HP AutoSYNC User’s Guide Abstract HP NonStop™ AutoSYNC™ software is a reliable, fault-tolerant, and automatic subsystem for replicating files through synchronization. This manual describes the uses, installation, and all other operational aspects of AutoSYNC. Product Version AutoSYNC AAR Supported Release Version Updates (RVUs) This manual supports all J-series RVUs, all H-series RVUs, and all G-series RVUs, until otherwise indicated in a replacement publication.
Document History Part Number Product Version Published 522580-012 NonStop AutoSYNC AAM (Update 12) March 2009 522580-013 NonStop AutoSYNC AAN (Update 13) January 2010 522580-014 NonStop AutoSYNC AAO (TCF) August 2010 522580-015 NonStop AutoSYNC AAQ (Update 14) January 2011 522580-016 NonStop AutoSYNC AAR (Update 15) June 2011
Legal Notices © Copyright 2010 Hewlett-Packard Development Company L.P. Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor's standard commercial license. The information contained herein is subject to change without notice.
HP AutoSYNC User’s Guide Index Tables . Legal Notices 1.
Contents 3. One-time Synchronization Authorize AutoSYNC 2-7 Privileged ASYSYNC/ASYSYNCP 2-8 Update the System Coldload and Shutdown Procedures Install the AutoSYNC EMS templates 2-9 Shutting Down AutoSYNC 2-9 Installing a New Version of AutoSYNC 2-9 Running Different Versions of AutoSYNC 2-10 2-8 3.
Contents 4.
Contents 5. OSS File Synchronization DELETE SYNCFILESET (and DELETE BATCHID) 4-6 SUSPEND SYNCFILESET (and SUSPEND BATCHID) 4-6 ACTIVATE SYNCFILESET (and ACTIVATE BATCHID) 4-6 Checking Status of Synchronizations 4-6 STATUS AUTOSYNC 4-6 STATUS SYNC 4-6 Security and Authorization 4-7 Bypassing Disabled Log Down 4-7 Scheduled Synchronization Flow 4-8 Monitor Process 4-8 ASYSYNCP process 4-9 ASYSYNC process 4-9 Variant for Synchronization over TCP/IP 4-10 5.
Contents 7. Synchronization Over TCP/IP TRIGGER examples 6-4 TRIGGERONFILESET Example 6-6 Note on Triggers that invoke SQLCI 6-6 7. Synchronization Over TCP/IP Overview 7-1 Configuring a TCP/IP connection 7-1 On both Source and Destination Systems 7-2 Set the TCP/IP port numbers for AutoSYNC On the Destination System 7-3 On the Source System 7-4 Synchronization over TCP/IP 7-5 Connection Setup 7-5 Note on ASYSYNCP Execute Access 7-5 Logging 7-6 7-2 8.
Contents 8.
Contents 9. Reporting Function RUN[D] 8-85 START MONITOR 8-86 STATS 8-87 STATUS AUTOSYNC 8-87 STATUS BATCHID 8-88 STATUS MONITOR 8-89 STATUS SYNC 8-89 STATUS SYNCFILESET 8-90 STOP MONITOR 8-92 ABEND MONITOR 8-92 SUSPEND BATCHID 8-92 SUSPEND SYNCFILESET 8-93 SUSPEND SYNCUSER 8-93 SYNCHRONIZE 8-93 TIME 8-110 UPDATE 8-111 UPGRADE MAPDB 8-114 VOLUME 8-114 9. Reporting Function Overview 9-1 REPORT Command 9-1 Report Output Format 9-3 A.
Contents B.
Contents C.
Contents Tables HP AutoSYNC User’s Guide—522580-016 x
What’s New in This Manual Manual Information HP AutoSYNC User’s Guide Abstract HP NonStop™ AutoSYNC™ software is a reliable, fault-tolerant, and automatic subsystem for replicating files through synchronization. This manual describes the uses, installation, and all other operational aspects of AutoSYNC.
What’s New in This Manual New and Changed Information on page 8-7, ALTER SYNCFILESET on page 8-35 and SYNCHRONIZE on page 8-93 for details. • • TRIGGERONFILESET no longer executes if no files have been replaced or purged as a result of the synchronization. See Overview paragraph in Section 6, Triggers and TRIGGERONFILESET command option description for ADD SYNCFILESET on page 8-7, ALTER SYNCFILESET on page 8-35 and SYNCHRONIZE on page 8-93.
About This Manual This manual describes the uses, installation, and all other operational aspects of HP AutoSYNC. AutoSYNC is a reliable, fault-tolerant, and automatic subsystem for replicating files through synchronization. File synchronization is a complementary alternative to data replication for disaster recovery. Data replication is used primarily for database files; it replicates each record update that occurs in a database transaction.
About This Manual General Syntax Notation each side of the list, or horizontally, enclosed in a pair of brackets and separated by vertical lines. For example: FC [ num ] [ -num] [ text] K [ X | D ] address-1 { } Braces. A group of items enclosed in braces is a list from which you are required to choose one item. The items in the list may be arranged either vertically, with aligned braces on each side of the list, or horizontally, enclosed in a pair of braces and separated by vertical lines.
About This Manual Notation for Messages Line Spacing. If the syntax of a command is too long to fit on a single line, each continuation line is indented three spaces and is separated from the preceding line by a blank line. This spacing distinguishes items in a continuation line from items in a vertical list of selections. For example: ALTER [ / OUT file-spec / ] LINE [ , attribute-spec ]... !i and !o.
About This Manual Notation for Messages Nonitalic text. Nonitalic letters, numbers, and punctuation indicate text that is displayed or returned exactly as shown. For example: Backup Up. lowercase italic letters. Lowercase italic letters indicate variable items whose values are displayed or returned. For example: p-register process-name [ ] Brackets. Brackets enclose items that are sometimes, but not always, displayed.
About This Manual Notation for Management Programming Interfaces Notation for Management Programming Interfaces The following list summarizes the notation conventions used in the boxed descriptions of programmatic commands, event messages, and error lists in this manual. UPPERCASE LETTERS. Uppercase letters indicate names from definition files; enter these names exactly as shown. For example: ZCOM-TKN-SUBJ-SERV lowercase letters.
About This Manual Change Bar Notation HP AutoSYNC User’s Guide—522580-016 xviii
1 AutoSYNC Overview Introduction HP NonStop AutoSYNC software (AutoSYNC) provides automatic synchronization for HP NonStop server disk files. File synchronization is a complementary alternative to data replication for disaster recovery. Data replication is used primarily for database files; it replicates each record update that occurs in a database transaction.
AutoSYNC Overview Product Overview Product Overview This section gives a brief overview of AutoSYNC; a more complete description can be found in later sections. File Sets AutoSYNC synchronizes Enscribe files and SQL tables between pairs of source and destination locations. These locations are described by file set patterns, such as $DATA.APP*.*OBJ.
AutoSYNC Overview Fault Tolerance and File Recovery To configure a scheduled synchronization, a user must be added to the AutoSYNC system. An AutoSYNC administrator enables users to schedule synchronizations and can specify a maximum priority and select CPUs that the user should use. The AutoSYNC administrator can also suspend and activate a user without changing the user’s configuration. Fault Tolerance and File Recovery Scheduled synchronizations are managed by a fault-tolerant process pair.
AutoSYNC Overview Reporting Reporting It is often difficult to measure how dynamic your processing environment really is. The reporting function allows you to evaluate the amount of change that occurs on the file sets you wish to synchronize, without performing the actual data movement. Reporting is useful for planning system migrations and for configuring the frequency of scheduled synchronizations, either for disaster recovery or for data distribution purposes.
AutoSYNC Overview System and Map Databases System and Map Databases Each system that has AutoSYNC installed has a single system database (SysDB) either in the subvolume $SYSTEM.ESCSYSDB or in the ESCSYSDB subvolume on some other disk volume. This database is a central reference point for other AutoSYNC components. You should only update SysDB through the AutoSYNC command interpreter, which ensures consistency between it and the other AutoSYNC components.
AutoSYNC Overview AutoSYNC in a Network of Systems HP AutoSYNC User’s Guide—522580-016 1-6
2 Installation This section discusses the procedure to follow for the initial installation of AutoSYNC. The topics covered are organized as follows: Prerequisites Installing AutoSYNC Installing a New Version of AutoSYNC Prerequisites The AutoSYNC (SysDB and MapDB) configuration is stored in a set of audited SQL tables. The HP NonStop TMF product must be installed and configured. Volumes containing the AutoSYNC configuration must be specified in the TMF configuration as audited data volumes.
Installation Move Files from the Product Media to the Installation Subvolume Also, if your system security is configured to prevent a user from performing a log down to another user ID without specifying a password, read the section “Bypassing Disabled Log Down” on page 4-7. Consult the softdoc file for any recent changes to the installation procedures. The entire process should take only a few minutes.
Installation Install the AutoSYNC Product Files Install the AutoSYNC Product Files Use the INSTALL macro to restore the product files from ZAUTOSYC to the operational subvolume; the default operational subvolume is $SYSTEM.AUTOSYNC. Do not install the files in $SYSTEM.SYSTEM or $SYSTEM.SYSnn. Then execute the INSTALL macro as follows:. VOLUME $.ZAUTOSYC RUN INSTALL You may specify a subvolume in which to install the AutoSYNC operational files.
Installation Create the AutoSYNC System Database templates and filter. To do this, enter the following commands in a separate window terminal window: ADD DEFINE =_EMS_TEMPLATES, CLASS MAP, FILE ZASYTMPL EMSDIST TYPE P, COLLECTOR $0, TEXTOUT $HOME, FILTER ASYFLTR Create the AutoSYNC System Database Each system running AutoSYNC maintains system-level parameters stored in a small set of tables to control the overall operation of the product. This set of tables is referred to as the System Database (SysDB).
Installation License AutoSYNC volume as the SysDB tables). You may also specify a particular catalog. See CREATE SYSDB on page 8-64 for details on the command. AutoSYNC 1? create sysdb on $data; --- SQL Catalog $DATA.ESCCATLG is being created. --- SQL Catalog $DATA.ESCCATLG has been created. --- SysDB table $DATA.ESCSYSDB.MAPDBS created. --- SysDB table $DATA.ESCSYSDB.REGISTRY created.
Installation Start the Monitor Process tables are created. See Appendix A, Operations and Management, for details on the additional MapDB tables. AutoSYNC 4? create mapdb; --- AutoSYNC MapDB table $DATA.ASYMAPDB.SYNCUSER created. --- Map Database for AutoSYNC created. --- System Database updated --- Changes will take effect when the Monitor is restarted. --- Starting AutoSYNC Monitor $ZASY. Monitor 1.5.
Installation Authorize AutoSYNC The following is displayed in the EMS log: 15:55 10JUN04 128,04,355 AutoSYNC Monitor $ZASY Starting Version 1.6.
Installation Update the System Coldload and Shutdown Procedures If system security is set such that the Authorizer cannot log down to another user ID, the following options are available: 1. Configure all scheduled synchronizations with the same user id as the AutoSYNC Authorizer. All other users must request one-time synchronizations. 2. Configure multiple AutoSYNC environments, each with its own MAPDB and monitor process.
Installation Install the AutoSYNC EMS templates Update the shutdown procedures to include the shut down steps described below in Shutting Down AutoSYNC. Install the AutoSYNC EMS templates The AutoSYNC EMS templates are in the file ZASYTMPL. Please refer to the HP EMS template installation instructions to perform this step. Shutting Down AutoSYNC For an orderly shutdown of AutoSYNC, the following steps are recommended: 1. Use the SUSPEND SYNCUSER command to suspend all AutoSYNC users. 2.
Installation Running Different Versions of AutoSYNC 9. Use the STATUS SYNC command to ensure that synchronizations have been activated. Note that the same version of AutoSYNC must be running on the source and destination systems. Updates to the software should be planned accordingly.
3 One-time Synchronization This section presents a basic overview of AutoSYNC processing and the various options available for performing a one-time synchronization of Guardian files. AutoSYNC also supports the synchronization of OSS files, as described in Section 5, OSS File Synchronization. Most of the features described in this section also apply to scheduled synchronization, described in Section 4, Scheduled Synchronization.
