f. OpenVMS FAQ -&- page 11)b @5z

HP OpenVMS Systems Documentation

 q> $"b,
Content starts here"D

The OpenVMS Frequently Asked Questions (FAQ)


 l n  
PreviousContentsIndex

l

5.42 Please help me with the OpenVMS BACKUP utility?

N

5.42.1 Why isn't BACKUP/SINCE=BACKUP working?



FIf you are seeing more files backed up than previously, you are seeingDthe result of a change that was made to ensure BACKUP can perform anFincrementation restoration of the files. In particular, if a directoryGfile modification date changes, all files underneath it are included in?the BACKUP, in order to permit incremental restoration should adirectory file get renamed.`

5.42.1.1 Why has OpenVMS gone through the agony of this change?



CWhen a directory is renamed, the modified date is changed. When theDrestoration needs to restore the directory and its contents, and theGrestoration should not result in the restoration of the older directory?name when a series of incremental BACKUPs are restored. Thus anAincremental BACKUP operation needs to pick up all of the changes.

GConsider performing an incremental restoration, to test the procedures.DThis testing was how OpenVMS Engineering found out about the problem>that was latent with the old BACKUP selection scheme---the oldGincremental BACKUP scheme would have missed restoring any files under a?renamed directory. Hence the change to the selection mechanismsTmentioned in Section 5.42.1.T

5.42.1.2 Can you get the old BACKUP behaviour back?



@Yes, please see the /NOINCREMENTAL qualifier available on recentEOpenVMS versions (and ECO kits). Use of this qualifier informs BACKUPAthat you are aware of the limitations of the old BACKUP behaviour%around incremental disk restorations.T

5.42.2 What can I do to improve BACKUP performance?



DUse the documented commands in the manual for performing incrementalDBACKUPs. Use the documented incremental procedures. Don't try to use2incremental commands in a non-incremental context.

FAlso consider understanding and then using /NOALIAS, which will likelyFbe a bigger win than will anything to do with the incremental BACKUPs,Hparticularly on system disks and any other disks with directory aliases.

5See the OpenVMS documentation for additional details.

DIgnoring hardware performance and process quotas, the performance ofCBACKUP during a disk saveset creation is typically limited by threefactors:

    @
  1. Default extend size
    The default behavior can have poorG performance, as the extend operation can involve extensive additionalE processing and I/O operations. Consider changing the default extend< value on the volume, or change the extend for the process:

     

    "
    $ set rms/extend=65000
    
    
    ?

  2. Output IO size
    The default IO size for writing an RMSG sequential file is 32 blocks, an increase from the value of 16 blocksI used on earlier versions. Setting this to the maximum of 127 can reduce, the number of IOs by almost a factor of 4:

     

    "
    $ set rms/block=127
    
    
    H
    Note that the performance might be better on some controllers if1 the block count is a multiple of 4 - e.g. 124G

  3. Synchronous writes to the saveset
    Starting with OpenVMS V7.3,G you can now persuade RMS to turn on write-behind for sequential filesD opened unshared. (Please see the V7.3 release notes or more recentB documentation for details.) Enabling the write-behind operationsE involves setting the dynamic system parameter RMS_SEQFILE_WBH to 1.2 This parameter is dynamic, and it can be enabledFand disabled without a reboot, and changes in its setting can and willDdirectly effect the running system. In order to get the full benefitEfrom write-behind operations, you also need to increase the RMS localBbuffer count from the default of 2 to a larger number. Raising theCvalue to 10 is probably a reasonable first estimate for this value.

     

    "
    $ run sys$system:sysmanPARAMETERS USE ACTIVE PARAMETERS SET RMS_SEQFILE_WBH 1PARAMETERS WRITE ACTIVEEXIT*$ SET RMS/BUFFER=10/EXTEND=65000/BLOCK=1278$ BACKUP source-specification ddcu:[dir]saveset.bck/SAVE
    
    
    

K

5.42.3 Why is BACKUP not working as expected?



CFirst, please take the time to review the BACKUP documentation, andBparticularly the BACKUP command examples. Then please download andGinstall the most current BACKUP eco kit. Finally, please please set theCprocess quotas per the System Management documentation. These steps#tend to resolve most problems seen.

