2.3x and 3.x Can't help you with that. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] New Library upgrade procedure Scenario 1.
Sun E220R (running Solaris 8) Netbackup Server with attached L1000 tape unit + 4 x DLT drives. Netbackup 5.0MP7. Hardware is beyond old but customer is reluctant to upgrade. We have told them we cannot support the L1000 unit and their backups are at risk. It has gone EOSL and parts are increasingly hard to source. The initial proposal is to replace with an SL24 tape unit and LTO3 drives. Problem :- We are requesting a SCSI Ultra320 HBA to install and attach the new SL24 unit onto. Problem is the card is not compatible with the Solaris 8 (release 06/00) which is installed. We therefore need to upgrade solaris. Disk space is very limited (2 x internal 18gig disks raid 1) and doing an OS upgrade wouldn't leave us with much of a regression path. Proposal 1. One thought is to buy 2 * 36 gig disks. We have a spare E220R we could build a Solaris9 release on as fresh install and it would include the mpt drivers for the card. We could then install Netbackup and attach the SL24 and get everything working in a test environment. Then ship the disks, hba and SL24 down to datacenter and swap them with the current disks in the server. Perform a reconfigure reboot and voil? hopefully all would be working. [...] Perhaps I'm missing something, but essentially you are just replacing the master server and adding another library, right? This should be no different that any other hardware+OS change, and very similar to a DR. So, recover to the existing state on the upgraded/replaced platform, then optionally upgrade NetBackup to the desired version. - Catalog backup(s) - Replace the old box or upgrade Solaris on it - Hardware connections as appropriate - Install the same rev of NetBackup - bprecover - Rediscover devices - Test - [catalog backup/upgrade NetBackup/test as desired] - Production and duplication of retiring drives and media. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] How to plan out policy(schedules), [...]
This is a really well-thought-out answer to his question. Although I don't agree with ALL of your recommendations (I don't like frequency-based schedules for fulls), this is actually a pretty good summary of what someone should do to setup a new backup environment. You've inspired me to blog about the same. (Of course, I may use my own opinions...) ;) Thank you, Curtis. I'm just a simple guy; complex setups make my hair hurt and I've have had enough Oops, forgot about moments that I use simplicity to minimize them. The other approach that I love (though I'd never implement it unless I had beaucoup time to get the coding right end-to-end) is the exact opposite: one person on this list (sorry, I don't rember who you are) who has a bajillion clients with a policy for each one with an automated setup, like a scratch-built backup provisioning system. Took a week for the concept to grow on me, and I can see it for a very experienced shop with a fluid mix of clients. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] How to plan out policy(schedules), [...]
All other advice you receive will be more complicated and that's fine if it makes more sense to you. My philosophy here is to make your NetBackup as simple and self-maintaing as possible; every exception (and there will be some) is a cost. I am very new to NBU. Our enviornment has 300 intell and 100 unix system.We have 1 master ( sun t5120) for entire enviornment and 2 media ( sun x4500 disk storage) for lan based backup. one media server ( sun t5220 ) for SAN client for exchange and RMAN. One more media server ( sun t5220 ) for NDMP backup. We also have stk 8500 tape library. So you need a (emphasis: a) windows policy, a standard, an Exchange, an Oracle and an NDMP. If you'll have long-running backups, set as long a checkpoint interval as you can stand. Give every policy some default priority, say, 1000 (sooner or later, you'll have something that should run as low-priority but if all the policies are 0, there's no lower setting). Set allow multiple data streams. Our plan is to have 45 days retention for all data Personally, I wouldn't save incremenal data for more than two fulls, but since a single retention period is simpler and meets your needs, let's use it. But rather than create a custom retention period let's use one that's already in NetBackup, say 2 months. (Though I love to fiddle with and customize things personally, I prefer to leave everything as stock as possible and still meet the business requirements--every customization adds to the list of things that have to be replicated in the future. If you have a Real Business Need for 45 days, or if the extra retention will cause you to buy more storage, then go ahead and customize a retention level.) and do increamental daiy and full on weekend. Don't do that. Figure out your backup window, say 2000-0600, and set full and incremental frequency-based schedules the same, every day. (Remember that the end time of a window is the last time a policy can automatically _start_, so if a queued backup starting at 0550 and running for a couple of hours is too late, use an earlier time than 0600.) Your weekly fulls will run every seven days (and that day's incremental will not). (There are half a dozen ways to avoid doing a disproportionate number of fulls on a weeknight; the simplest is to just add, say, 10% of your clients to policies every weekday and the rest on Friday; a client/policy's first backup will be a full.) For windows and standard policies selection lists: ALL_LOCAL_DRIVES. Set up (you and/or your clients) excludes for database files, for netbackup/db/images and your disk STUs (yes, include your NetBackup servers in the standard policy for simplicity) and for any other data the client doesn't want/need