One-time Synchronization File Sets File Sets The file sets describe the collection of files to be synchronized. A file set is a standard file pattern, such as \MYSYS.$DATA*.OBJ*.SERV*. There are a few restrictions on the source and destination file sets. Source file set Destination file set File Set Comparison Source file set The source file set must describe only local files. If the file set contains a system name (for example \NEWYORK), the system name must be the local Expand node name.
One-time Synchronization • • • • File Set Comparison Files that are opened for update. Copying files that are being updated often results in corrupt files. You can request such files to be synchronized by using the OPENUPDATE option. Broken or corrupt files. Excluded files. Files that match a file name or pattern in the specified list of excluded files are not synchronized. Files in subvolumes $.ZYQ*.*: These subvolumes are excluded from Guardian file sets because they contain OSS files.
One-time Synchronization Data Transfer and File Replacement Data Transfer and File Replacement Once the list of files to be synchronized is complete, AutoSYNC transfers the files using an efficient data-streaming protocol. Reading the source files, transferring the data from the source to destination system (or systems), and replacing the destination files is parallelized and proceeds at maximum throughput.
One-time Synchronization Ownership and Security Default PURGE Security To be replaced by AutoSYNC, existing destination files must be secured for purge access. If the destination is on a remote system, the files must be secured for remote purge access. Note that the recommended method to prevent the synchronization of a file is to list the file in the exclude file using the EXCLUDE option. However, you can also avoid synchronizing a destination file by setting PURGE security of the destination file to“-”.
One-time Synchronization Alternate Key Files LICENSE and PROGID File Attributes Files that have the LICENSE or PROGID attribute set are synchronized but the setting of these attributes are not carried over to the destination file set. Refer to Preserving Security Attributes on the Destination on page 3-18 for a method to preserve these attributes.
One-time Synchronization SQL Tables and Catalogs Example 2 File $DATA.DB.ACCT has its alternate key ACCTAK on subvolume $KEYS.DBKEYS. To synchronize the file and alternate key file to identically named files on the destination system, use the following command: SYNC $*.DB*.ACCT* TO \DEST.*.*.*; To synchronize the $DATA.DB.
One-time Synchronization SQL Views SQL index tables cannot be synchronized by themselves. AutoSYNC does not synchronize an index that is created after the table has been synchronized, unless the SQLREDEF option is specified (see Partitioned Files and Tables below); otherwise, the table must be modified or purged at the destination to cause the table and indexes to be resynchronized as a unit. SQL Views Protection Views If you synchronize a SQL table, protection views are implicitly synchronized.
One-time Synchronization Partitioned Files and Tables If source partitions are synchronized to identical volume names on the destination, specify the synchronization as: SYNC $DATA.DB.PARTFILE TO \.*.*.*; Identical name mapping is implied and no MAPNAMES file is required. If some of the source partition volume names are different on the destination, use the MAPNAMES option to describe the mapping of the secondary partition names. Note.
One-time Synchronization Partitioned Files and Tables Partitioned file To synchronize the partitioned file $DATA*.DB.PARTFILE to different subvolume and file names $DATA*.BACKDB.BACKFILE, the command must specify the primary partition mapping, SYNC $DATA01.DB.PARTFILE TO $DATA01.BACKDB.BACKFILE, MAPNAMES mapnames...; and the mapnames file must contain an entry for each secondary partition, as follows: $DATA02.DB.PARTFILE TO $DATA02.BACKDP.BACKFILE $DATA03.DB.PARTFILE TO $DATA03.BACKDP.BACKFILE ...so on..
One-time Synchronization Partitioned Files and Tables which says that everything in the source file set will be mapped to the same destination file set. Therefore, if you synchronize files and tables to the same location on the destination, you do not need to supply a mapnames-file.
One-time Synchronization Synchronization Options To find out the file name and catalog mapping being supplied to RESTORE, specify KEEPBACKUPINFILE and examine the ZRSnnnnn file in the destination MapDB subvolume. Synchronization Options Numerous options provide flexibility for synchronization. Most of options apply to both one-time and scheduled synchronizations.
One-time Synchronization Excluding Specific Files System swap files are also considered temporary and they are not synchronized. Excluding Specific Files You may create an edit file containing a list of volume, subvolume, and file names or patterns to be excluded from the source file set. Exclusion does not apply to the destination file set. To preserve efficiency when exclude lists are large, patterns are limited to trailing asterisks only.
One-time Synchronization Replacing Open Destination Files All other synchronization options specified apply to the selection of files to purge in the destination file set. For example, if the AUDIT option is specified, the audited files in the extra destination subvolumes are purged. Use the REPORT function to display the extra subvolumes that will be purged if SUBVOLPURGE is used in the synchronization.
One-time Synchronization Changing File Ownership starts, it checks the last modified timestamp for the file and synchronizes the file only if the OPENUPDATE interval has elapsed since the last update. In addition to specifying the OPENUPDATE option, the following conditions must be met: ° ° ° The file must not be audited. The file cannot be opened for exclusive access. Recovery must be enabled for the fileset. Changing File Ownership By default, the ownership of each synchronized file is preserved.
One-time Synchronization Synchronizing based on Binder or Linker timestamp the destination subvolume, but the primary source partition is older than the primary destination partition, the index will not be synchronized. Synchronizing based on Binder or Linker timestamp Certain operations performed on an object file cause the MODIFICATION timestamp of the file to change but do not alter the file in ways that require synchronization.
One-time Synchronization Compression exist; the PERSISTENT ON attribute retains the SAFEGUARD file-protection record after the file is purged. Compression To reduce the amount of data that is transferred between the source and a remote destination, AutoSYNC can compress the data. The LZW compression algorithm typically eliminates 50 percent to 60 percent of the volume of the data.
One-time Synchronization BACKUP and RESTORE Programs Saved recovery files are purged when the replacement file has been transferred successfully. Recovery occurs immediately if the destination ASYSYNC process detects the failure that would leave the file in a corrupt state. If, however, a cpu or media failure prevents immediate recovery, any subsequent synchronization of the affected file sets will perform the recovery.
One-time Synchronization Preserving Guardian LICENSE and PROGID Attributes Running the Destination Synchronization Process as a Local Process Preserving Guardian LICENSE and PROGID Attributes The LICENSE and PROGID attributes apply to object files. They are managed carefully by the NSK file system to preserve system security. The basic security rules that apply to all operations, including those performed by AutoSYNC, are as follows: • • Only the local Super user ID can set the LICENSE attribute.
One-time Synchronization Preserving Other Security Attributes Preserving Other Security Attributes You can also replace destination files even if you have elected to keep local purge access to the files. To keep the PURGE security of a file on the destination system as “O” or “G” for example, you will need to make the ASYSYNC or ASYSYNCP (if synchronizing over TCP/IP) privileged, allowing it to run as a local process.
4 Scheduled Synchronization As mentioned previously, most one-time synchronization features described in Section 3, One-time Synchronization also apply to scheduled synchronizations. This section describes features and concepts that are specific to scheduled synchronizations.
Scheduled Synchronization The AutoSYNC Authorizer Scheduled synchronizations, however, are managed by the AutoSYNC monitor process. The AutoSYNC monitor acts as an agent for the AutoSYNC users, but also must ensure that each user has access rights to the files that are to be synchronized. The following pages describe how AutoSYNC is granted sufficient authority to act as an agent for multiple users, but in a way that prevents granting enhanced access rights to those users.
Scheduled Synchronization • • AutoSYNC Users SUSPEND SYNCUSER ACTIVATE SYNCUSER Caution. The AutoSYNC administrator can manipulate the AutoSYNC configuration to gain access to files that may not be secured for access to the administrator user ID. Thus, the AutoSYNC authorizer must exercise discretion when designating the administrator. AutoSYNC Users To configure file sets to be synchronized, you must be a AutoSYNC user; the administrator makes you a user through the ADD SYNCUSER command.
Scheduled Synchronization Data Access Control Data Access Control Users of AutoSYNC, especially the authorizer, should understand how multiple users are managed by AutoSYNC, a non-privileged subsystem. The monitor process reads the SYNCUSER table to determine which users are configured to use AutoSYNC. The SYNCUSER table should be secured to prevent unauthorized access. For each AutoSYNC user, the monitor accesses the SFggguuu table to fetch the synchronizations configured by the user.
Scheduled Synchronization Interval, Start, and Stop Times Interval, Start, and Stop Times The basic scheduling control is the INTERVAL between synchronizations. The default interval is 5 minutes, but you can specify another interval. In addition, you can specify both START and STOP times. If you specify only START, AutoSYNC performs a daily synchronization at the specified time and the INTERVAL, if specified, is ignored.
Scheduled Synchronization ALTER SYNCFILESET (and ALTER BATCHID) ALTER SYNCFILESET (and ALTER BATCHID) You can change any of the file set options, including the source and destination file sets. See ALTER SYNCFILESET on page 8-35 and ALTER BATCHID on page 8-27. Altering a SYNCFILESET will cause it to be rescheduled and performed at the first opportunity. DELETE SYNCFILESET (and DELETE BATCHID) The synchronization is terminated and deleted from the schedule.
Scheduled Synchronization Security and Authorization Security and Authorization A scheduled synchronization is managed by the AutoSYNC monitor process. The monitor process starts ASYSYNCP processes for each AutoSYNC user who has been added and who has configured synchronization file sets. Initially, the ASYSYNCP process runs under the user ID of the authorizer, but when accessing files, the ASYSYNCP process uses only the user ID and access rights of the user who configured the file sets.
Scheduled Synchronization Scheduled Synchronization Flow 2. At the TACL prompt, SQL compile ASYSYNCP: SQLCOMP/IN ASYSYNCP/NOREGISTER ON 3. At the TACL prompt, accelerate ASYSYNC AXCEL ASYSYNCP, ASYSYNCP or OCA ASYSYNCP 4. Then, as SUPER.SUPER, issue the following command FUP LICENSE ASYSYNCP Note. Disabling log down simply prevents use of another user ID by the Super user. The Super user still has unfettered access to all data on the system.
Scheduled Synchronization ASYSYNCP process The monitor maintains (and checkpoints) a table of ASYSYNCP processes and periodically checks that these processes are still alive. If any process fails, or if there are network or CPU failures that affect these processes, the monitor takes an appropriate corrective action. Simple failures will be retried three times immediately and then once every 30 minutes.
Scheduled Synchronization Variant for Synchronization over TCP/IP The names of files that are missing from the destination, as well as files that meet other criteria for synchronization, are returned to the ASYSYNCP process. After the exchange of file information, the data stream of file data is received and used to create or replace the destination copies of the source files.
5 OSS File Synchronization This section describes the features and options available for the synchronization of OSS files and directories. These features apply to both one-time synchronization and scheduled synchronization.
OSS File Synchronization Source File Set with these commands, the complete OSS pathname is displayed. Pathnames wrap to the next line if necessary. AutoSYNC synchronizes the following OSS disk files: • • • • Directories Regular files Symbolic links Hard links Unlike an empty Guardian file system subvolume, an empty directory is synchronized.
OSS File Synchronization Links Purging Destination Files If a file, a link or a directory exists in the destination file set but does not exist in the source, it can be purged on the destination by specifying the PURGE option. If PURGE is specified, subordinate directories that do not exist in the source are purged. This differs from the purge processing for Guardian subvolumes. Entire Guardian subvolumes are purged only if the SUBVOLPURGE option is specified in addition to the PURGE option.
OSS File Synchronization • Links By default, when ALLSYMLINKS is specified, a symbolic link is synchronized also when the symbolic link points outside the synchronization file set. In this case, the symbolic link may point to a file that does not exist. A message is written to the logfile: File not in source file set, synchronizing uncontained symbolic link. • When NO ALLSYMLINKS is specified, AutoSYNC does not synchronize a symbolic link that points outside the synchronization file set.
OSS File Synchronization Displaying OSS Files and Directories Another side effect of the source and destination timestamps inherent mismatch is that the TIMEEXACT option is ignored for symbolic links. Displaying OSS Files and Directories For convenience, the AUTOSYNC command LS can be used to display information and statistics on OSS files, directories, and directory trees. The AUTOSYNC command LS is equivalent to the OSS shell command ls and has several options.