CBACKUP has a very complex interface, and there are numerous commandGexamples and extensive user documentation available. For a simpler user:interface for BACKUP, please see the documentation for theBACKUP$MANAGER tool.

2As for recent BACKUP changes, oddities, bugs, etc:

    H
  • A change made in OpenVMS V6.2 WILL cause more files to be includedC into a file-based BACKUP saveset using /SINCE=BACKUP as all filesE underneath any directory with a sufficiently recent (selected) dateA will be included in the saveset. This change was deliberate andC intentional, and was mandated by the need to provide a functional= incremental restoration.
    Without the inclusion of theseD apparently-extra files, an incremental saveset can NOT be reliably restored.F
  • As part of the OpenVMS V6.2 change, the /SINCE command---withoutH the specification of the =BACKUP keyword---selected more files than itG should have. This is a bug. This bug has been remedied in the OpenVMSB BACKUP source code and in some of (all of?) the BACKUP ECO kits.


+When working with BACKUP, you will want to:

    H
  • Ensure you have your process quotas set per the recommendations inC the OpenVMS System Management documentation. Deviation from theseA values can and will lead to access violation (ACCVIO) and other untoward behaviour.I
  • Get the current BACKUP ECO kit and install it BEFORE you attempt to troubleshoot any problems.H
  • Learn about the /NOINCREMENTAL (new) and /NOALIAS (V6.2 and later)B command qualifiers. The former qualifier returns to the pre-V6.2D behaviour of the /SINCE file selection mechanism, while the latterD (specified with /IMAGE) reduces the replication of files on systemI disks and other disks with file alias and directory alias entries. BothI of these can reduce the numbers of files that will be selected and thusI included into the saveset.
    Learn what /IGNORE=INTERLOCK means. This command probably does notGprovide what you think it does---those file system interlocks that thisFcommand is ignoring were implemented for a reason, after all. IgnoringGthese interlocks can lead to missed data and potentially to corruptionsFto individual files stored within the output saveset, corruptions that>may or may not be reported. For details on this BACKUP command6qualifier, please see the Ask The Wizard topic (2467).


*When working with the BACKUP callable API:

    I
  • Build your applications with the most current BACKUP API available.I Changes made to the V7.1-2 and V7.2 API were incompatible with the V7.1H and V7.2-1 and later APIs, and this incompatibility was repaired via aH BACKUP ECO kit. Do NOT build your application with the versions of theI BACKUP API that shipped with V7.1-2 and V7.2, as these are incompatibleA with the BACKUP API constants that were used on other versions.
K

5.42.4 How do I fix a corrupt BACKUP saveset?



CBACKUP savesets can be corrupted by FTP file transfers and by toolsFsuch as zip (particularly when the zip tool has not been asked to saveGand restore OpenVMS file attributes or when it does not support OpenVMS<file attributes), as well as via other means of corruptions.

EIf you have problems with the BACKUP savesets after unzipping them orAafter an FTP file transfer, you can try restoring the appropriate"saveset attributes using the tool:

 

"
+$ @RESET_BACKUP_SAVESET_FILE_ATTRIBUTES.COM




AThis tool is available on the OpenVMS Freeware (in the [000TOOLS]@directory). The Freeware is available at various sites---see theCFreeware location listings elsewhere in the FAQ---and other similar.tools are also available from various sources.

DIn various cases, a SET FILE/ATTRIBUTES command can also be used. AsBthe parameters of this command must be varied as the target BACKUP:saveset attributes vary, this approach is not recommended.

:Also see the "SITE VMS", /FDL, and various other@file-attributes options available in various FTP tools. (Not all9available FTP tools support any or all of these options.)

