Re: Adding a second CSS
Hi Greg, Many thanks for all the info! Yep, dynamic I/O is pretty cool - no more waiting for a POR (something which was never a problem for us in the past as we ran in BASIC mode on our own box, but now we are on an LPAR sharing a big z10 with some other groups, so we need to get concurrence from EVERYBODY to get downtime). Have a great weekend and Thanks again! -Mike -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Friday, March 19, 2010 10:34 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Adding a second CSS Grrreat! They are stored in HSA and will survive an IPL but without a couple more steps not a POR. To make them permanent, 1.) update the IOCP 2.) write out a slot, one not in use or the next slot, according to local traditions eg 'IOCP iocpFN ( ICP DYN WRTAn' <-- were An is the slot you want to write the IOCP into **Note** ICP is machine specific/ z9 2094 and the old LPAR parameter isn't needed on z9's 3) then point the SE at the slot to be active 'CP SET IOCDS_ACTIVE An' You don't need to do a POR but should one happen, and if the HMC is set to "use the active slot" or you update the 'reset profile', you'll be covered again, all as local traditions, change/risk mgt/etc dictate. Enjoy! I can barely think of anything I enjoy doing more then DynIOs... at work that is.. X-<8^) On Fri, Mar 19, 2010 at 8:22 AM, Michael Coffin wrote: > Now, my last (hopefully!) question is this. Are these modifications stored > in the HSA, and/or will they survive an IPL of z/VM in the TEST LPAR? I > realize we'll still need a POR to make these permanent (absent using HCD, > which we haven't deployed yet) - but for the time being are these changes > "permanent" unless/until a POR is performed, or will they disappear when the > z/VM TEST LPAR is IPL'd? > > Thanks for your help Greg, MUCH appreciated. :) > > -Mike -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Grrreat! They are stored in HSA and will survive an IPL but without a couple more steps not a POR. To make them permanent, 1.) update the IOCP 2.) write out a slot, one not in use or the next slot, according to local traditions eg 'IOCP iocpFN ( ICP DYN WRTAn' <-- were An is the slot you want to write the IOCP into **Note** ICP is machine specific/ z9 2094 and the old LPAR parameter isn't needed on z9's 3) then point the SE at the slot to be active 'CP SET IOCDS_ACTIVE An' You don't need to do a POR but should one happen, and if the HMC is set to "use the active slot" or you update the 'reset profile', you'll be covered again, all as local traditions, change/risk mgt/etc dictate. Enjoy! I can barely think of anything I enjoy doing more then DynIOs... at work that is.. X-<8^) On Fri, Mar 19, 2010 at 8:22 AM, Michael Coffin wrote: > Now, my last (hopefully!) question is this. Are these modifications stored > in the HSA, and/or will they survive an IPL of z/VM in the TEST LPAR? I > realize we'll still need a POR to make these permanent (absent using HCD, > which we haven't deployed yet) - but for the time being are these changes > "permanent" unless/until a POR is performed, or will they disappear when the > z/VM TEST LPAR is IPL'd? > > Thanks for your help Greg, MUCH appreciated. :) > > -Mike -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Hi Greg, Yep, that was it - and I think I understand the syntax a little better now that we have a working set of parms (I was also confusing the values of CSS and RCSS). So now I have all of the CHPIDS online to my TEST LPAR: q chpids 0 1 2 3 4 5 6 7 8 9 A B C D E F 0x + + + + + + + + . . . . . . . . 1x . . . . + + + + . . . . . . . . 2x + + + + + + + + + + + + + + + + 3x . . . . . . . . . . . . . . . . 4x . . . . . . . . . . . . . . . . 5x . . . . . . . . . . . . . . . . 6x . . . . . . . . . . . . . . . . 7x . . . . . . . . . . . . . . . . 