OSS File Synchronization • • Ownership and Security The user is SUPER.SUPER. The Super ID always has permission to synchronize files for other user IDs. The SYNCHRONIZE or ADD SYNCFILESET option OWNER is specified and the user is the same as the specified owner or, the OWNERGROUP is specified and the group ID is the same as the specified owner. In these cases the destination files are all owned by the user.
OSS File Synchronization Changing Default Ownership of Symbolic Links Changing Default Ownership of Symbolic Links To set the owner of a synchronized symbolic link to a user different than the user performing the synchronization or to set the owner to the user ID specified through the OWNER or OWNERGROUP option, the destination ASYSYNC process, or ASYSYNCP if synchronizing over TCP/IP, has to be privileged.
OSS File Synchronization Synchronization Options WARNING. By configuring a callable ASYSYNC or ASYSYNCP, the systems administrator allows a syncuser to perform operations on the remote system, that would have previously required the knowledge of the local password. This possibility is particularly critical to remember if the syncuser is SUPER.SUPER.
OSS File Synchronization • • • • • • • Synchronization Options KEEPNOPURGEUNTIL MAPNAMES MAXBACKUP NETWORKSECURE SUBVOLPURGE USEBINDERTIME ZZINCLUDE The following options are not supported: • • COMPRESS RENAMEOPEN HP AutoSYNC User’s Guide—522580-016 5-9
OSS File Synchronization Synchronization Options HP AutoSYNC User’s Guide—522580-016 5-10
6 Triggers This section gives a basic overview of AutoSYNC triggers. Triggers allow a user to run a TACL or OSS shell command for each synchronized file that was successfully moved to the destination system (TRIGGER) or run a TACL or OSS shell command after a complete file set has been successfully synchronized (TRIGGERONFILESET).
Triggers Commands The feature is invoked by specifying TRIGGER or TRIGGERONFILESET option in the SYNCHRONIZE, ADD SYNCFILESET, or ALTER SYNCFILESET commands: SYNCHRONIZE … TRIGGER “command-text”; or SYNCHRONIZE … TRIGGERONFILESET “command-text”; ADD SYNCFILESET … TRIGGER “command-text”; or ADD SYNCFILESET … TRIGGERONFILESET “command-text”; Commands TRIGGER Specify a complete TACL or OSS command line that includes built-in arguments that are replaced by appropriate file names.
Triggers TRIGGERONFILESET If no archive name is found, AutoSYNC will pass the placeholder question mark (?). For OSS, files the absolute pathname is specified. Recovery files have a name chosen by AutoSYNC. Such files are usually purged by AutoSYNC when the replacement file has been synchronized, but triggers may preserve these files by renaming them • #MAPNAMES is replaced by the mapnames-file that is configured for the syncfileset using the MAPNAMES option.
Triggers Examples Examples TRIGGER examples 1. SQL-compile the synchronized object file myslqprg. SQLCOMP is invoked in the TACL macro sqlcmpme that sets up the defines for the compilation: SYNC trigtest.mysqlprg TO \sys2.$data.trigtest.*, TRIGGER "sqlcmpme #syncfile", LOGFILE synctest.log; Where sqlcmpme uses obey file mydefine before launching SQLCOMP: ?tacl macro #frame #push :file #set :file %1% system \sys02 obey mydefine sqlcomp /in [:file], out $s.#sqlcomp/catalog \sys2.vol.
Triggers TRIGGER examples where the TRIGCOND macro is: ======================================================== == Name: TRIGCOND == Purpose: Sample conditional sqlcomp depending on file == code and OS version == Usage: == SYNC $data.test.sqobj TO \paris.$datab.testb.
Triggers TRIGGERONFILESET Example TRIGGERONFILESET Example Change the library pointer in object files GETME and GETMETOO to GETUL after synchronizing the entire object subvolume to a location with a different name:y SYNC $data.objd.* TO \paris.$datab.objp.
Triggers Note on Triggers that invoke SQLCI Instead, the user should use the following command: SYNC $data.test.sqacct TO \paris.$datab.testb.
Triggers Note on Triggers that invoke SQLCI HP AutoSYNC User’s Guide—522580-016 6-8
7 Synchronization Over TCP/IP AutoSYNC supports the use of a direct TCP/IP network connection to communicate between the source and destination AutoSYNC processes. This section describes how to setup synchronizations over TCP/IP and how AutoSYNC establishes and manages the connection. The topics covered are organized as follows: Overview Configuring a TCP/IP connection On both Source and Destination Systems On the Destination System On the Source System .
Synchronization Over TCP/IP On both Source and Destination Systems On both Source and Destination Systems Set the TCP/IP port numbers for AutoSYNC A TCP/IP SERVICES file is used to specify the port number for use by defined services. The name of the service is ASYSYNC-ZASY where ZASY is the name of the AutoSYNC monitor process $ZASY that is used on both systems to coordinate the TCP/IP connections.
Synchronization Over TCP/IP On the Destination System On the source system, the ASYSYNC process for a scheduled synchronization inherits the define from the monitor process. For a one-time synchronization, the ASYSYNC process inherits the define directly from the AUTOSYNC CI process. There is no default port number; if you do not configure a port number, synchronization over TCP/IP is not possible. Consult your local networking specialist to choose an appropriate port number.
Synchronization Over TCP/IP On the Source System Thus, you must specify all TCPIP, TCPSAM or TCP6SAM processes that correspond to those addresses, or the synchronization will fail to make the connection. Consult a networking specialist to determine the names of the TCPIP, TCPSAM, and TCP6SAM processes. Use the following SCF commands to understand your local networking configuration: LISTDEV TCPIP and INFO PROCESS $, DETAIL.
Synchronization Over TCP/IP Synchronization over TCP/IP to the source and destination file sets. If the destination is on a remote system, the destination file set must include the Expand node name of the remote system. ADD SYNCF[ILESET[ src-fileset TO \systname.dest-fileset, HOST { name | IP address } [, sync-options]; Synchronization over TCP/IP As in synchronizations over an Expand link, the source ASYSYNCP receives its file sets from the monitor or Autosync CI.
Synchronization Over TCP/IP Synchronization over TCP/IP Consequently, ASYSYNCP EXECUTE security must be set to "N" if any non SUPER.SUPER Syncusers are configured to perform synchronizations. Logging When synchronizing over an Expand link, the destination ASYSYNC opens the log file SLggguuu through Expand and writes records directly to the log.
8 Command Interpreter The AUTOSYNC command interpreter is the primary interface for invoking utility functions and for controlling and monitoring the AutoSYNC. This chapter describes the AutoSYNC command interpreter and is organized as follows: Running the Command Interpreter Command Syntax Command Summary Command Descriptions Running the Command Interpreter AUTOSYNC is invoked through the TACL run command or equivalent. The TACL syntax is: [RUN] $SYSTEM.AUTOSYNC.
Command Interpreter Command Syntax Command Syntax AUTOSYNC is a conversational-mode command interpreter. Most commands can be continued over many lines with no need for a continuation character; as a consequence, each command must be terminated with a semi-colon (;). A few commands, such as FC and RUN, are single-line commands and are terminated by the end of line; such commands can be continued by placing an ampersand (&) at the end of the line.
Command Interpreter Command Summary Table 8-1. AutoSYNC Commands (continued) Command Name Description ALTER LOCAL Changes the value of a AutoSYNC local parameter ALTER MAPDB Changes attributes of MapDB and the monitor. ALTER MONITOR Changes attributes of a monitor process. ALTER SYNCFILESET Modifies configuration of scheduled synchronization by file set. ALTER SYNCUSER Modifies configuration of an AutoSYNC user. AUTHORIZE Sets the AutoSYNC authorizing user ID.
Command Interpreter Command Summary Table 8-1. AutoSYNC Commands (continued) Command Name Description STATUS BATCHID Displays status information for a specific batchid. STATUS MONITOR Displays the status of a monitor process. STATUS SYNC Displays status for a synchronization process. STATUS SYNCFILESET Displays status information for a specific syncfileset. STOP MONITOR Stops the monitor process. SUSPEND BATCHID Terminates and suspends scheduled synchronization for specified BATCHID.
Command Interpreter Command Descriptions Table 8-2. General Purpose Commands Command Name Description LOG Outputs a copy of AUTOSYNC command input and output to a file. OBEY Executes a sequence of AUTOSYNC commands from a file. OUT Directs all AUTOSYNC command output to a file. PURGEDATA Clears all data from a file. RENAME Renames an Enscribe file RESET Resets completion code. RUN[D] Executes an external process during an AUTOSYNC session. STATS Examines an Enscribe structured file.
Command Interpreter ACTIVATE SYNCFILESET * | batchid | batchid - batchid a collection of file sets by BATCHID or a range of BATCHIDs. Up to 32 ranges can be entered. A BATCHID is an integer and is of the same format as an alternate key tag, that is, two characters in quotes or a number between 1 to 32767. If batchid-batchid is specified, the second batchid must be greater than the first. * designates all configured batchids for the user.
Command Interpreter ADD SYNCFILESET ADD SYNCFILESET Adds a file set to the scheduled synchronizations. See the Synchronization Options on page 3-12 for an overview of synchronization.
Command Interpreter ADD SYNCFILESET common-options (continued) { TRIGGER { tacl-command | oss-shell-command } } { TRIGGERONFILESET { tacl-command | oss-shell-command } } guardian-only-options: { AUDITED } { AUTOCREATECAT[ALOG] } { BACKUP backup-program } { CATALOG SQL-catalog } { COMPRESS } { KEEPNOPURGEUNTIL } { MAPNAMES mapnames-file } { MAXBACKUP num-of-files } { NETWORKS[ECURE] | NO NETWORKS[ECURE] } { NOSAFEGUARD | NO NOSAFEGUARD } { RECOVER[Y] | NO RECOVER[Y] } { RENAMEO[PE
Command Interpreter ADD SYNCFILESET OSS-only-options: { ALLHARDLINKS | NO ALLHARDLINKS } { ALLSYMLINKS | NO ALLSYMLINKS } { OWNERG[ROUP] { group-id | group-user } } { SECURE “rwxrwxrwx” } src-file-set a file set on the local system that is the source of the synchronization operation. If a system name is specified, it must be the local Expand node name. dest-file-set the location of the file set that is to be updated.
Command Interpreter ADD SYNCFILESET OSS file set definition • • • • • The source and destination file sets must be specified with the absolute pathname. Wild-card characters are not supported; specify an absolute directory name or an absolute file name. If the source file set is a single file, the file is moved to the destination specified. If the directory does not exist at the destination location, the directory is created.
Command Interpreter ADD SYNCFILESET instructs AutoSYNC to exclude the files listed in exclude-file from synchronization. exclude-file an edit file that contains names of the files, subvolumes or disk volumes in the syncfileset that are to be excluded from the synchronization. An exclude file can be shared among syncfilesets. When starting a new synchronization ASYSYNC rereads the exclude list if the exclude file has been updated. Each entry must be on a separate line.
Command Interpreter ADD SYNCFILESET instructs AutoSYNC to exclude the files listed in the excludep-file from being purged on the destination, if the files do not exist on the source. EXCLUDEPURGE only applies when PURGE or SUBVOLPUREGE are also specified. Otherwise, EXCLUDEPURGE is ignored. excludep-file an edit file that contains the names of the files, subvolumes or disk volumes in the destination syncfileset that are excluded from being purged.
Command Interpreter ADD SYNCFILESET (for synchronization over TCP/IP only) address of the destination system specified as an internet address or the configured internet name of the host system. tcpip-address is specified in the dotted decimal format nnn.nnn.nnn.nnn (example: 16.107.164.200). host-name is the name by which the destination system is known on the internet, specified as a character string of 49 or fewer characters (example: foxii.cac.cpqcorp.net). There is no default value for this option.
Command Interpreter ADD SYNCFILESET LISTA[LL] includes all file names that are processed by BACKUP and RESTORE in the log. OPENU[PDATE] [ interval [SECONDS | MINUTES | HOURS] ] specifies that a source file that is opened for update is eligible for synchronization. By default, AutoSYNC does not synchronize files that are opened for update. OPENUPDATE overrides the default behavior.