GBrowser downloads (via FTP) and incorrect (binary or ascii FTP transferFmodes) are notorious for causing RMS file corruptions and particularly@BACKUP saveset corruptions. You can sometimes help encourage theCbrowser to select the correct FTP transfer type code (via RFC1738):

    G
  • ftp://host/urlname.ext;type=i ! request ftp image/binary transferE
  • ftp://host/urlname.ext;type=a ! request ftp ascii/text transfer


EYou can also often configure the particular web browser to choose theBappropriate transfer mode by default, based on the particular fileFextensions, using a customization menu available in most web browsers.EYou can select that the specific file extentions involved use the FTPGbinary transfer mode, which will reduce the number of corruptions seen.V

5.42.5 How do I write a BACKUP saveset to a remote tape?



@How to do this correctly was described at DECUS long ago. On theCOpenVMS host with the tape drive, create the following SAVE-SET.FDLfile:

 

"
RECORD%        FORMAT                  fixed$        SIZE                    8192




Then create BACKUP_SERVER.COM:

 

"
$ !?$ ! BACKUP_SERVER.COM - provide remote tape service for BACKUP.$ !
$ set noon$ set rms/network=16$ allocate mka500 tapedev2$ mount/nounload/over:id/block=8192/assist tapedev0$ convert/fdl=SAVE-SET sys$net tapedev:save-set.$ dismount/unload tapedev$ stop/id=0




AOn the node where you want to do the backup, use the DCL command:

 

"
$ backup -    srcfilespec -8    node"user pwd"::"task=backup_server"/block=8192/save




FOne area which does not function here is the volume switch; multi-reelDor multi-cartridge savesets. Since the tape is being written throughEDECnet and RMS and the magtape ACP, BACKUP won't see the media switchGand will split an XOR group across the reel boundary. BACKUP might well?be willing to read such a multi-reel or multi-cartridge savesetF(directly, not over the net) as the XOR blocks are effectively ignoredDuntil and unless needed for error recovery operations. BACKUP likely@will not be able to perform an XOR-based recovery across reel orcartridge boundaries.

FUnfortunately BACKUP can't read tapes over the network because the RMSCfile attributes on a network task access look wrong; the attributes)reported include variable length records.S

5.42.6 How to perform a DoD security disk erasure?



ASometimes refered to as disk, tape, or media declassification, as>formatting, as pattern erasure, or occasionally by the genericFreference of data remanence. Various references to the US Deparment of/Defence (DoD) or NCSC "Rainbow Books". documentation are also seen in this context.

@While this erasure task might initially appear quite easy, basicEcharacteristics of the storage media and of the device error recoveryFand bad block handling can make this effort far more difficult than itmight initially appear.

@Obviously, data security and sensitivity, the costs of exposure,>applicable legal or administrative requirements (DoD, HIPPA or@otherwise), and the intrinsic value of the data involved are all=central factors in this discussion and in the decision of the?appropriate resolution, as is the value of the storage hardware involved.

EWith data of greater value or with data exposure (sometimes far) more@costly than the residual value of the disk storage involved, theDphysical destruction of the platters may well be the most expedient,Feconomical, and appropriate approach. The unintended exposure of a badGblock containing customer healthcare data or of credit card numbers canEquite be costly, of course, both in terms of the direct loss, and the1longer-term and indirect costs of such exposures.

@Other potential options include the Freeware RZDISK package, theEOpenVMS INITIALIZE/ERASE command (and potentially in conjunction withthe $erapat system service)@and OpenVMS Ask The Wizard (ATW) topics including (841), (3926),E(4286), (4598), and (7320). For additional information on sys$erapat, ?see the OpenVMS Programming Concepts manual and the OpenVMS VAXhFexamples module SYS$EXAMPLES:DOD_ERAPAT.MAR. Some disk controllers andGeven a few disks contain support for data erasure. Some DSSI Disk ISEs,a for instance.e

s;For the prevention of casual disk data exposures, a generice>INITIALIZE/ERASE operation is probably sufficient. This is notGcompletely reliable, particularly if the data is valuable, or if legal,>Dadministrative or contractual restrictions are stringent---there mayDwell be revectored blocks that are not overwritten or not completelyEoverwritten by this erasure, as discussed above, and these blocks cannGobviously contain at least part of most any data that was stored on theiGdisk -- but this basic disk overwrite operation is likely sufficient tot,prevent the typical information disclosures.