8x + + + + + + + + + + + + + + + + 9x + + + + + + + + + + + + + + + + Ax . . . . . . . . . . . . . . . . Bx . . . . . . . . . . . . . . . . Cx . . . . . . . . . . . . . . . . Dx . . . . . . . . . . . . . . . . Ex . . . . . . . . . . . . . . . . Fx . . . . . . . . . . . . . . . . I modified the control units, eg: cp modify cu 4000 css 1 add_css_access chpid 80 84 88 8C 81 85 89 8D cuadd 0 And added all of the IODevs, eg: cp def dev 4000-4080 unitadd 00 cu 4000 css 1 dasd par TEST All of the DASD and control units came online normally. :) Now, my last (hopefully!) question is this. Are these modifications stored in the HSA, and/or will they survive an IPL of z/VM in the TEST LPAR? I realize we'll still need a POR to make these permanent (absent using HCD, which we haven't deployed yet) - but for the time being are these changes "permanent" unless/until a POR is performed, or will they disappear when the z/VM TEST LPAR is IPL'd? Thanks for your help Greg, MUCH appreciated. :) -Mike -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Thursday, March 18, 2010 1:27 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Adding a second CSS Hi Mike, yeppurr it looks like I had my ones and zeros backwards CP MODify CHPID 80 CSS 1 ADD_CSS_access RCSS 0 INITial_access TEST ACCess_list TEST On Thu, Mar 18, 2010 at 1:15 PM, Michael Coffin wrote: > Hi Greg, > > TEST is in CSS 1. The DASD on CHPID 80 that I'm trying to give TEST access > to is in CSS 0. Here are snips from our IOCP: > > RESOURCE PARTITION=((CSS(0),(sys1,2),(sys2,1),(sys3* > T,4),(sys4,3),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,* > B),(*,C),(*,D),(*,E),(*,F)),(CSS(1),(TEST,1),(*,2),(*,3)* > ,(*,4),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,B),(*,C),(* > *,D),(*,E),(*,F))) > CHPID PATH=(CSS(0),80),SHARED, * > PARTITION=((sys1,sys2,sys3,sys4),(=)), * > PCHID=290,TYPE=FC > > CNTLUNIT CUNUMBR=4000,PATH=((CSS(0),80,84,88,8C,81,85,89,8D)),* > UNITADD=((00,256)),CUADD=0,UNIT=2107 > IODEVICE ADDRESS=(4000,128),CUNUMBR=(4000),STADET=Y,UNIT=3390 > > Thanks for the info about "adding" access vs. "replacing", I was pretty > concerned about that. :) > > -Mike > > > -Original Message- > From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On > Behalf Of Gregg > Sent: Thursday, March 18, 2010 12:16 PM > To: IBMVM@LISTSERV.UARK.EDU > Subject: Re: Adding a second CSS > > Hi Mike, > what CSS is Test in? Maybe I got my 1's and 0's backwards... would > you paste, from the IOCP the RESOURCE PARTITION statement, CHPID 80 > and the CNTLUNIT statement for CU 4000? > My read earlier was that CHPID 80 is currently defined in CSS 1 and I > assumed TEST was in CSS 0. > The modify adds the lpnames to the ACC/INIT lists. > We've 'spanned' channels channels here, where we first had to DEFine > in one CSS and then MODify for the second. > > > On Thu, Mar 18, 2010 at 7:31 AM, Michael Coffin > wrote: >> Hi Greg, >> >> Ooops, yes - CAPS is a typo in this email (not in the actual command), >> should be TEST. :) >> >> I tried the CP MODIFY with the parms you suggested, but I'm still getting >> HCP0864E: >> >> CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST >> ACCess_list TEST >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> >> Perhaps, does this command need to be executed on a system that already > has >> access to CSS 0? Are we in a "chicken and egg" scenario where I'm >> requesting access to the CSS, but I'll never get it because I don't > already >> have access to the CSS? >> >> PS: Do you know if the ACCESS_LIST needs to be the COMPLETE list of > systems >>
Re: Adding a second CSS