Command Interpreter ADD SYNCFILESET If this option is omitted, the priority of ASYSYNC is set to the priority of the syncuser. Note. Be aware that running ASYSYNC at high priority may impact system performance as the process is CPU bound. PURGE specifies that any file that is in the destination file set but is not in the source file set will be purged. By default AutoSYNC does not purge any file in the destination file set.
Command Interpreter ADD SYNCFILESET When the START option is specified the INTERVAL is ignored and one synchronization per day is performed at the start time. If the start time is larger than the stop time, then it is assumed the synchronization window crosses into the next day. If STOP is omitted, the synchronization occurs once a day at start time. In this case the INTERVAL option is ignored. STOP hh:mm:ss the ending time of the synchronization window.
Command Interpreter ADD SYNCFILESET a TACL command or OSS shell command that is executed by AutoSYNC for each file in the synchronized file set after the file has been moved to the destination. The command line contains four built-in arguments. Each argument starts with a pound sign (#) and is terminated by a TACL delimiter character such as a space or a comma, or by an OSS shell delimiter such as a space or a semicolon (;) and so on. Arguments can be specified in either upper or lower case.
Command Interpreter ° ° ° ° ADD SYNCFILESET #FSETCOUNT is the number of files in the source file set. #SYNCCOUNT is the number of files replaced on the destination. #COUNTNOTBKUP is the number of files not backed up. #COUNTNOTREST is the number of files that were not restored. guardian-only-options AUDITED instructs AutoSYNC to synchronize audited files. By default, audited files are not synchronized.
Command Interpreter ADD SYNCFILESET This option overrides the global parameter [AUTO]SYNCCOMPRESS Note Compression is CPU-intensive and can reduce total throughput unless the connection between source and destination systems is slow. KEEPNOPURGEUNTIL specifies that the NOPURGEUNTIL attribute of a source SQL table or alternate index is preserved on the destination.
Command Interpreter ADD SYNCFILESET The entries follow file mapping entries and are preceded by the keyword "CATALOGS:", including the colon (:), as follows: $DATA TO $BACK \NEWYORK.$DATA*.DB TO $BACK.BACKUPDB $DATA*.DB.PARTFILE TO \TOKYO.* CATALOGS: $CATS.BACKCAT FOR $BACK.*.* $CATS.DBCAT FOR $BACK.BACKUPDB.* \TOKYO.$CATS.DBCAT for \TOKYO.*.*.* CATALOGS entries are checked for proper syntax, and passed along to the RESTORE process.
Command Interpreter ADD SYNCFILESET written to the logfile to notify the user that the Safeguard setting for the source file has been turned off on the destination. The default is NO NOSAFEGUARD and the ACL for a source file is propagated to the synchronized destination file. RECOVER[Y] | NO RECOVER[Y] enables or disables the recovery option. If recovery is enabled, AutoSYNC saves a copy of an existing destination file to an archive subvolume before it restores a later copy of the file.
Command Interpreter ADD SYNCFILESET the security of destination files. If this option is omitted, the security of a destination file is the same as that of the source file unless NETWORKSECURE is on and the destination files are on a destination system. SQLREDEF instructs AutoSYNC to synchronize SQL tables based on differences in the redefinition timestamps, in addition to differences in modified timestamps.
Command Interpreter ADD SYNCFILESET By default, ZZ* files and system swap files are not synchronized. OSS-only-options When describing synchronization options for OSS files below, the phrase "uncontained links" refers to files that are hard links or symbolic links that have references to files that reside outside of the synchronization file set. An uncontained hard link is a link to an inode where another link to the same inode resides outside the synchronization file set.
Command Interpreter ADD SYNCUSER the security of the destination file set. "rwxrwxrwx" settings are for user, group and others. The accepted values are: ° ° ° ° "r" grants read access "w" grants write access "x" grants execute access "-" denies access for read, write and execute setting. If this option is omitted, the security of the destination file set is the same as that of the source file set.
Command Interpreter ADD SYNCUSER The log file is used by AutoSYNC to display messages pertaining to the synchronization operations it performs on behalf of this user. ADD SYNCU[SER] { user-id | user-name } [ ,option... ]; option is: { BACKUP backup-program } { CPU n | (n1, n2, n3...
Command Interpreter ADD SYNCUSER suppresses the generation of a class of EMS events that are considered optional. Suppressing the optional messages reduces the volume of output sent to the EMS collector. The types of messages suppressed are the following: • • Synchronization process START and STOP messages. Messages reporting configuration changes made by the administrator to a file set or a batch ID. Events that report failures are never suppressed and are always logged to the EMS collector.
Command Interpreter ALTER BATCHID items include system messages, operations such as starting and stopping processes, creating and purging files, opening and closing files, and so on. ALTER BATCHID Modifies the options of the syncfilesets in the specified BATCHID range. ALTER BATCH[ID] batchid-range [, batchid-range ], ... {, alter-option... }; batchid-range is { * | batchid | batchid - batchid } * | batchid | batchid - batchid a collection of file sets by BATCHID or a range of BATCHIDs.
Command Interpreter ALTER FILE ALTER FILE Similar to the FUP ALTER (or SQLCI ALTER TABLE) commands, changes certain attributes of Enscribe files or mapped tables. ALTER FILE file-set {, option...
Command Interpreter ALTER FILE BUFFERED | NO BUFFERED turns buffering on or off. CLEARONPURGE | NO CLEARONPURGE erases disk free space when files are purged. CODE numeric file code of the file. file-code is an integer between 0 and 65535. Codes 100 to 999 are reserved for use by HP. LOCKLENGTH length the byte count of the key for generic locks. length is between 0 and the primary key length of the file. MAXEXTENTS size the maximum disk allocation extents.
Command Interpreter ALTER GLOBAL VERIFIEDWRITES | NO VERIFIEDWRITES sets the mode of file writes: verified or not verified. The default is NO VERIFYWRITES. ALTER GLOBAL Sets or changes global AutoSYNC values. Global values are stored in the SYSDB table called REGISTRY and affect all processes executed using the SYSDB.
Command Interpreter ALTER GLOBAL enables or disables the recovery option. If recovery is enabled, AutoSYNC saves a copy of a destination file to an archive subvolume before it restores a later copy of the file. If a failure occurs during the synchronization and the file is corrupted, the archived copy is recovered and replaces the corrupted file. This setting can be overridden by setting the syncfileset attribute or one-time sync option RECOVERY. The default is ON.
Command Interpreter ALTER LOCAL ALTER LOCAL Sets or changes a local parameter value. Local parameters are the same as global parameters and have the same purpose, but they are in effect for the current Monitor (MapDB) only. Local parameter values are stored in the MapDB table REGISTRY. Setting a local parameter value overrides the global (default) value.
Command Interpreter ALTER MAPDB monitor process is restarted. These attributes can also be changed using ALTER MONITOR. ALTER MAPDB subvol { , option... } option is: { BACKUP cpu-number } { HOMETERM [ file-name | NONE ] } { MONITOR monitor-name } { PRIMARY cpu-number } { PRIORITY process-priority } subvol the name of the MapDB subvolume. If this command is used to change the name of the monitor, the current monitor process must first be stopped.
Command Interpreter ALTER MONITOR a process priority (in the range of 1 to 199) for the monitor process The PRIMARY, BACKUP, and PRIORITY options specify process creation parameters that are used when starting the monitor process. ALTER MONITOR Modifies attributes of an existing monitor process. The changes take effect when the monitor process is restarted. These attributes can also be changed using ALTER MAPDB. ALTER MONITOR [ process-name ] {, option...
Command Interpreter ALTER SYNCFILESET ALTER SYNCFILESET Modifies the options of the syncfilesets in the specified fset-id range. ALTER SYNCF[ILESET] fset-range [, fset-range ], ... {, options...
Command Interpreter ALTER SYNCFILESET common-options: (continued) { START hh:mm:ss | NO START } { STOP hh:mm:ss | NO STOP } { SUSPEND } { TRIGGER { tacl-command | oss-shell-command }| NO TRIGGER } { TRIGGERONFILESET { tacl-command | oss-shell-command } | NO TRIGGERONFILESET } { TIMEEX[ACT] | NO TIMEEX[ACT] } { TO file-set } guardian-only-options: { AUDITED | NO AUDITED } { AUTOCREATECAT[ALOG] } { BACKUP backup-program | NO BACKUP } { CATALOG SQL-catalog | NO CATALOG } { COMPRESS | NO
Command Interpreter ALTER SYNCFILESET guardian-only-options: (continued) { SQLREDEF | NO SQLREDEF } { SUBVOLPURGE | NO SUBVOLPURGE } { USEBINDERTIME | NO USEBINDERTIME } { ZZINC[LUDE] | NO ZZINC[LUDE] } OSS-only-options: { ALLHARDLINKS | NO ALLHARDLINKS } { ALLSYMLINKS | NO ALLSYMLINKS } { OWNERG[ROUP] {{ group-id | group-user } | NO OWNERG[ROUP]} { SECURE “rwxrwxrwx” | NO SECURE } * | fset-id | fset-id - fset-id identifies a synchronization file-set ID or a range of file-set IDs.
Command Interpreter ALTER SYNCFILESET All file sets that have the same batchid for a given user and destination system are synchronized in sequence by a single ASYSYNC process. To have more than one ASYSYNC process for the same destination system, use different BATCHIDs. EXCLUDE exclude-file | NO EXCLUDE instructs AutoSYNC to exclude (or remove exclusion of) the files listed in exclude-file from synchronization.
Command Interpreter ALTER SYNCFILESET while exclude-file entry: /home/group/user/destdir/A/KA* or /home/group/user/destdir/A/KA/ excludes directory: /home/group/user/destdir/A/KA/ Note that some performance degradation could occur if an exclude file containing a long list of files is shared by multiple syncfilesets.
Command Interpreter ALTER SYNCFILESET a file set on the local system that is the source of the synchronization operation. If a system name is specified, it must be the local Expand node name. For file set formats, see Guardian file set definition and OSS file set definition on page 8-10. FROMO[WNER] user-id | NO FROMOWNER instructs Autosync to (or not to) synchronize only the source files owned by the specified user ID.
Command Interpreter ALTER SYNCFILESET The name of the input file is in the format of ZBKnnnnn for BACKUP and ZRSnnnnn for RESTORE, where nnnnn is a serial number. The files are placed in the MapDB subvolume. The exact file name is displayed in the syncuser’s log file SLnnnggg. Input files are not created if no files require synchronization. For BACKUP, the input file is always created when files are synchronized.
Command Interpreter • ALTER SYNCFILESET the global SYNCRECOVER parameter is set to OFF. Note for OSS file sets. AutoSYNC only synchronizes updates that are actually sent to the disk process. Any updates that are buffered in the application process (such as performed by the file system or C runtime) are not synchronized until they are transmitted to the disk process. OWNER [ user-id | user-name ] | NO OWNER the owner of the destination file set.
Command Interpreter ALTER SYNCFILESET Extra SQL tables and views are purged explicitly, indexes and secondary partitions are purged implicitly, with the main table. OSS files and directories: ° If PURGE is specified, subordinate directories not present in the source file set are purged on the destination. Note that this is different than the processing of Guardian subvolumes. Guardian subvolumes are not purged unless SUBVOLPURGE is specified.
Command Interpreter ALTER SYNCFILESET If STOP is omitted but START is specified, synchronization happens once a day at the specified START time. TIMEEX[ACT] | NO TIMEEX[ACT] a file is synchronized if the modification timestamp of the source does not exactly match the modification timestamp of the destination file. By default, a file is synchronized only when the source file has been modified after the destination file was last modified.
Command Interpreter ALTER SYNCFILESET specified, AUTOSYNC substitutes the placeholder question mark (?) for the mapnames-file. This option is applicable to Guardian syncfilesets only. TRIGGERONFILESET { tacl-command | shell-command } | NO TRIGGERONFILESET specifies a tacl or OSS shell command that is executed by AutoSYNC after the complete file set has been successfully synchronized. This is in contrast with the TRIGGER option that executes the tacl or shell command for each file that is synchronized.
