Link list is one of the most commonly misunderstood z/OS features. When link list is built at IPL, MVS uses the specified list of libraries but does not subsequently access programs via those DSNs. Instead, a table of disk extents is created that maps the named libraries across all the volumes represented. Individual libraries can be renamed, deleted, expanded or contracted, whatever. Link list knows nothing of these changes. He still plods along steadfastly with the original extent list: beginning and ending track location of each library *at IPL* regardless of what data may or may not be there later on. The main reason for the XCFAS enqueue on link list libraries is to prevent accidental deleting or otherwise mangling of a library or its contents.
For decades the only way to change link list was to IPL. Now we have dynamic link list rebuild via SETPROG, which directs z/OS to create a new link list based on *current* library locations and contents. This may encompass resized or relocated libraries, new libraries, or libraries removed from link list. It's important to understand that merely changing a link list library is some way is pointless without a corresponding SETPROG. S106 abend most often results from some unexpected change to a link list library. FETCH tries to read a load module from the expected location but gets an I/O error because the module is no longer there or has been pushed into a new disk extent that link list does not know about. It's possible to nullify the XCFAS enqueue, but you really have to know what you're doing. It's useful mainly if you have to manipulate a library that has the same name as a link list library but is not physically the same one. That situation is rare. Any other use is a recipe for disaster. . . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile [email protected] > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Lizette Koehler > Sent: Saturday, February 20, 2016 06:42 AM > To: [email protected] > Subject: [Bulk] Re: DASD device not going offline > > Is this a production or dev/test environment? > > If prod, I would go through proper change control > > But no, you should not have to IPL. The S106 may indicate your APF is not > pointing to the correct volumes now. > > > > What has happened is the LINKLST or APF LIST does not match the catalog. > So you need to do > > D PROG,LINKLIST > > See what volumes are pointed to by the datasets. > > D PROG,APF > See what volumes are pointed to by the datasets. > > > Next check your definitions in the CSVLLAxx and PROGxx members. If needed, > create a new CSVLLAxx member and PROGxx member that points those files to > the correct volumes. If everything is correct, you should be able to correct > your issue. > > > For the PROGxx member, here is a sample. You need to create it for your > needs. SAMPLE ONLY, you may or may not need everything in this member. > > /* */ > /* Sample PROGxx member */ > /* */ > /* */ > LNKLST DEFINE NAME(SAMPLLA) COPYFROM(CURRENT) > LNKLST ADD NAME(SAMPLLA) DSNAME(SYS1.MYLINKLB), > AFTER(SYS1.AFTER.LIB) > LNKLST DELETE NAME(SAMPLLA) DSNAME(SYS1.LINKLB.OLD) > LNKLST ACTIVATE NAME(SAMPLLA) > LNKLST UPDATE JOB(*) > APF ADD DSNAME(SYS1.MYLINKLB) VOLUME(&SYSR3) > > In this sample, I am creating a new LINKLST named SAMPLLA, I am adding a > load lib called SYS1.MYLINKLB after the dataset SYS1.AFTER.LIB. > I am deleting the entry SYS1.LINKLB.OLD. > I then activate the new SAMPLLA linklist I tell all jobs to be updated to the > new > Linklst And I add an APF For SYS1.MYLINKLB for indirect on SYSRES volume 3. > > You may or may not need all of these actions. > > Notice, I do not shut-down and restart LLA in this process. I do not > unallocated the Linklst in this process. I only build the new Linklst and > activate > it. > > > But you need to slowly go through the APF and LINKLST and make sure the > datasets and volumes are now matching what you need. There is still > something not quite correct between the two. > > Lizette > > > > -----Original Message----- > > From: IBM Mainframe Discussion List [mailto:[email protected]] > > On Behalf Of Mainframe Mainframe > > Sent: Saturday, February 20, 2016 7:17 AM > > To: [email protected] > > Subject: Re: DASD device not going offline > > > > I think because, these dataset were in diff volume and after making > > indriect catalog, updated catalog entry, we getting this error. > > > > Now I tried updating linklist dynamically Issue the