backed up and make the client responsible for managing it--that's why exclude/include lists are on the client in the first place. For database policies, have all clients use the same name and location for the script. As of now we not planning for cloning ( VAULT ( offsite)). There is alos plan to migrate data if LAN media server ( x4500) fills to 80% to Tape library. NDMP and SAN client backup will go directly to 8500. That sounds as if you're going to have one copy of some backups on disk (only) and one copy of others on tape (only). If the loss of backups due to failed disk or failed tape is acceptable, fine. If that's not acceptable, use Storage Lifecycle Policies to do the duplications; SLPs can easily do duplications integrated into the backup period rather than the big vault batches usually done in off-hours. Two SLPs (one disk-to-tape, the other tape-to-tape) will cover both duplications in your setup. Use the storage unit groups for destinations. stgunit, stggroup, Storage units are almost self-defining. Device discovery for the tape drives and supply a disk path to create basic disk storage units. Reduce the fragment size only if you will be doing significant smounts of individual file restores. Multiplex the tape STUs to something like 8 or even 32 (and control the actual multiplexing used in the schedules) Since you want some backups on tape and others on disk, create one group for all the tape storage units and one for all the disk STUs and set them to load balance. pools etc ? One production pool. Either use DataStore or make _one_ up for your datacenter. With your current plans, you don't need an offsite/duplicate pool or offsiteDB pool, You will need a scratch pool (and set up a barcode rule to put all new tapes into it) and a test pool (do not use your production tapes for testing). Do add a duplicate pool if you ,make duplications as above. Set up the hot catalog backup, full backup, every day. Tell your clients what you will provide them (as you detailed in your message); do not ask them what they would like. They probably know nothing about backup and recovery, what they do know will be wrong, and you will have 300 unique windows policies and 300 unix policies. Just
Re: [Veritas-bu] 6.5.3 on the FTP site
I got the first BUG. YE Ok, it's just the JAVA GUI, but it's something they should see at the first check. Try to open the Activity Monitor, Filter, Advanced Search and then try to do a search for a Policy. This really sucks. You remembered to apply the JAV patch? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] FW: Multi Stream Backups
Both the Information Store Backup and the Mailbox Backup run at the same time. (Not best practice, but they do). The two streams of the Information Store backups starts and run. The four streams of the Mailbox backup start but only two run and two queue. Summary: STU = 1 drive x 8 streams per drive. Multiplexing of each policy's schedule = 4. Allow multiple data steams = yes Master server global setting for max jobs per client = 99. Problem: Although the storage unit allows 8 streams to it and the two policies produces 6 streams between them, only 4 run and 2 queue. Why? Any ideas? You answered your own question without realizing it--see chapter 3 of the Veritas NetBackup System Administrator's Guide, Volume II for how multiplexing works. Each schedule has multiplexing set to 4. The storage unit setting of 8 then becomes immaterial. The lower value (schedule or STU) always rules. Since there are four streams going to the tape (two IS and two MBX, and the latter two streams are in a schedule that allows only four streams to be multiplexed, those streams will wait until a multiplex slot opens up. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Running jobs for hot backups of NBU catalog
Since some days ago I'm researching about, what I think it's, a problem (or a bug?) in a new installation of Symantec NBU 6.5.2 on AIX 5.3. [parent job of catalog backup doesn't end] Have you seen technote http://support.veritas.com/docs/292450 regarding the level of the C++ runtime libraries and AIX patchsets? If you're not at 9.0.0.3 minimum, it might be worth a look. There are other notes in the platform compatibility list to check. (This appears to be AIX week on the list--and for me... I'm ready for that to end.) ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] SAP Scheduled Job Fails with 6 and then never
If it gets a failure (say 96, 6 or 150), the job fails to re-run the same day, but does re-run on the next backup window. Schedule is set to run every 2 hours and I have a window of 2 hours set during the day. If I get no failures, all is well. But one failure, just one seems to blow the schedule away! What does that mean? In some cases, if I re-create a new policy, it seems to be ok, until it fails again. Simon - are your schedules frequency-based? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Error 200
All of our policies are setup the same, i.e. the weekly and monthly backups are calendar based and the daily incrementals are frequency. This has never caused a problem before. Of course, I wouldn't let the fact that Veritas has put this statement Note: A policy can contain more than one schedule. Symantec recommends, however, that calendar-based and frequency-based schedule types are not mixed within the same policy. Under some conditions, schedule types that are combined in one policy can cause unexpected results. in every NetBackup Administrator's Guide since calendar-based schedules were introduced, deter me, either. No-siree, and even if I remembered that somebody with calendar and frequency scheds in the same policy comes up with a problem just like this every year, well, I'd ignore that, too. Pffft. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] What are default exclusion in 6.5.2 ?