Hi Mike, yeppurr it looks like I had my ones and zeros backwards CP MODify CHPID 80 CSS 1 ADD_CSS_access RCSS 0 INITial_access TEST ACCess_list TEST On Thu, Mar 18, 2010 at 1:15 PM, Michael Coffin wrote: > Hi Greg, > > TEST is in CSS 1. The DASD on CHPID 80 that I'm trying to give TEST access > to is in CSS 0. Here are snips from our IOCP: > > RESOURCE PARTITION=((CSS(0),(sys1,2),(sys2,1),(sys3* > T,4),(sys4,3),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,* > B),(*,C),(*,D),(*,E),(*,F)),(CSS(1),(TEST,1),(*,2),(*,3)* > ,(*,4),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,B),(*,C),(* > *,D),(*,E),(*,F))) > CHPID PATH=(CSS(0),80),SHARED, * > PARTITION=((sys1,sys2,sys3,sys4),(=)), * > PCHID=290,TYPE=FC > > CNTLUNIT CUNUMBR=4000,PATH=((CSS(0),80,84,88,8C,81,85,89,8D)),* > UNITADD=((00,256)),CUADD=0,UNIT=2107 > IODEVICE ADDRESS=(4000,128),CUNUMBR=(4000),STADET=Y,UNIT=3390 > > Thanks for the info about "adding" access vs. "replacing", I was pretty > concerned about that. :) > > -Mike > > > -Original Message- > From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On > Behalf Of Gregg > Sent: Thursday, March 18, 2010 12:16 PM > To: IBMVM@LISTSERV.UARK.EDU > Subject: Re: Adding a second CSS > > Hi Mike, > what CSS is Test in? Maybe I got my 1's and 0's backwards... would > you paste, from the IOCP the RESOURCE PARTITION statement, CHPID 80 > and the CNTLUNIT statement for CU 4000? > My read earlier was that CHPID 80 is currently defined in CSS 1 and I > assumed TEST was in CSS 0. > The modify adds the lpnames to the ACC/INIT lists. > We've 'spanned' channels channels here, where we first had to DEFine > in one CSS and then MODify for the second. > > > On Thu, Mar 18, 2010 at 7:31 AM, Michael Coffin > wrote: >> Hi Greg, >> >> Ooops, yes - CAPS is a typo in this email (not in the actual command), >> should be TEST. :) >> >> I tried the CP MODIFY with the parms you suggested, but I'm still getting >> HCP0864E: >> >> CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST >> ACCess_list TEST >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> >> Perhaps, does this command need to be executed on a system that already > has >> access to CSS 0? Are we in a "chicken and egg" scenario where I'm >> requesting access to the CSS, but I'll never get it because I don't > already >> have access to the CSS? >> >> PS: Do you know if the ACCESS_LIST needs to be the COMPLETE list of > systems >> to have access to the CSS, or only the system you are adding via the CP >> MODIFY command? In other words, will the system(s) specified in > ACCESS_LIST >> be added to the existing access list, or REPLACE the existing access list? >> >> -Mike >> >> -Original Message- >> From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On >> Behalf Of Gregg >> Sent: Thursday, March 18, 2010 5:50 AM >> To: IBMVM@LISTSERV.UARK.EDU >> Subject: Re: Adding a second CSS >> >> Mike, is "init CAPS" a typo/some email malfunction? >> >> The CSS value is the target CSS and the RCSS value "is the reference >> channel-subsystem image from which the channel-path characteristics >> are to be copied" >> INITial_access is optional, but ACCess_list is required and as you >> know the INIT and ACC lpname(s list) do not need to be the same in the >> INIT and ACC lists >> I believe the modify chpid should be >> CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST >> ACCess_list TEST >> >>> The spindle I’m trying to reach is 4030 on CU 4000 on CHPID 80. So > first >> I >>> try to define the CHPID for our LPAR (TEST): >>> >>> CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 >>> othersys3 TEST INIT TEST >>> >>> But I get: >>> >>> HCPCCO864E Partition TEST does not have access to CSS 0 >>> >>> So I try to authorize our partition: >>> >>> cp modify chpid 80 add access_list TEST >>> HCPCCO864E Partition TEST does not have access to CSS 0 >>> >>> H, well the chpid exists (even though it shows as undefined in QUERY >>> CHPIDS), so I try to modify the chpid to add the CSS: >>> >>> cp modify chpid 80 css 0 add_css rcss 0 init CAPS >>> HCPZPM6704E Missing token at end of line >> >> -- >> Gregg Reed >> "No Plan, survives execution" >> > > > > -- > Gregg Reed > "No Plan, survives execution" > -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Hi Greg, TEST is in CSS 1. The DASD on CHPID 80 that I'm trying to give TEST access to is in CSS 0. Here are snips from our IOCP: RESOURCE