following MVS > commands: > > SETPROG LNKLST,UNALLOCATE > > P LLA > > SET PROG=09 > > S LLA,SUB=MSTR > > SETPROG LNKLST,ALLOCATE > > > > But after issuing this SETPROG LNKLST,ALLOCATE command, I am getting > > below error. > > > > CSV540E LNKLST SET LNKLST00 is in error. > > DATASET SET SYS1.SERBLINK > > HAS VOLUME ID DOESNT NOT MATCH WITH CATALOG. > > > > Now do I need to IPL system, So that LNKLIST can look at dataset in > > correct volume and catalog or is there any way we can do this dynamically. > > > > On Sat, Feb 20, 2016 at 7:15 PM, Mainframe Mainframe < > > [email protected]> wrote: > > > > > These below 4 datasets were directly was pointing into primary res > > > volume, So I changed back to point to indirect catalog. > > > SYS1.SCSFMOD0 > > > SYS1.SISFLOAD > > > SYS1.SASMMOD1 > > > SYS1.SERBLINK > > > > > > > > > > > > On Sat, Feb 20, 2016 at 6:28 PM, Eric Mendelson < > > > [email protected]> wrote: > > > > > >> Did you bring in a new link list. Looks like a dataset is missing > > >> or one dataset went into secondary extents > > >> > > >> Sent from my iPhone > > >> > > >> > On Feb 20, 2016, at 7:52 AM, Mainframe Mainframe < > > >> [email protected]> wrote: > > >> > > > >> > Thanks for help . > > >> > I think I found the problem, it was related to indirect logging. > > >> > I made necessary changes and changed dataset to indirect catalog > > >> > but now I am > > >> not > > >> > able to access SDSF, I am getting below error while accessing panel. > > >> > > > >> > ISPF Dialog Error Command ===> > > >> > > > >> > > > >> > > > **************************************************************** > ****** > > ******** > > >> > * ISPD212 > > >> > * > > >> > * > > >> > * > > >> > * Invalid PGM name > > >> > * > > >> > * Link to 'ISFISP' failed, abend code = x'00000106'. > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * Current dialog statement: > > >> > * > > >> > * SELECT PGM(ISFISP) PARM() NEWAPPL(ISF) PASSLIB NOCHECK > > >> > SCRNAME(SDSF) > > >> > * > > >> > * > > >> > * > > >> > * Enter HELP command for further information regarding this error. > > >> > * > > >> > * Press ENTER key to terminate the dialog. > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > * > > >> > > > >> > > > >> > > > >> > On Sat, Feb 20, 2016 at 5:28 PM, Steve Horein > > >> > <[email protected]> > > >> > wrote: > > >> > > > >> >> The presence of XCFAS looks indicative of a LNKLST'ed dataset on > > >> >> the > > >> the > > >> >> volume. > > >> >> This may help: > > >> >> > > >> >> > > >> https://www- > > > 01.ibm.com/support/knowledgecenter/SSLTBW_2.1.0/com.ibm.zos.v2r1.ieae2 > > 00/remen > > q.htm > > >> >> > > >> >> > > >> >> On Sat, Feb 20, 2016 at 5:47 AM, Mainframe Mainframe < > > >> >> [email protected]> wrote: > > >> >> > > >> >>> Hello Steve, > > >> >>> result is > > >> >>> D U,DASD,ALLOC,376B,1 > > >> >>> IEE106I 04.48.06 UNITS ALLOCATED 730 > > >> >>> UNIT JOBNAME ASID JOBNAME ASID JOBNAME ASID JOBNAME > ASID > > >> >>> 376B *MASTER* 0000 XCFAS 0006 > > >> >>> > > >> >>> > > >> >>> > > >> >>> On Sat, Feb 20, 2016 at 5:10 PM, Steve Horein > > >> >>> <[email protected] > > >> > > > >> >>> wrote: > > >> >>> > > >> >>>> What is the result of D U,DASD,ALLOC,<dasd ucb>,1 > > >> >>>> > > >> >>>> On Sat, Feb 20, 2016 at 5:06 AM, Mainframe Mainframe < > > >> >>>> [email protected]> wrote: > > >> >>>> > > >> >>>>> Hello, > > >> >>>>> One of my dasd device not going offline. I tried all > > >> >>>>> possible > > >> >>> way > > >> >>>>> but still not success. I am not able to find, who is holding > > >> >>>>> this > > >> >>> device. > > >> >>>>> We are in Sysplex env and from other system, this device is > > >> >>>>> offline. > > >> >>>>> > > >> >>>>> 1) I am able to make offline this device from all other > > >> >>>>> system > > >> except > > >> >>>> from > > >> >>>>> the system where it is required to be used. > > >> >>>>> > > >> >>>>> 2) I tried checking link to CHPID as well but I did see any > > >> >>>>> path > > >> >> going > > >> >>>> bad. > > >> >>>>> > > >> >>>>> 3) I also check conjunction using D GRS,C but nothing i could find > > >> >>>>> . > > >> >>>>> > > >> >>>>> > > >> >>>>> Finally when I try forcing device making offline using force > > >> command, > > >> >>>> this > > >> >>>>> goes in boxed status. > > >> >>>>> Can you please help, if any other way to find reason for this > > >> >>>>> and > > >> >> solve > > >> >>>>> this issue. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