Command Interpreter ALTER SYNCFILESET The security of the newly created destination catalog is set to the syncuser's default Guardian security. If the catalog already exists, the option is ignored. To register the object in a different catalog, use the CATALOG option in addition to AUTOCREATECATALOG. BACKUP backup-program | NO BACKUP the name of the BACKUP object program to use on the source system when a file set is synchronized.
Command Interpreter ALTER SYNCFILESET The mapnames-file is an edit or entry-sequenced file located on the destination system. File name mapping: The mapnames-file entries are similar to, but not exactly the same as, the MAP NAMES option of RESTORE. Each line of the MAPNAMES file contains a single mapping of the form. source-file-pattern TO destination-file-pattern The file name can be partially qualified; AutoSYNC replaces the missing portion of the name with the wild card asterisk (*).
Command Interpreter ALTER SYNCFILESET When processing large file sets, AutoSYNC divides the synchronized files into subsets, each processed by a separate BACKUP process. BACKUP performance decreases significantly with the number of files it processes. To avoid performance degradation, AutoSYNC starts multiple BACKUP processes, in sequence, and assigns to each no more than the configured num-of-files. The default is 1000.
Command Interpreter ALTER SYNCFILESET specifies (or disables the option) that destination unaudited files that are opened for shared READ access or running object files are to be renamed to allow AutoSYNC to replace them. The syncuser must have PURGE privileges to the destination file for the RENAME to succeed. Both the old and new file names are recorded in an AutoSYNC index file called AAINDEX, located in the MapDB subvolume.
Command Interpreter ALTER SYNCFILESET instructs AutoSYNC to process (or not) subvolumes in the destination file set even if the subvolumes do not exist on the source file set. All other synchronization options specified apply to the selection of files to purge in the destination file set. For example, if the AUDIT option is specified, the audited files in the extra destination subvolumes are purged. The REPORT function also lists the extra subvolumes if both PURGE and SUBVOLPURGE are specified.
Command Interpreter ALTER SYNCFILESET If NO ALLHARDLINKS is specified, uncontained hard links are not synchronized. The default in ALLHARDLINKS. ALLSYMLINKS | NO ALLSYMLINKS synchronizes all symbolic links in the specified syncfileset, including uncontained symbolic links. A warning is written into the log file when an uncontained symbolic link has been synchronized. If NO ALLSYMLINKS is specified, uncontained symbolic links are not synchronized.
Command Interpreter ALTER SYNCUSER ALTER SYNCUSER Modifies the parameters that were specified for a AutoSYNC user through the ADD SYNCUSER command. ALTER SYNCU[SER] { user-id | user-name } [ ,option... ]; option is: { BACKUP backup-program | NO BACKUP } { CPU { n | ( n1, n2, n3 ...
Command Interpreter ALTER SYNCUSER EMSSUPPRESS | NO EMSSUPPRESS suppresses the generation of a class of EMS events that are considered optional. Suppressing the optional messages reduces the volume of output sent to the EMS collector. The types of messages suppressed are the following: • • Synchronization process START and STOP messages. Messages reporting configuration changes made by the administrator to a file set or a batch ID.
Command Interpreter AUTHORIZE messages, operations such as starting and stopping processes, creating and purging files, opening and closing files, and so on. AUTHORIZE By default, the user that starts the AutoSYNC monitor process authorizes all sync users to perform scheduled synchronizations. Thus, in a multiuser environment, the monitor process must be started by the super user or the group manager (n, 255) of all sync users.
Command Interpreter CLEAR AUTOSYNC integer arithmetic. If any operand contains a decimal point, the expression is evaluated to four decimal places of accuracy. CALC expression; expression is { constant } { expression op const } { ( expression ) } op is { + | – | * | / } constant an ordinary decimal constant with or without a decimal point, or a binary, octal, or hexadecimal integer (prefixed by %b, %, or %h, respectively.
Command Interpreter COPY COPY Copies records or rows from an input file to an output file or displays them to an output device. It is analogous in function to the FUP COPY command. At this time, some of the FUP options pertaining to tape devices are not supported.
Command Interpreter COPY { SKIPMATCH } { TRIM trim-character } { UNLOADIN | NO UNLOADIN } { UNSTRUCTURED } { VARIN } out-options is: { BLOCKOUT <1 to 32767> } { EBCDICOUT } { FOLD } { PAD pad-character } { RECOUT <1 to 4096> } { REWINDOUT | NO REWINDOUT } { UNLOADOUT | NO UNLOADOUT } { UNSTROUT } { UPDATE } { VAROUT } display-options is { O[CTAL] } { D[ECIMAL] } { H[EX] } { A[SCII] } { BYTE[S] } { NO HEAD } in-file the name of the file that is input to the copy co
Command Interpreter COPY out-file the name of the file that is destination of the copy command. It can be a process, a tape, terminal, a printer or disk file. Supported disk files include edit files, spooler files, Enscribe structured and unstructured files. If the out-file is omitted, the destination of the COPY is the AutoSYNC out file, which is by default the terminal control-options. COUNT num-records the number of records or rows to be copied. If omitted, all records are copied.
Command Interpreter COPY key-spec is the alternate key tag (1 or 2 bytes) designating the alternate key to be used for positioning. ALTKEY key-value [, key-value] the alternate key of the starting record or row. The format of key-value is described above. UPSHIFT converts lowercase characters to uppercase. in-options BLOCKIN the number of bytes in an input block that is requested in a single physical read operation. When BLOCKIN is not specified, the RECIN value is used.
Command Interpreter COPY REVERSE the input file is to be READ from the starting record in reverse order. REWINDIN | NO REWINDIN (magnetic tapes only) the tape is rewound (or not rewound) when the EOF is read from the tape. If NO REWINDIN is specified, the tape remains positioned without rewinding.The default is REWINDIN. This option also applies to labeled tapes. SHARE the file is to be opened in shared-exclusion mode. The default is protected.
Command Interpreter COPY EBCDICOUT the output characters are translated from ASCII to their EBCDIC equivalents. FOLD the input records or rows that are longer than the output record length are divided into as many output records as needed to copy the entire input record or row. PAD pad-character the input record containing fewer bytes than the output record length is padded with pad-character. Specify pad-character as a single byte character in quotation marks or as an integer in the range of 0 to 255.
Command Interpreter CPUS display-options AUTOSYNC display options differ from FUP display options. AUTOSYNC applies all output data transformations (RECOUT, BLOCKOUT, PAD, and so on) before converting the data to display format, whereas FUP does it the other way around. As a result, AUTOSYNC only formats the data itself and displays it as it would be written to a disk file. Entry-sequenced record keys are displayed in the 64-bit format. O[CTAL] the output is in octal and ASCII format.
Command Interpreter CREATE MAPDB Since the MapDB is a set of SQL tables, the volume on which it is created must be TMF protected. CREATE MAPDB subvol [ , option... ]; or CREATE MAPDB ON volume [ , option... ]; option is { BACKUP cpu-number } { CATALOG SQL-catalog } { HOMETERM file-name } { MONITOR process-name } { PRIMARY cpu-number } { PRIORITY process-priority } { SECURE “rwep” } subvol the subvolume where the MapDB table is to be created.
Command Interpreter CREATE SYSDB The PRIMARY, BACKUP, and PRIORITY, and HOMETERM options specify process creation parameters to be used when starting the monitor process. MONITOR process-name the name of the monitor process to be associated with the MapDB. The name must have 4 characters. The default is $ZASY. See the note below. PRIMARY cpu-number the CPU number (in the range of 0 to 15) of the primary monitor process.
Command Interpreter DELAY If you create SYSDB on a subvolume other than $SYSTEM.ESCSYSDB, then some precautions must be taken, mainly to avoid creating more than one SYSDB, since AUTOSYNC searches all disks for the SYSDB subvolume and issues an error if more than one is found. SECURE “rwep” sets the security of the SYSDB tables. DELAY Suspends the execution of the AutoSYNC command interpreter for the specified interval.
Command Interpreter DELETE SYNCUSER * designates all configured syncfilesets for the syncuser. To display the file set ID’s and BATCHIDs, use INFO SYNCFILESET. DELETE SYNCUSER Removes synchronization authorization defined for a user. DELETE SYNCU[SER] user-id [ ,KEEP ]; user-id the Guardian user ID or user name to remove from the list of users authorized to perform synchronization. KEEP instructs AutoSYNC to retain the configuration files SFggguuu and SLggguuu associated with the user being deleted.
Command Interpreter DISABLE AUTOSYNC an Enscribe queue file, entry-sequenced file, the terminal or $NULL (or equivalent). dequeue-options See command options for COPY on page 8-56. DISABLE AUTOSYNC Instructs AutoSYNC to reject synchronizations originating from the specified system or from all systems. Use the CLEAR AUTOSYNC command to allow synchronization from the disabled source systems. CLEAR AUTOS[YNC] FROM [ \system-name | * ]; \system-name an Expand system name.
Command Interpreter FACTOR FACTOR Accepts any numeric expression and calculates the prime factors of the integer part of the expression. FACTOR constant; constant an ordinary decimal constant with or without a decimal point, or a binary, octal, or hexadecimal integer prefixed by %b, %, or %h, respectively. FC and ! Execute previous commands found in the command history. FC permits a command to be edited before execution, while ! executes a command without editing.
Command Interpreter FILES displays detailed information. If DETAIL is not specified, AUTOSYNC displays one line of information per file in the file-set. FID file-set is equivalent to FILEINFO file-set, DETAIL. FILES Displays the 8-character filename only, for the filename or filename pattern that is supplied. Note that the AUTOSYNC FILES command is not the same as the TACL or FUP FILES commands. FILES file-set; file-set TACL-style file name pattern specifying a collection of files.
Command Interpreter HISTORY is an AUTOSYNC command. Help displays the syntax and description of the command. Multi-word commands are entered with hyphens. For example, to obtain help on ADD SYNCFILESET, type: HELP add-syncfileset; If DETAIL is specified, a description of the command parameters is displayed in addition to the syntax. If EXAMPLES is specified, Help displays examples of the command. If neither DETAIL nor EXAMPLES is specified, Help displays both as possible subtopics.
Command Interpreter INFO BATCHID \system-name an Expand system name. * designates all systems that have this system as a destination. If FROM is omitted, AUTOSYNC displays the remote access control for all systems. INFO BATCHID plays the synchronization parameters for the syncfilesets in the specified BATCHID ranges. This command is analogous in functionality to INFO SYNCFILESET. INFO BATCH[ID] [ batchid-range [, batchid-range ],...
Command Interpreter INFO GLOBALS INFO GLOBALS Displays a complete list of the AutoSYNC global parameter values. INFO GLOBALS [, OBEYFORM ]; OBEYFORM displays a list of the ALTER GLOBAL commands that were entered to modify the default global values. See ALTER GLOBAL above for a list of valid AutoSYNC global parameters and values. INFO LOCALS Displays a complete list of global and local values. Global parameter values are the default values in effect for all configured monitors in the system.
Command Interpreter INFO SYNCFILESET INFO SYNCFILESET Displays the synchronization parameters for one file set or all specified file sets. INFO SYNCF[ILESET] [ fset-range [, fset-range ],...
Command Interpreter INFO SYNCUSER SYSTEM \system-name the name of the destination system for the syncfileset. This option is only required if the destination system was configured using the asterisk (*) in ADD SYNCFILESET or ALTER SYNCFILESET. Otherwise, the option is ignored. If system-name is omitted, the output of the most recent synchronization is displayed. OBEYFORM displays the text of the ADD SYNCF[ILESET] command that was used to configure the attributes for the file set.
Command Interpreter LABELDISPLAY (LD) LABELDISPLAY (LD) Displays label information for the specified table, such as catalog name, audit compression, columns data types, primary and alternate key columns, partitions, allocated extents, and so on. The command can be abbreviated to LD. L[ABEL]D[ISPLAY] sql-table [ ALLP[ARTS] | ALLI[NDEXS] ] sql-table a SQL table or View. ALLP[ARTS] displays the labels of all partitions of the table. ALLI[NDEXES] displays the labels of all indexes of the table.