PARTITION=((CSS(0),(sys1,2),(sys2,1),(sys3* T,4),(sys4,3),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,* B),(*,C),(*,D),(*,E),(*,F)),(CSS(1),(TEST,1),(*,2),(*,3)* ,(*,4),(*,5),(*,6),(*,7),(*,8),(*,9),(*,A),(*,B),(*,C),(* *,D),(*,E),(*,F))) CHPID PATH=(CSS(0),80),SHARED,* PARTITION=((sys1,sys2,sys3,sys4),(=)), * PCHID=290,TYPE=FC CNTLUNIT CUNUMBR=4000,PATH=((CSS(0),80,84,88,8C,81,85,89,8D)),* UNITADD=((00,256)),CUADD=0,UNIT=2107 IODEVICE ADDRESS=(4000,128),CUNUMBR=(4000),STADET=Y,UNIT=3390 Thanks for the info about "adding" access vs. "replacing", I was pretty concerned about that. :) -Mike -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Thursday, March 18, 2010 12:16 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Adding a second CSS Hi Mike, what CSS is Test in? Maybe I got my 1's and 0's backwards... would you paste, from the IOCP the RESOURCE PARTITION statement, CHPID 80 and the CNTLUNIT statement for CU 4000? My read earlier was that CHPID 80 is currently defined in CSS 1 and I assumed TEST was in CSS 0. The modify adds the lpnames to the ACC/INIT lists. We've 'spanned' channels channels here, where we first had to DEFine in one CSS and then MODify for the second. On Thu, Mar 18, 2010 at 7:31 AM, Michael Coffin wrote: > Hi Greg, > > Ooops, yes - CAPS is a typo in this email (not in the actual command), > should be TEST. :) > > I tried the CP MODIFY with the parms you suggested, but I'm still getting > HCP0864E: > > CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST > ACCess_list TEST > HCPCCO864E Partition TEST does not have access to CSS 0 > > > Perhaps, does this command need to be executed on a system that already has > access to CSS 0? Are we in a "chicken and egg" scenario where I'm > requesting access to the CSS, but I'll never get it because I don't already > have access to the CSS? > > PS: Do you know if the ACCESS_LIST needs to be the COMPLETE list of systems > to have access to the CSS, or only the system you are adding via the CP > MODIFY command? In other words, will the system(s) specified in ACCESS_LIST > be added to the existing access list, or REPLACE the existing access list? > > -Mike > > -Original Message- > From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On > Behalf Of Gregg > Sent: Thursday, March 18, 2010 5:50 AM > To: IBMVM@LISTSERV.UARK.EDU > Subject: Re: Adding a second CSS > > Mike, is "init CAPS" a typo/some email malfunction? > > The CSS value is the target CSS and the RCSS value "is the reference > channel-subsystem image from which the channel-path characteristics > are to be copied" > INITial_access is optional, but ACCess_list is required and as you > know the INIT and ACC lpname(s list) do not need to be the same in the > INIT and ACC lists > I believe the modify chpid should be > CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST > ACCess_list TEST > >> The spindle Im trying to reach is 4030 on CU 4000 on CHPID 80. So first > I >> try to define the CHPID for our LPAR (TEST): >> >> CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 >> othersys3 TEST INIT TEST >> >> But I get: >> >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> So I try to authorize our partition: >> >> cp modify chpid 80 add access_list TEST >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> H, well the chpid exists (even though it shows as undefined in QUERY >> CHPIDS), so I try to modify the chpid to add the CSS: >> >> cp modify chpid 80 css 0 add_css rcss 0 init CAPS >> HCPZPM6704E Missing token at end of line > > -- > Gregg Reed > "No Plan, survives execution" > -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Hi Mike, what CSS is Test in? Maybe I got my 1's and 0's backwards... would you paste, from the IOCP the RESOURCE PARTITION statement, CHPID 80 and the CNTLUNIT statement for CU 4000? My read earlier was that CHPID 80 is currently defined in CSS 1 and I assumed TEST was in CSS 0. The modify adds the lpnames to the ACC/INIT lists. We've 'spanned' channels channels here, where we first had to DEFine in one CSS and then MODify for