SMS activation - RACF AUDIT
Hi there I am trying to set up auditing of various storage management tasks. Within RACF I have set various dataset profiles with AUDIT SUCESS(UPDATE) so that they are logged as SMF/RACF type 80s - this works fine. I then tried to do something similar whenever someone activated a new SMS config. But that doesn't seem to work. I have updated the RACF FACILITY STGADMIN.IGD.ACTIVATE.CONFIGURATION so that it has AUDIT(ALL). When I activate a new SMS config I can see the entry in the SMF - type 43 subtype 3. But when I then run the program to list the RACF records it isn't there. This is the job I am running. So I can see the type 42/3 record in SMFDATA but no equivalent entry in the OUTDD dataset. I am a Storage/Sysprog person - only a little knowledge of RACF. Thanks //STEP0001 EXEC PGM=IFASMFDP //SYSPRINT DD SYSOUT=* //ADUPRINT DD SYSOUT=* //OUTDDDD DISP=(,CATLG),DSN=BD.SL780A.SMF.SQ.TEMPE, // SPACE=(CYL,(1500,250),RLSE),DSORG=PS,LRECL=32756,REC // UNIT=(DISK,3) //SMFDATA DD DISP=SHR,DSN=BD.MOSMFZS.SMFGEN //SMFOUT DD DUMMY //SYSINDD * INDD(SMFDATA,OPTIONS(DUMP)) OUTDD(SMFOUT,TYPE(0:255)) ABEND(NORETRY) USER2(IRRADU00) USER3(IRRADU86) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: SMS activation - RACF AUDIT
Hi there Thanks for the tip. It looks like it is the other way around. If you activate via ISMF then it doesn't get logged into RACF type 80. But if I issue SETSMS SCDS (dsname) it does get picked up. I think I will raise a question with IBM. Thanks -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
VTOC audit trail - init of new dasd volumes
Hello Does anyone know how I could track dasd volumes being initialised? This would be for audit purposes. I can look at DCOLLECT info and see live SYS1.VTOCIX.volser datasets. This only looks at dasd available when the job was run. This shows when they were created but not what they were called before. I can look at SMF data to show when the ICKDSF program was run. This gives a job name but not the userid. And doesn't show any volser details. I would like to find out when the ICKDSF program was run and the volser name before and afterwards. I could scan the syslogs but this only show the UCB address not the volsers. Does anyone have any ideas? Thanks -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Dataset FLASH/SNAP question
We have a project which wants to copy about 1tb of data from prod to dev. All the data will be in the same EMC DMX. All the data will be in the same sysplex/SMSplex. The datasets will be renamed when copied. The data is DB2 and IMS - so lots of little files. The process will be repeated every 1-2 months. Minimal outage to the source (prod) is required.. We have looked at EMC Timefinder SNAP at dataset level but there is a bug whereby it seems to randomly rename the target dataset. Problem is with EMC but little progress. We also have FDRABR and FDRinstant (from Innovation) which we currently use for our backups. We are looking at FDRCOPY and it seems similar to Timefinder SNAP but it works. Does anyone else do something similar? Any problems? Things to look out for? Thanks in advance -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: EMC Compatible Peer Features
Hello We use GDPS and EMC DMX dasd. We did use IBM ESS but replaced with EMC dasd. Works fine. There was very little we had to do to change our DR procs. GDPS works much the same whatever the dasd. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: EMC Compatible Peer Features
EMC uses SRDF in PPRC emulation mode. We also use Flashcopy emulation via GDPS too -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How do you do a standalone restore in a virtual tape enviroment
yes - virtual tapes wil be duplexed. Question is how to get the starter/driving system up. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How do you do a standalone restore in a virtual tape enviroment
Got the answer now. Thanks. see page 324 of the TS7740 Redbook. http://www.redbooks.ibm.com/abstracts/sg247312.html?Open Standalone Mount This allows defined users with the correct authority to instruct a VTS to manually mount a virtual volume on a virtual drive. This function enables you to mount a logical volume on a selected virtual drive. If you have an operating system image written to a logical volume, performing a standalone mount of a logical volume to a virtual drive can allow you to boot from that mounted logical volume. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
How do you do a standalone restore in a virtual tape enviroment
We currently have a tape environment which is a mixture of virtual and non- virtual tape. We are soon to be going to a complete virtual tape environment. In the unlikely event that we might loose all our data we have a process where we could restore from our tape backups. But how could we do this in a virtual tape environment? There are 4 sets of tapes. At the moment sets 1,2,3 are all 3590 non-virtual tapes. A list of the tapes are stored in a word doc (for tape sets 2 and 3). Tape 1 set - standalone IPL volume. Created yearly. Stored in a drawer. Tape 2 set - backup of a a mini (starter) MVS system (3*model9s) which we could IPL. On this starter system are some sample jobs to do the restores etc. Created weekly. Tape 3 set - back up of FDR catalog, TLMS/VMF backup, jobs to init dasd and restore from backups. Created daily. A list of the tape volsers is stored off-mainframe in a word document. Tape 4 set - FDR backups Created daily Assume we have no MVS systems. We would use standalone IPL (tape set 1) to restore the starter system (tape set 2). We would then IPL starter system and restore data on tape set 3 as this would only be a few tapes. The restored data from tape set 3 would the be used to reinit the dasd volumes and do FDR restores from tape backups. I know that we could replace the standalone IPL on a CD instead of a tape. We could also FTP the data from tape set 3 to an off-mainframe server. But how could we create/read the starter tape in virtual envionment? We have a DR site and we PPRC our data cross site and have procedures for that recovery scenario. But this process is to cover the unlikely event that all the data becomes corrupt at both sites and we have to restore from tape. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
HSM secondary space management - order?
Does anyone know how HSM determines the order to move eligible datasets from ML1 to ML2? We have HSM secondary space management set up to run between (0900 1600) and it usually runs out of time. However when you look at the MIG logs it seems to process the datasets in alphabetical order eg. ARC0734I ACTION=MIGRATE FRVOL=ML1087 TOVOL=A06490 TRACKS= 1 DSN=AA.UNITSUPP.VP.UPLOAD.OSELECT.GDG.G0001V00 Some of our bigger datasets seem to be at the end of the alphabet and so never seem to get processed by secondary space management. Is there a way to change it so that it processes the bigger datasets first? Thanks -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Identifying SASDBs in SMS ACS routines
I am trying to write some SMS ACS code to identify SAS databases. SASDBs seem to have these characteristics - DSORG=PS and RECFM=FS. I want to set up a separate SMS Storage group for SASDBs but I can't seem to find a way to code something in the ACS routines. I can do a check for DSORG=PS but there doesn't seem anything for the RECFM. Any ideas? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