Command Interpreter LIST SYNCPROCESS displays files opened for write or update access only. LIST SYNCPROCESS Obtains and displays information for all ASYSYNC and ASYSYNCP processes that are running from the current Autosync product subvolume. The synchronization processes listed are source and destination processes. The current product subvolume is the subvolume of the autosync command interpreter on the local system. This command helps to manage the AutoSYNC environment.
Command Interpreter LOG If DETAIL is omitted, the items displayed are: User ID, program file name, and process creation time. MONITOR monitor-name lists the synchronization processes that are running under a specific monitor process. If MONITOR is omitted, all synchronization processes running in the AutoSYNC product subvolume are listed. SOURCE displays the synchronization processes that are running as source synchronization processes.
Command Interpreter MONITOR If oss-pathname is omitted, the root directory is assumed. -a lists all entries in the directory, including the "." (dot) and ".." (dot dot) and other files whose names begin with a ".". -i displays inode number for each file listed. -l displays the mode, number of links, group.owner, size, time of last modification for each file, and pathname. If the file is a special file, the size field instead contains the major and minor device numbers.
Command Interpreter OBEY OBEY Reads and executes a sequence of AUTOSYNC commands from another device. The commands are executed serially until an end-of-file condition is detected. An OBEY file may not contain an OBEY command. OBEY filename; filename the file containing a sequence of AUTOSYNC commands. OPEN Select a current monitor process and associated MapDB. OPEN process-name; process-name the name of a configured monitor process.
Command Interpreter RENAME RENAME Renames a non audited Enscribe file, like the TACL RENAME command. RENAME old-file , new-file old-file the Enscribe file to rename. new-file the new name for the file. RESET Changes all of the environmental variables to their original settings.It is most often used to change the condition code. RESET; RESET GLOBAL Removes a previously set global value from the configuration and reset the parameter to the default value.
Command Interpreter RESET LOCAL RESET LOCAL Removes a previously set local value from the configuration and resets the parameter to the default value of the corresponding global parameter. RESET LOCAL { parameter | * }; parameter is: { AUTOSYNC } { [AUTO]SYNCCOMPRESS } { [AUTO]SYNCOVERIP } { [AUTO]SYNCRECOVER } { EMSCOLLECTOR } { SYNCTCPPROC[ESSES] } * all global parameters are set to their original default value.
Command Interpreter REPORT REPORT Generates a report that compares source files and destination files. For additional information on the reporting function, consult Section 9, Reporting Function.
Command Interpreter REPORT fset-id identifies a synchronization file-set ID or a range of file-set IDs. Up to 32 ranges can be entered. If fset-id - fset-id is specified, the second fset-id must be greater than the first. * designates all configured syncfilesets for the current syncuser. USER user-ID | ALLU[SERS] [, USER user-ID | ALLU[SERS] USER user-ID a configured syncuser. REPORT creates a report for the syncuser file sets that are in the fset-range specified.
Command Interpreter REPORT displays the file name, file attributes and selection criteria of each selected file. If DETAIL is omitted, the report shows only summary statistics for the file set, but it does not report on specific files. ALL short form to specify all DETAIL options described below. INSYNC displays all files from the source and destination file set that have already been synchronized and are still in sync at the time the report is generated.
Command Interpreter RUN[D] NEWERINDEST displays all files in the destination file set where the destination file is newer than the source file. It overrides the other DETAIL options INSYNC, NOTINSYNC and EXCLUDED. The option is ignored if TIMEEXACT has also been specified. The sync options are applied at the time the files are compared.
Command Interpreter START MONITOR CPU EXTSWAP LIB NOWAIT PRI DEBUG IN MEM OUT SWAP DEFMODE INSPECT NAME PFS TERM command any command line to be passed to the process in the startup message. The RUN[D] command is not a multiline command and is not terminated with a semicolon. Any semicolon is passed to the process as part of the startup command. Refer to Table 8-3 for a list of standard programs, such as EDIT and FUP, that can be executed by simply typing the program name.
Command Interpreter STATS If this option is not specified and not configured in the MapDB, the monitor process hometerm is set to the system console. The monitor process writes the messages that would normally go to its hometerm to the EMS collector $0 or the collector specified through the EMSCOLLECTOR global parameter. Also, AUTOSYNC prints a warning to this effect. See “ALTER GLOBAL” on page 8-30 for details about the EMSCOLLECTOR global.
Command Interpreter STATUS BATCHID STATUS BATCHID Obtains and displays status information from the synchronization processes configured in the specified batchid. The command is functionally equivalent to STATUS SYNC with the BATCHID option. STATUS BATCHID batchid, [, option ]; option is { ACTIVE } { BRIEF } { DETAIL } { SYSTEM system-name } { USER { user-id | user-name } | ALLUSERS } batch-id a batchid number.
Command Interpreter STATUS MONITOR STATUS MONITOR Determines the status of monitor processes by sending them an inquiry message. STATUS MONITOR [ * | process-name | mapdb ]; process-name the process name of the monitor process to be queried. The default is the current monitor process for the session. mapdb a MapDB subvolume. Since each MapDB has an associated monitor process, this option is simply an alternative method to specify the process to be queried.
Command Interpreter STATUS SYNCFILESET displays status information for ASYSYNC processes that are currently in the process of synchronizing file sets. BATCH[ID] batchid displays status information for the synchronization file sets in the specified batchid. The same information can be displayed using STATUS BATCHID. BRIEF displays the synchronization process name, the destination system, the number of active file sets in the batchid and status of the synchronization process.
Command Interpreter STATUS SYNCFILESET The command is functionally equivalent to STATUS SYNC with the FSETID option. STATUS SYNCFILESET fset-id, [, option ]; option is ACTIVE } { BRIEF } { DETAIL } { SYSTEM system-name } { USER { user-id | user-name } | ALLUSERS } fset-id a synchronization file set. ACTIVE displays status information for ASYSYNC processes that are currently in the process of synchronizing file sets.
Command Interpreter STOP MONITOR STOP MONITOR Performs an orderly shutdown of a monitor process. All scheduled synchronizations are stopped. STOP MONITOR [ process-name | mapdb ]; process-name the process name of the monitor process to be stopped. The default is the current monitor process for the session. mapdb a MapDB subvolume. Since each MapDB has an associated monitor process, this is simply an alternative method to specify the process to be stopped.
Command Interpreter SUSPEND SYNCFILESET SUSPEND SYNCFILESET Stops the synchronization for a given file set. The synchronization operations can be resumed by using the ACTIVATE SYNCFILESET command. Active synchronization is stopped after the current file set is synchronized. SUSPEND SYNCF[ILESET] fset-range [, fset-range ] ,...; fset-range is { * | fset-id | fset-id - fset-id } * | fset-id | fset-id - fset-id identifies a synchronization file-set ID or a range of file-set IDs.
Command Interpreter • • SYNCHRONIZE Files starting with ZZ and system swap files, unless ZZINCLUDE is specified Files in subvolumes $vol.ZYQ*, because they are OSS files If an eligible source file has a modification timestamp that is later than the corresponding destination file, the destination file is replaced. Timestamp comparisons account for differences in time zones and system clocks.
Command Interpreter SYNCHRONIZE Section 3, One-time Synchronization discusses synchronization options in greater detail.
Command Interpreter SYNCHRONIZE guardian-option: { BACKUP backup-program } { CATALOG SQL-catalog } { COMPRESS | NO COMPRESS } { EXCLUDE exclude-file } { KEEPNOPURGEUNTIL } { MAPNAMES mapnames-file } { MAXBACKUP num-of-file } { NETWORKS[ECURE] | NO NETWORKS[ECURE] } { NOSAFEGUARD | NO NOSAFEGUARD } { RECOVER[Y] | NO RECOVER[Y] } { RENAMEO[PEN] } { RESTORE restore-program } { SECURE "rwep" } { SQLREDEF } { SUBVOLPURGE } { USEBINDERTIME } { ZZINC[LUDE] } continued } HP Aut
Command Interpreter SYNCHRONIZE oss-option: { ALLHARDLINKS | NO ALLHARDLINKS } { ALLSYMLINKS | NO ALLSYMLINKS } { OWNERG[ROUP] { group-id | group-user } } { SECURE "rwxrwxrwx” } source-file-set a guardian or OSS file set that is to be synchronized. If a system name is specified, it must be the local Expand node name. dest-file-set the location of the file set that is to be updated.
Command Interpreter SYNCHRONIZE If a system is specified, it must be the local system; exclusion pertains only to source files, not destination files. ° If the PURGE option is specified, to exclude files from being purged on the destination, use the EXCLUDEPURGE option. Without EXCLUDEPURGE: ° A file on the exclude list is purged on the destination (it is not excluded from purging), if it does not exist on the source.
Command Interpreter SYNCHRONIZE Each entry must be on a separate line. If a line starts either with the word "COMMENT", or the characters "--", the line is treated as a comment and is ignored. Note that exclusion applies only to destination files; it does not apply to files on the source. Consequently, if a system name is specified, it must be the name of the destination system. Guardian file sets: The exclude file supports all legal disk file patterns.
Command Interpreter SYNCHRONIZE KEEPB[ACKUPINFILE] instructs AutoSYNC to retain the BACKUP and RESTORE input files. By default, these files are temporary and they are purged after the synchronization has completed. This option is used for troubleshooting only. The name of the input file is in the format of ZBKnnnnn for BACKUP and ZRSnnnnn for RESTORE, where nnnnn is a serial number. The files are placed in the MapDB subvolume. The exact file name is displayed in the LOGFILE.
Command Interpreter SYNCHRONIZE The time elapsed is computed from the last time the file was modified. The candidate files can be opened for WRITE or READ-WRITE access. The exclusion mode must be SHARED or PROTECTED. The Autosync global SYNCRECOVER parameter must be set to ON. OPENUPDATE is ignored if: • • • a file is opened for EXCLUSIVE access or, a file is audited or, the global SYNCRECOVER parameter is set to OFF. Note for OSS file sets.
Command Interpreter SYNCHRONIZE Extra SQL tables and views are purged explicitly, indexes and secondary partitions are purged implicitly, with the main table. OSS files and directories: ° If PURGE is specified, subordinate directories not present in the source file set are purged on the destination. Note that this is different than the processing of Guardian subvolumes. Guardian subvolumes are not purged unless SUBVOLPURGE is specified.
Command Interpreter SYNCHRONIZE ° #SYNCFILE is replaced with the destination file that has been created or replaced by AutoSYNC. For OSS files the absolute pathname is specified. ° #SOURCEFILE is replaced with the name of the original source file. For OSS files the absolute pathname is specified. ° #RECOVERFILE is replaced with the name of the file that was recovered by AutoSYNC. If no archive name is found, AUTOSYNC substitutes the placeholder question mark (?) for the file name.
Command Interpreter SYNCHRONIZE The default is to synchronize only non audited files. AUTOCREATECAT[ALOG] instructs AutoSYNC to create (or not to create) a SQL catalog when a SQL object that requires a catalog is moved to a destination where the catalog does not exist. The catalog is created in the same location as the catalog on the source system. If the source catalog volume does not exist on the destination system, the catalog is placed on the volume of the destination tables.
Command Interpreter SYNCHRONIZE KEEPNOPURGEUNTIL is not supported for Enscribe files, consistent with FUP and RESTORE behavior. MAPNAMES mapnames-file specifies the mapping of a source fileset to a destination fileset and catalog name mapping. The mapnames-file is an edit or entry-sequenced file located on the destination system. File name mapping: The mapnames-file entries are similar to, but not exactly the same as, the MAP NAMES option of RESTORE.
Command Interpreter SYNCHRONIZE page 3-8 and File and Catalog Name Mapping using MAPNAMES on page 3-10. MAXBACKUP num-of-files the maximum number of files to be processed by one BACKUP process. When processing large file sets, AutoSYNC divides the synchronized files into subsets, each processed by a separate BACKUP process. BACKUP performance decreases significantly with the number of files it processes.
Command Interpreter SYNCHRONIZE This option overrides the global parameter [AUTO]SYNCRECOVER. By default, recovery is enabled. RENAMEO[PEN] specifies that destination unaudited files that are opened for shared READ access or running object files are to be renamed to allow AutoSYNC to replace them. The syncuser must have PURGE privileges to the destination file for the RENAME to succeed. Both the old and new file names are recorded in an AutoSYNC index file called AAINDEX, located in the MapDB subvolume.