1>You will want to consult with your site security officer, yourEcorporate security or legal office, with HP Services or your preferediCservice organization, or with a firm that specializes in erasure ortCdata declassification tasks. HP Services does traditionally offer a-%secure disk declassification service.eQ

5.42.7 How to enable telnet virtual terminals?

t

oATo enable virtual terminal support for telnet and rlogin devices,e?add the following logical name definitions into SYLOGICALS.COM:<

> 
a
"
/$ DEFINE/SYSTEM/EXECUTIVE TCPIP$RLOGIN_VTA TRUE3/$ DEFINE/SYSTEM/EXECUTIVE TCPIP$TELNET_VTA TRUEh

U




fGIn SYSTARTUP_VMS.COM, ensure that a command similar to the following is invoked:

e 
i
"
N$ SYSMAN IO CONNECT VTA0/NOADAPTER/DRIVER=SYS$LOADABLE_IMAGES:SYS$TTDRIVER.EXE




tEIn MODPARAMS.DAT, add the following line or (if already present) masknBthe specified hexidecimal value into an existing TTY_DEFCHAR2, and5perform a subsequent AUTOGEN with an eventual reboot:e

t 
o
"
TTY_DEFCHAR2 = %X20000

s


lThis value is TT2$M_DISCONNECT.s

KDOn older TCP/IP Services---versions prior to V5.0---you will have to"perform the following UCX command:

s 
g
"
$ UCXo$UCX> SET CONF COMM/REMOTE=VIRTUAL

P
N

5.42.7.1 Volume Shadowing MiniCopy vs MiniMerge?



o@MiniMerge support has been available for many years with OpenVMSEhost-based volume shadowing, so long as you had MSCP controllers (eg:CFHSC, HSJ, or HSD) which supported the Volume Shadowing Assist known as""Write History Logging".

eDIf you are interested in mini-merge and similar technologies, pleaseBsee the Fibre Channel webpage and the information available there:

i

-EMini-Merge support was originally intended to be controller-based andg&was expected with HSG80 series storageAcontrollers and was expected to require ACS 8.7 and OpenVMS AlpharV7.3-1..

Host-based Mini-Mergee (HBMM)pGis now available for specific OpenVMS releases via a shadowing ECO kit,tBand is also present in OpenVMS V8.2 and later. HBMM applies to theCHSG80 series and---like host-based volume shadowing---to most other '(all other?) supported storage devices.i

eFThe following sections describe both Mini-Copy and Mini-Merge, and can provide a basis for discussions.6

5.42.7.1.1 Mini-Copy?



b?A Shadowing Full Copy occurs when you add a disk to an existing Dshadowset using a MOUNT command; the entire contents of the disk areBeffectively copied to the new member (using an algorithm that goesGthrough in 127-block increments and reads one member, compares with theeCtarget disk, and if the data differs, writes the data to the targetuFdisk and loops back to the read step, until the data is equal for thatC127-block section). (This is one of the reasons why the traditional Arecommendation for adding new volumes to a shadowset was to use arDBACKUP/PHYSICAL copy of an existing shadowset volume, simply becauseEthe reads then usually matched and thus shadowing usually avoided theeneed for the writes.)n

t@If you warn OpenVMS ahead of time (at dismount time) that you'reGplanning to remove a disk from a shadowset but re-add it later, OpenVMSaEwill keep a bitmap tracking what areas of the disk have been modified Ewhile the disk was out of the shadowset, and when you re-add it latert@with a MOUNT command OpenVMS only has to update the areas of theEreturned disk that the bit-map indicates are now out-of-date. OpenVMSmDdoes this with a read source / write target algorithm, which is muchEfaster than the shenanigans the Full Copy does, so even if all of the 9disk has changed, a Mini-Copy is faster than a Full Copy.o7

5.42.7.1.2 Mini-Merge?

.