Can anybody provide me list what are not backed up by default ( files,folder or filesystem ) for windows and unix clients ? The manual is your friend. Veritas NetBackup System Administrator's Guide, Volume 1, under the heading Files that are excluded from backups by default ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] WIN32 21 on XP
On all I receive WIN32 21 messages Backup from client pco0625: WRN - Removable Storage Management: unable to export database (WIN32 21: The device is not ready. ), the normal fix excluding files in system32 or disabling service is not working or not available. I've tried searching the internet, but no success as well. You didn't find it on support.veritas.com? Your error message is there, and it's also on Google 250 times. As I remember, there are three related articles on the support site (disable RSM, make the microsoft piece that complains about RSM being disabled shut up, and something else Windowy). But perhaps your issue is the normal fix excluding files in system32 or disabling service is not working or not available. Suggest you expand on that to the group. XP? You're running a media server on XP? ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] LTO4 Media Type
We currently have a lot of investment in LTO3 Drives and Tapes and moving forward with LTO4. Is there a way we can se LTO4 Media Type to be different than HCART3 in Netbackup? Yes, call it whatever you want. Either manually or with barcode rules, bring your new LTO4 media in as hcart. Make your LTO4 drives also hcart. If your LTO3 drives and LTO3 media are both hcart3, NetBackup will only put hcart3 tapes in hcart3 drives, hcart into hcart, 8mm into 8mm... The problem is that if you have only 1 scratch pool. Mixing LTO-3 and LTO-4 can cause you some grief. Unless I'm having a memory parity error--always possible--I used to run 5.x (at least, probably earlier), a scratch pool and two drive types with corresponding media. ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Anyone setup the new clustered HDS VTL on
We're unable to see the robot. It's a : Protect tier VTL v2.1.1 Library is presented as P3000 It has LTO2 drives being emulated This is the broken one: /dev/sg/c0tw1000c973d212l0: ## - robot should appear here /dev/sg/c0tw1000c973d212l1: removable dev type 1h IBM ULTRIUM-TD2 5AT0 /dev/sg/c0tw1000c973d212l2: removable dev type 1h IBM ULTRIUM-TD2 5AT0 Should look like this: /dev/sg/c0tw1000c96c5060l0: removable dev type 8h ATL P3000 0100 /dev/sg/c0tw1000c96c5060l1: removable dev type 1h QUANTUM DLT7000 0100 Any ideas, Is your VTL in the hardware compatibility list? Maybe I misunderstand what VTL you have, but the only Hitachi I see in the list is VF100 Series. The list also shows the required Inquiry String--it shouldn't identify itself with the string of the library it's going to emulate. ALACRIT^VTLA for the VF100, for instance. Early October HCL: ftp://exftpp.symantec.com/pub/support/products/NetBackup_Enterprise_Serv er/284599.pdf ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Vmoprcmd in version 6.5
In the past we have used vmoprcmd -h vahfcnas1-old -d pr | awk
'$1~/[0-9]/ {print $3}' to find the pending requests for a single
media server. We have seen that the fields have been changed and that
all the pending requests for all media servers are returned.
Is there a
replacement for this command.
Our issue is that we will have to parse the output of each line of
pending requests to determine which remote site is to be notified.
You may want to prepare a Plan B; the 6.0 Commands manual says
The following usage is provided for backward compatibility only:
/usr/openv/volmgr/bin/vmoprcmd [-h volume_database_host] {-d
[pr | ds | ad] | -dps [drive_name]}
and the 6.5 manual says
-d [pr | ds | ad]
This command is supported for pre-NetBackup 6.0 systems.
There's also this from the 6.5 Release Notes (similar language in the
6.0 Release Notes and Release Impact Bulletin:
Device monitor interface
│ Default display information for the vmoprcmd command line has been
changed to include the status of all Media Servers, and all drives'
paths on
all media servers and all pending requests on all media servers. The -d
option provides the pre-NetBackup 6.0 display format and functionality.
│ Additional options have been added to the vmoprcmd command line to
manage and display the drive's status per-path. Refer to the NetBackup
Command document for more information about these commands.
│ The Device Monitor in the NetBackup Administration Console provides
additional host/path specific information for drives, only when the
drive is
selected. The default display is an overview of all drives' status.
___
Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
Re: [Veritas-bu] Reduce Volume Pools
First, if this is one of the class of I do not appear to have a reply to any emails I have sent for the past 2 weeks :-( emails, then, FYI, I haven't seen any of them so perhaps you really have had an outbound mail problem. Quick question; I am attempting to reduce the amount of volume pools that are in use for each policy we have. I want to group them together, but one thing I noticed is that they have different retentions. I used to think that was a big no no, but could someone advise if this is an issue in 5.1 MP5. I know the default is turned off. The aim of the excercise is to try to use the smallest number of pools possible, and maximize the amount of data into the tapes, so they get full. I kind of would like to fill the tapes as much as poss It's not at all clear to me what you are proposing. If it's using allow_multiple_retentions_per_media, then, yes, you would no longer have, say, a separate tape from pool X with 2-week retention daily incremental data and another pool X with three-month retention weekly fulls and another with one-year-retention monthly fulls--they could all be on the same tape if that tape were available when those backups ran. But unless you have a toy robot that only holds five tapes, this is usually a really bad idea. Why is left as an exercise for the reader. The minimum number of pools to use is one, and that, IMO, should be your (impractical) goal. Use NetBackup, or name it after your datacenter or your cat, but only use additional pools if there's a good technical reason to. A Unix pool and a Windows pool is not a good technical reason. onsite and offsite to ensure, for example, that an inline-tape-copy will produce separated data so that one tape can be sent away, is. Don't forget that multiple media servers don't write to each others' tapes (except in 6.5, if you choose the option to allow it). ___ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