Command Interpreter SYNCHRONIZE By default, AutoSYNC compares only modification timestamps of source and destination SQL tables. Note. A SQL view or index is synchronized automatically with the primary partition of the base table. If the redefinition date of an index or view in the source subvolume is newer than in the destination subvolume but the primary source partition is older than the primary destination partition, the view or index will not be synchronized.
Command Interpreter SYNCHRONIZE ALLHARDLINKS | NO ALLHARDLINKS synchronizes all files that are hardlinks in the specified syncfileset, including uncontained hardlinks. When synchronizing an uncontained hard link, a new file is created in the destination file set and all contained links to the same inode are linked to this file. A warning is written into the log file when an uncontained hard link has been synchronized. If NO ALLHARDLINKS is specified, uncontained hard links are not synchronized.
Command Interpreter TIME If this option is omitted, the security of the destination file set is the same as that of the source file set. The exception is write security for the owner: when a file or a directory is synchronized and the write access was denied for the owner in the source file, write access is granted in the destination file to allow subsequent synchronizations. Write access is only upgraded when the syncuser is the owner of the file and not SUPER.SUPER.
Command Interpreter UPDATE UPDATE Modifies field values in records from an Enscribe file. Most of the parameters are identical to COPY.
Command Interpreter UPDATE file the name of the Enscribe file that is to be modified by the command. selection-options COMParelen <1-to-255 > sets the compare length for generic positioning on the record key (primary or alternate) specified in the FIRST KEY option. The compare length must be less than or equal to the key specified. COUNT num-records is the number of records or rows to be copied. If omitted, all records are copied.
Command Interpreter UPDATE ALTKEY key-value [, key-value ] the alternate key of the starting record or row. The format of key-value is described above. EXACT requests exact positioning on the record key (primary or alternate) specified in the FIRST KEY option. FROMLAST positions on the last record in the key range specified in the FIRST KEY option. REVERSE the input file is to be READ from the starting record in reverse order. SHARE the file is to be opened in shared-exclusion mode.
Command Interpreter UPGRADE MAPDB UPGRADE MAPDB Creates additional tables in a configured MapDB (that is, the MapDB subvolume and monitor process are registered in the SysDB). AUTOSYNC prompts the user to issue the command if and when necessary. UPGRADE MAPDB subvol; subvol the name of a MapDB to upgrade. VOLUME Changes the default volume and/or subvolume for filename expansion. It can be abbreviated to V. Volume [ volume | subvolume | volume.
9 Reporting Function This section describes the AutoSYNC reporting function. The topics covered are organized as follows: Overview REPORT Command Report Output Format Overview The AutoSYNC reporting function compares source and destination file sets and reports on the status of source and destination locations. It can aid in system migration by providing ad-hoc reports on migrated subvolumes, volumes or systems during the migration process.
Reporting Function REPORT Command The report can be written to a disk file. If no report file is specified, the report is displayed at the terminal.
Reporting Function Report Output Format Report Output Format The default REPORT output displays statistics of the files in sync, not in sync and excluded files; the list of potential entries generated are formatted as follows: Autosync Report timestamp Report Summary fileset: Source: source-fset, Destination: destination-fset n Files in source fileset n Subvolumes in source fileset n Volumes excluded n Subvolumes excluded n Files excluded from sync n Audited n Corrupt n Excluded specified n
Reporting Function Report Output Format If DETAIL is specified, the filename, file attributes, status indicator and the selection criteria are displayed for each file; in addition, the summary statistics are displayed at the end of the report. Filenames are grouped by subvolume. The extra files in the destination subvolume (files that are not in the source file set) are displayed first, in alphabetical order; for OSS file sets, extra files are listed last in alphabetical order.
Reporting Function Report Output Format The following example shows a sample report for a Guardian file set, where the EXCLUDE, PURGE and EXCLUDEPURGE options are specified: : AutoSYNC 1? report sync testdata.* to \dest.$data.testdata.*, purge, AutoSYNC 1+ excludep \dest.$data.jojoexcl.excludep, exclude jojoexcl.exclude; Starting SYNCfset $SDATA.testdata.* ($X1KLD) File $SDATA.testdata.RCRDEXIT excluded from synchronization. File $SDATA.testdata.RDEMP excluded from synchronization. File $SDATA.testdata.
Reporting Function Report Output Format The following example shows a sample report for a OSS file set, where the PURGE is specified: AutoSYNC 5? AutoSYNC 31? report sync /home/cssw/johanne to /E/YOSPRD/home/cssw/johanne,purge; Starting SYNCfset /home/cssw/johanne ($X0YP8) -- Report completed for fileset -- /home/cssw/johanne -- to -- /E/YOSPRD/home/cssw/johanne. --- REPORT completed successfully. Autosync Report 8/24/07 12:37:44.
A Operations and Management This appendix describes the management of AutoSYNC. The topics covered are the following: AutoSYNC Configuration Database The AutoSYNC Monitor Monitor Commands Configuring and Using an Alternate AutoSYNC Environment AutoSYNC Defines Takeover Procedures Migrating the Configuration Database AutoSYNC Configuration Database The configuration information is stored in two subvolumes: 1. SysDB: global parameters and licensing information 2.
Operations and Management MapDB MapDB MapDB stores configuration parameters for scheduled synchronizations. The tables in the MapDB subvolume are: • • • • A REGISTRY table to store local parameters. Local parameters are analogous to global parameters stores in the SysDB REGISTRY table, but they apply only to a specific MapDB/monitor pair.
Operations and Management • The AutoSYNC Monitor In FUP, rename the current log file and create the new file: RENAME SLggguuu, OLDLOG SET LIKE OLDLOG SET EXT (primary-extent, secondary-extent) MAXEXTENTS maxextents CREATE SLggguuu EXIT • In AutoSYNC CI, reactivate the synchronizations for the syncuser: ACTIVATE SYNCUSER userID; The log file is opened once, when the synchronization process starts up. SUSPEND stops the synchronization process and ACTIVATE restarts the synchronization process.
Operations and Management Configuring the Monitor Process Configuring the Monitor Process The monitor is configured and started when the MapDB is created through the CREATE MAPDB command. The monitor has an assigned default process name, $ZASY, reserved for AutoSYNC. The monitor configuration can be changed by using the ALTER MONITOR command. You can configure process priority, process CPUs, and the monitor event-logging device.
Operations and Management Stopping a Monitor Stopping a Monitor The monitor process should be available for scheduled synchronizations to occur. If the monitor is stopped, synchronizations in progress will continue, but none are started. In normal situations, however, the monitor process should be stopped only when the system is being shut down. Use the STOP MONITOR command to stop the monitor. Monitor Commands The monitor accepts interactive command requests from AUTOSYNC.
Operations and Management LOG LOG The LOG command starts (and stops) the logging of monitor activity. LOG { TO file [CLEAR] | STOP } TO file [CLEAR] the name of a log file. CLEAR empties the file before logging commences. If file already exists, information is appended to the end of the file unless CLEAR is specified. If the file name ends with a number (for example, LOG001), additional files will be allocated with incrementing numbers when the first file becomes full. STOP stops logging.
Operations and Management SECURITY LICENSE requests information about the AutoSYNC license for the current system. TCPIP requests information about the configured listener processes. SECURITY The SECURITY command instructs the monitor to accept further commands only from a specific set of process access ids. SECURITY { N | C | U | A | G | O | - } The command option is a single character and is a Guardian security specifier.
Operations and Management Installing an Alternate Version of AutoSYNC AutoSYNC can use the same SysDB concurrently. See Configuring an Alternate SysDB below for details. Installing an Alternate Version of AutoSYNC If you wish to run different versions of AutoSYNC on your system, you first must install the alternate version in its own product subvolume. Run the INSTALL macro from the installation subvolume that contains the desired version, specifying the alternate product subvolume of your choice.
Operations and Management Configuring an Alternate Monitor Process Create an alternate MapDB Create the alternate MapDB using the CREATE MAPDB command, specifying the alternate monitor process name. This command starts the monitor process. In the example below, the user creates a test MapDB in $DATA.ASZMAPDB with monitor $ZASZ, from the same alternate product subvolume $data.asyaakts created as an example above in Installing an Alternate Version of AutoSYNC. 1> $data.asyaakts.
Operations and Management Configuring an Alternate Monitor Process In the example below, the user wishes to disable synchronization over TCP/IP for the new environment, for testing purposes and does so by setting the local SYNCoverip parameter: AutoSYNC 2? info globals; AutoSYNC Global Settings AutoSYNC SYNCcompress SYNCrecover SYNCoverIP SYNCTCPPROCesses EMSCollector ON OFF ON ON $ZTC0,$ZTC1 $0 AutoSYNC 3? alter local syncoverip off; --- AutoSYNC Local AUTOSYNCOVERIP updated.
Operations and Management • Configuring an Alternate SysDB Specify the monitor process name on the AUTOSYNC run command: 3> $data.asyaakts.autosync $ZASZ HP Nonstop(tm) AutoSYNC(tm) Command Interpreter(T2713H01) - System \ATOM (C)2006 Hewlett Packard Development Company, L.P. (C)2006 Carr Scott Software Incorporated • In AUTOSYNC, use the OPEN monitor-process command. 4> $data.asyaakts.
Operations and Management Configuring an Alternate SysDB Add the SysDB define Use the =_AUTOSYNC_SYSDB class catalog define to specify an alternate SysDB subvolume. Do not use the ESCSYSDB subvolume name. 1> add define =_autosync_sysdb,class catalog,subvol $data1.mysysdb 2> autosync HP Nonstop(tm) AutoSYNC(tm) Command Interpreter(T2713H01) - System \ATOM (C)2006 Hewlett Packard Development Company, L.P. (C)2006 Carr Scott Software Incorporated *Warning* The SysDB cannot be found.
Operations and Management AutoSYNC Defines define is the only way to specify an alternate SysDB. Use the ENV command to verify which SysDB and MapDB you are currently accessing. Autosync 1? env; ---------------------------------------CI Session Environment ---------------------------------------Process: $Z6LD (7,163) (priority 145) Volume $RWC.AUTOSYNC Dictionary None SQL catalog None SysDB $DATA1.MYSYSDB (9607180a) MapDB $DATA.
Operations and Management Takeover Procedures The =_AUTOSYNC_SYSDB DEFINE is class CATALOG and specifies a subvolume for the SysDB tables. If this DEFINE is present when AUTOSYNC is started, it does not look for the normal SysDB, but uses the one specified. Thus, it operates only with MapDBs and monitor processes that are created under the specified SysDB. A typical =_AUTOSYNC_SYSDB DEFINE is added by using the following TACL command: ADD DEFINE =_AUTOSYNC_SYSDB,CLASS CATALOG,SUBVOL $vol.
Operations and Management Migrating the Configuration Database followed, the old primary system will immediately start synchronizing possibly out-ofdate files to the new primary system. To prevent older files on the original primary system from replacing newer files on the current primary system, the user must disable synchronization from the old primary system using the DISABLE AUTOSYNC command:.
Operations and Management Migrating an AutoSYNC Source System On the New Source System 1. Install AutoSYNC as described in Installing AutoSYNC on page 2-1. When creating the new MapDB, use the INFO MAPDB output collected in infom as a reference. 2. Authorize Autosync, using the STATUS MONITOR output collected in infom as a reference. 3.
Operations and Management • • Migration Steps for an AutoSYNC Destination System Validate the new configuration using the INFO GLOBAL, INFO SYNCUSER and INFO SYNCFILESET commands. Proceed with your usual testing suites, using one time syncs, the REPORT command and activating syncfilesets progressively. Migration Steps for an AutoSYNC Destination System On the Existing Source System Moving a destination system requires updates to the syncfileset definitions on the source system. 1.
Operations and Management Migration Steps for an AutoSYNC Destination System HP AutoSYNC User’s Guide—522580-016 A-18
B Problem Resolution This appendix discusses how to diagnose AutoSYNC problems, how to use the tools and how to report problems. The topics covered are: Common Security Problems Troubleshooting Tools Submitting Problems Common Security Problems Security related problems are common when first configuring AutoSYNC and performing synchronizations. Security rules enforced by SafeGuard and Guardian must be taken into consideration when planning and installing AutoSYNC.