gBA Shadowing Merge is initiated when an OpenVMS node in the cluster;(which had a shadowset mounted) crashes or otherwise leavesnDunexpectedly, without dismounting the shadowset first. In this case,?OpenVMS must ensure that the data is identical, since ShadowingIGguarantees that the data on the disks in a shadowset will be identical.0DIn a regular Merge operation, Shadowing uses an algorithm similar to@the Full Copy algorithm (except that it can choose either of theGmembers' contents as the source data, since both are considered equally>Fvalid), and scans the entire disk. Also, to make things worse, for anyDread operations in the area ahead of what has been merged, ShadowingGwill first merge the area containing the read data, then allow the read to occur.r

uDA Merge can be very time-consuming and very I/O intensive. If a nodeGcrashes, the surviving nodes can query to determine what exact areas ofaDthe disk the departed node was writing to just before the crash, andFthus Shadowing only needs to merge just those few areas, so this tendsDto take seconds, as opposed to potentially requiring many minutes or$even hours for a regular full Merge.l

5.43 Please explain DELETE/ERASE and File Locks?



DDELETE/ERASE holds the file lock and also holds a lock on the parentEdirectory for the duration of the erasure. This locking can obviouslyDcause an access conflict on either the file or on the directory---itBmight well pay to rename files into a temporary directory locationDbefore issuing the DELETE/ERASE, particularly for large files and/orBfor systems with multiple overwrite erase patterns in use; for anyCsystems where the DELETE/ERASE erasure operation will take a while.oZ

5.44 Managing File Versions?



iDSome applications will automatically roll file version numbers over,@and some will require manual intervention. Some will continue toCoperate without the ability to update the version, and some will bewFunable to continue. Some sites will specifically (attempt to) create aCfile with a version of ;32767 to prevent the creation of additionalm files, too.u

tETo monitor and resolve file versions, you can use commands including: 

n 
s
"
#$ SET FILE/VERSION_LIMIT=n filenamer+$ SET DIRECTORY/VERSION_LIMIT=n [directory]c




nFAnd you can also monitor file version numbers, and can report problemsDwith ever-increasing file versions to the organization(s) supporting@the application(s) generating files with ever-increasing versionFnumbers for details on potential problems, and for any recommendationsGon resetting the version numbers for the particular product or package.mIf required, of course.e

fGThe following pair of DCL commands---though obviously subject to timingaEwindows--- can be used to rename all the versions of a file back down 3to a contiguous sequence of versions starting at 1:e

 
h
"
!$ RENAME file.typ;*   RENAME.TMP;l$ RENAME RENAME.TMP;* file.typ;a

t


AFThe key to the success of this RENAME sequence is the specification ofD(only) the trailing semicolon on the second parameter of each of theRENAME commands.

aCYou may also see the numbers of files reduced with DELETE commands, Gwith multiple directories, or with PURGE commands such as the followingm examples: 

e 
/
"
$ PURGE/BEFORE="-2-"$ PURGE/BEFORE="TODAY-2-"d$ PURGE/KEEP=10"

s


tYou can use DFU (Freeware)D to quickly and efficiently scan for all files with large(r) version numbers:a

e 
s
"
DFU SEARCH/VERSION=MINIMUM=nnnnA

e


oEIf you are creating or supporting an application, selecting temporary?or log file filenames from among a set of filenames---selectingt@filenames based on time, on process id, on the day of week, week!number, or month, on the f$unique_D lexical (V7.3-2 and later), etc---is often useful, as this approachE more easily permits on-line adjustments to the highest file versionslF and easily permits on-line version compression using techniques shownB above. With differing filenames, you are less likely to encounterD errors resulting from files that are currently locked. You can alsoF detect the impending version number limit within the application, andG can clean up older versions and roll the next file version creation to ;1 or such.

uIAlso see Section 9.4.lh

5.45 Host-based Volume Shadowing and RAID?



oEHost-based Volume Shadowing (HBVS) is Disk Mirroring is RAID Level 1.n

s@HBVS is capable of shadowing devices of different geometries, ofFdifferent block counts (with dissimilar device shadowing; allowing forFmixtures of hardware) and---with dynamic volume expansion---of growingEvolumes on the fly, and HBVS is capable of shadowing/mirroring/raid-1/Eoperations across cluster configurations up to the full span---pleasepFsee the Cluster SPD for the current supported span; the supported spanGlimit is currently multiple hundreds of kilometers---of a cluster. HBVSx8can be layered onto controller (hardware) RAID, as well.