Problem Resolution Scheduled Synchronization Problems Scheduled Synchronization Problems The source synchronization process creation fails with security violation • • Verify that the ASYSYNCP object file in the source product subvolume is PROGID’ed to the Authorizer. Verify that the Authorizer has permission to log down to the sync user ID. If not, make ASYSYNCP callable, to bypass the restriction. See Bypassing Disabled Log Down on page 4-7 for details.
Problem Resolution Errors executing triggers Synchronization of OSS files fails with error 4001 • Check the owner of the source file. If the user is not the owner of the source file, only SUPER.SUPER can synchronize the file or directory. Errors executing triggers • • Verify that the user has sufficient privileges to execute the trigger commands on the destination. Before configuring a trigger, log on as the user on the source system and run the trigger commands from a TACL prompt.
Problem Resolution Troubleshooting Tools Troubleshooting Tools AUTOSYNC Commands Synchronization Logs Events in the EMS Log AutoSYNC Tracing AUTOSYNC Commands Checking the Environment The overall processing environment can be checked using the CI command ENV. It displays the names of the current MapDB and SysDB, along with the Guardian OS RVU, the serial number of the NonStop system, and so on. Checking the Configuration To list the values of system wide parameters, use INFO GLOBALS.
Problem Resolution • ° ° ° ° ° Synchronization Logs d— ASYSYNC-D - destination system t— trigger command output b— BACKUP r— RESTORE p— Report output line A text message If the synchronization is over EXPAND, both source and destination synchronization processes open the log file on the source system and write their respective messages into the log file. If the synchronization is over TCP/IP, only the source synchronization process opens the log file.
Problem Resolution Synchronization Logs Autosync 1? info syncf 2,log last 1; Autosync User: 100,1 (sysman.user1) ID Source fileset Destination fileset Batch Intvl --- ------------------------ --------------------------------- ----- ----2 $DATA.STUFF* \SYSB.$DATA.STUFF.* 0 2 Log Starting Time: 2001-02-09 14:46:53.
Problem Resolution Synchronization Logs One-Time Synchronizations One-time synchronizations are launched using the SYNCHRONIZE command. AUTOSYNC logs the output of a such synchronizations in a temporary log file that is purged after the synchronization has completed. To retain the log file, use the LOGFILE option of the SYNCHRONIZE command. The format and content of the log file are identical to the SLggguuu log for scheduled synchronizations.
Problem Resolution Events in the EMS Log Events in the EMS Log The AutoSYNC monitor and the synchronization processes report informational and critical events to the EMS log on the source system. The Monitor events are sent to the Monitor’s home terminal. To direct all AutoSYNC events (from ASYSYNC processes and the Monitor) to the same collector, configure the Monitor’s home terminal to the local collector.
Problem Resolution AutoSYNC Tracing In both one time and scheduled synchronizations, the output of the trace is written to the synchronization log file. Every operation performed by the sending and receiving ASYSYNC process is recorded in the trace.
Problem Resolution Submitting Problems The following example traces a synchronization attempt where no files require synchronization. Autosync 3? info syncf 1,log last 1; Autosync User: 100,1 (SYSMAN.USER1) ID Source fileset Destination fileset Batch Intvl ---- ------------------------- --------------------------------- ----- --1 $DATA2.STUFF.* \SYSB.$DATA2.STUFF.* "JC" 1 Log Starting Time: 2001-02-09 16:39:11.373 s-PROCESS_CREATE_(\SYSB.$DATA2.JCC153.ASYSYNC,\SYSB.$Y2769:24726817) s-FILE_OPEN_(\SYSB.
Problem Resolution AutoSYNC Component Failures AutoSYNC Component Failures When the monitor process, AUTOSYNCCI or a synchronization process abend, they produce a saveabend file. The monitor displays a message on its hometerm or the configured EMS collector process and the command interpreter displays a message on its out file.
Problem Resolution Scheduled Synchronization Failures ° The log of the synchronization, either by PAKíng the log file (SLnnnggg) or by sending the output of the INFO SYNCF... LOG option ° A VPROC listing of the ASYSYNC(P) process When the log does not yield sufficient diagnostic information, trace the syncfileset and send the trace along with the remaining supporting data.
C Examples These examples are intended to assist you in becoming acquainted with the use of the AutoSYNC product. These examples assume that the product has been successfully installed, licensed and that the monitor process is running. One-Time Synchronization Guardian file set SYNCHRONIZE the source file set $DATA.JOJOX*.* to the destination file set \DEST.$DATA.*.
Examples OSS file set OSS file set To synchronize the local directory tree home/dev/jojo to the destination node NSK2:i AutoSYNC 10? sync /home/dev/jojo to /E/NSK2/home/dev/jojo, timeexact; Starting SYNCfset /home/dev/jojo Source Directory: /home/dev/jojo/ ( .
Examples Scheduled Synchronization Scheduled Synchronization Single-User Environment In the single-user AutoSYNC environment, you install and configure AutoSYNC. The following commands show how to configure a single-user scheduled synchronization: Autosync 1? ADD SYNCUSER 110,125, MAXPRIORITY 130; --------- MapDB table $F08.ASYMAPDB.SF110125 created. MapDB table $F08.ASYMAPDB.SL110125 created. AutoSYNC user 110,125 (OPER.JOE) added. Monitor $ZASY cache refreshed.
Examples Multiple-User Environment Autosync 8? DELETE SYNCFileset 1; --- AutoSYNC fileset ID 1 deleted. --- Monitor $ZASY cache refreshed. Autosync 9? INFO SYNCFileset; AutoSYNC User: 110,125 (OPER.JOE) ID Source fileset Destination fileset Batch Intvl ---- ------------------------- --------------------------------- ----- ----0 $DATA.GROUP0.* \PROD.$DATA.GROUP0.* 0 5 Autosync 10? STATUS SYNC; SYNCuser: 110,255 (OPER.
Examples Multiple-User Environment configure a multiple-user environment where SUPER.SUPER is the AUTHORIZER. After logging on as SUPER.SUPER: Autosync 1? AUTHORIZE AUTOSYNC, ADMINistrator 110,255; User 110,255 (OPER.MGR) has been authorized as AutoSYNC administrator. The AutoSYNC monitor process must be started by user 255,255 (SUPER.SUPER). --- Monitor $ZASY cache refreshed. Log on as OPER.MGR: Autosync 2? ADD SYNCUSER OPER.JOE, PRIORITY 140; --------- MapDB table $F08.ASYMAPDB.SF110125 created.
Examples Multiple-User Environment Autosync 7? INFO SYNCFileset 0, DETAIL; AutoSYNC User: 110,125 (OPER.JOE) SyncID: 0 Source fileset: $DATA.GROUP1.* Destination fileset: \PROD.$DATA.GROUP1.
D Error Messages This appendix describes the EMS messages generated by AutoSYNC.
Error Messages Informational messages Effect. The monitor stops. Recovery. Restart the monitor. 103 Monitor takeover Cause. The monitor primary process or its CPU has failed. Effect. The monitor backup process has taken over. The monitor is ready to service requests. Recovery. Informational message. No action is required. 104 Monitor backup created Cause. The monitor primary process has created a backup process. Effect. The monitor process is fault tolerant. Recovery. Informational message.
Error Messages Informational messages 107 Monitor status info Cause. The monitor process has displayed status information. Effect. Status information is placed in the EMS log. Recovery. Informational message only; no corrective action is needed. 108 Monitor started ASYSYNC process Cause. The monitor process has started an ASYSYNC process to perform AutoSYNC synchronization. The message contains the user id that configured the synchronization, the destination node, and the process name. Effect.
Error Messages Informational messages 111 ASYSYNC process lost Cause. An ASYSYNC process has stopped or abended, but no process-stop message was received. This may be caused by a CPU failure. Effect. Informational message only; no corrective action is needed. Recovery. No action is required. The ASYSYNC process is restarted automatically. 112 Monitor cache refreshed Cause. Cached configuration information has been refreshed from the MapDB tables. Effect.
Error Messages Informational messages 115 ASYSYNC process failed Cause. The monitor process has attempted to start an ASYSYNC process to perform AutoSYNC synchronization, but the operation failed. The message contains PROCESS_CREATE_ error and error detail code. Effect. AutoSYNC processing for the specified user id is not active. Recovery. Correct the cause of the failure and the process will be restarted automatically. 120 ASYSYNC process started via TCP/IP Cause.
Error Messages Informational messages 123 Trigger TACL or OSS shell process abended Cause. An ASYSYNC process performed the TRIGGER option; it started a TACL process or an OSS shell passing the trigger command. TACL or the OSS shell abended. The message contains the user ID, destination node name, BATCHID, the file-set ID, or a -1 if not available and an the error text or the error and error detail code. Effect. Effect.
Error Messages Informational messages 143 MAPDB inconsistent Cause. Configuration information in the MapDB tables is inconsistent. Effect. The configuration information is not available on request to the monitor. Recovery. Correct the configuration information and issue the MONITOR REFRESH NOW command. 200 ASYSYNC log access error Cause. An ASYSYNC process detected an error accessing a SYNC user logfile. Effect. The error number and the nature of the problem is described in the message. Recovery.
Error Messages Critical event messages Critical event messages 402 Monitor intentionally abended Cause. The monitor process has been intentionally abended to produce a saveabend file for problem diagnosis. Effect. The primary monitor process abends. The backup process takes over and continues processing. Recovery. Send the saveabend file to product support. 403 Invalid license Cause. An operation was attempted, but no license for that service has been installed. Effect. The operation is rejected.
Error Messages Critical event messages 413 Memory pool allocation error Cause. An attempt to allocate memory in the extended segment failed. Effect. The process abends. Recovery. Since the product usually resizes the segment to satisfy memory requests, this error should not occur. Report this error to product support. 414 Allocate segment error Cause. Allocation of the extended segment failed. Effect. The process abends. Recovery. The error number is provided in the message.
Error Messages Critical event messages 432 ASYSYNC version mismatch Cause. A source or destination ASYSYNC process detected a version mismatch. The process that failed and the starting process, either the AutoSYNC monitor or the source ASYSYNC process if the destination ASYSYNC process failed, have different versions. Effect. The ASYSYNC process that detected the version mismatch abends. Recovery. Install the same AutoSYNC version on the source and destination systems.
Index A Administrator 4-2 Alternate key files 3-6 ASYSYNC or ASYSYNCP privileged processes 2-8 ASYSYNCP process authorizing 2-7 Authorizer 4-2 AUTOSYNC command interpreter 8-1 AutoSYNC license 2-5, 3-1 SysDB subvolume A-1 Exact modification time stamp 3-12 Excluded Files 3-13 Excluded Files and Purge EXCLUDEPURGE 3-13 Excluded subvolumes and Purging EXCLUDEPURGE 3-13 F Files Opened for Update 3-14 B L BATCHID 4-5 License and Progid attributes 2-1 Preserving when synchronizing 2-8, 3-6 Log down 2-2 C
Index O O Open Destination Files 3-14 Operational product files 2-2 OSS Destination file set 5-2 empty directory 5-2 hard links 5-3 pathname length 5-1 purging extra destination files 5-3 Source file set 5-2 Symbolic link 5-3 ZASYnnnn name extension 5-2 OSS pathname /E directory 5-1 Ownership Changing in symbolic links 5-7 OSS file sets 5-5 P Partitioned files and tables 3-8 Purging Destination Subvolumes SUBVOLPURGE 3-13 Purging Files in Destination 3-13 Purging Subvolumes in Destination 3-13 R REGISTR
Index Special Characters ZYQ*.
Index Special Characters HP AutoSYNC User’s Guide—522580-016 Index-4
Content Feedback First Name: __________________ Phone: _____________________ Company: ___________________ Last Name: _________________ e-mail address: ______________ (All contact information fields are required.) If you’re reporting an error or omission, is your issue: Minor: I can continue to work, but eventual resolution is requested. Major: I can continue to work, but prompt resolution is requested. Critical: I cannot continue to work without immediate response.