rCFor information on host-based striping, please see the StorageWorks RAID product.a 

u


 i il n  r
PreviousNextContentsIndex

 

#6b
$ !?$ ! BACKUP_SERVER.COM - provide remote tape service for BACKUP.$ !
$ set noon$ set rms/network=16$ allocate mka500 tapedev2$ mount/nounload/over:id/block=8192/assist tapedev0$ convert/fdl=SAVE-SET sys$net tapedev:save-set.$ dismount/unload tapedev$ stop/id=0




AOn the node where you want to do the backup, use the DCL command:ref="vmsfaq_008.html#bottom_008">PreviousY Nextt\ Contents^ Index u

 

p#l6fwand ECO levels are currentVIS DISK$WEB3 .2CB(y筗;; E!!OGO-SMALL.GIF;BEAVIS DISK$WEB3 .2CB(y筗<< <$.GIF;BEAVIS DISK$WEB3 .2CB(y筗== D ORACLE-Q.GIF;BEAVIS DISK$WEB3 .2CB(y筗>> W3 PRIM_CATEG_HEADER-FACTSHEET.GIF;BEAVIS DISK$WEB3 .2CB(y筗?? G#2OVMARCHIVES.GIF;BEAVIS DISK$WEB3 .2CB(y筗@@ C5PARTNER.GIF;BEAVIS DISK$WEB3 .2CB(y筗AA E!5QUESTIONS.GIF;BEAVIS DISK$WEB3 .2CB(y筗BB B2QSPECS.GIF;BEAVIS DISK$WEB3 .2CB(y筗CC ?2RAC.GIF;BEAVIS DISK$WEB3 .2CB(y筗DD D 2SERVCERT.GIF;BEAVIS DISK$WEB3 .2CB(y筗EE D 3OD7-2001.GIF;BEAVIS DISK$WEB3 .2CB(y筗FF C2UPGRADE.GIF;BEAVIS DISK$WEB3 .2CB(y筗GG E! RED_ARROW.GIF;BEAVIS DISK$WEB3 .2CB(y筗HH N*"VERSE_HEADER-8I-9I.GIF;BEAVIS DISK$WEB3 .2CB(y筗II ?/RDB.GIF;BEAVIS DISK$WEB3 .2CB(y筗JJ ^: SECOND_CATEG_HEADER-COMMUNICATIONS.GIF;BEAVIS DISK$WEB3 .2CB(y筗KK B6NTACTS.GIF;BEAVIS DISK$WEB3 .2CB(y筗A/KK L  vR.HTDOCS.SOLUTIONS.ORACLE.IMAGES'SECOND_CATEG_HEADER-EVENTS.GIF;BEAVIS DISK$WEB3 .2CB(y筗 A4LINKS.GIF;BEAVIS DISK$WEB3 .2CB(y筗 C4PATCHES.GIF;BEAVIS DISK$WEB3 .2CB(y筗 C5RODINFO.GIF;BEAVIS DISK$WEB3 .2CB(y筗 >4QA.GIF;BEAVIS DISK$WEB3 .2CB(y筗 H$4RELATIONSHIP.GIF;BEAVIS DISK$WEB3 .2CB(y筗 C5OADMAPS.GIF;BEAVIS DISK$WEB3 .2CB(y筗 D 4SERVCERT.GIF;BEAVIS DISK$WEB3 .2CB(y筗  D 4TRAINING.GIF;BEAVIS DISK$WEB3 .2CB(y筗  A TABLE.GIF;BEAVIS DISK$WEB3 .2CB(y筗  C UPDATED.GIF;BEAVIS DISK$WEB3 .2CB(y筗  N* VERTICAL_DIVISION3.GIF;BEAVIS DISK$WEB3 .2CB(y筗  F" Y