the second. On Thu, Mar 18, 2010 at 7:31 AM, Michael Coffin wrote: > Hi Greg, > > Ooops, yes - CAPS is a typo in this email (not in the actual command), > should be TEST. :) > > I tried the CP MODIFY with the parms you suggested, but I'm still getting > HCP0864E: > > CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST > ACCess_list TEST > HCPCCO864E Partition TEST does not have access to CSS 0 > > > Perhaps, does this command need to be executed on a system that already has > access to CSS 0? Are we in a "chicken and egg" scenario where I'm > requesting access to the CSS, but I'll never get it because I don't already > have access to the CSS? > > PS: Do you know if the ACCESS_LIST needs to be the COMPLETE list of systems > to have access to the CSS, or only the system you are adding via the CP > MODIFY command? In other words, will the system(s) specified in ACCESS_LIST > be added to the existing access list, or REPLACE the existing access list? > > -Mike > > -Original Message- > From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On > Behalf Of Gregg > Sent: Thursday, March 18, 2010 5:50 AM > To: IBMVM@LISTSERV.UARK.EDU > Subject: Re: Adding a second CSS > > Mike, is "init CAPS" a typo/some email malfunction? > > The CSS value is the target CSS and the RCSS value "is the reference > channel-subsystem image from which the channel-path characteristics > are to be copied" > INITial_access is optional, but ACCess_list is required and as you > know the INIT and ACC lpname(s list) do not need to be the same in the > INIT and ACC lists > I believe the modify chpid should be > CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST > ACCess_list TEST > >> The spindle I’m trying to reach is 4030 on CU 4000 on CHPID 80. So first > I >> try to define the CHPID for our LPAR (TEST): >> >> CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 >> othersys3 TEST INIT TEST >> >> But I get: >> >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> So I try to authorize our partition: >> >> cp modify chpid 80 add access_list TEST >> HCPCCO864E Partition TEST does not have access to CSS 0 >> >> H, well the chpid exists (even though it shows as undefined in QUERY >> CHPIDS), so I try to modify the chpid to add the CSS: >> >> cp modify chpid 80 css 0 add_css rcss 0 init CAPS >> HCPZPM6704E Missing token at end of line > > -- > Gregg Reed > "No Plan, survives execution" > -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Hi Greg, Ooops, yes - CAPS is a typo in this email (not in the actual command), should be TEST. :) I tried the CP MODIFY with the parms you suggested, but I'm still getting HCP0864E: CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST ACCess_list TEST HCPCCO864E Partition TEST does not have access to CSS 0 Perhaps, does this command need to be executed on a system that already has access to CSS 0? Are we in a "chicken and egg" scenario where I'm requesting access to the CSS, but I'll never get it because I don't already have access to the CSS? PS: Do you know if the ACCESS_LIST needs to be the COMPLETE list of systems to have access to the CSS, or only the system you are adding via the CP MODIFY command? In other words, will the system(s) specified in ACCESS_LIST be added to the existing access list, or REPLACE the existing access list? -Mike -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Thursday, March 18, 2010 5:50 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Adding a second CSS Mike, is "init CAPS" a typo/some email malfunction? The CSS value is the target CSS and the RCSS value "is the reference channel-subsystem image from which the channel-path characteristics are to be copied" INITial_access is optional, but ACCess_list is required and as you know the INIT and ACC lpname(s list) do not need to be the same in the INIT and ACC lists I believe the modify chpid should be CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST ACCess_list TEST > The spindle Im trying to reach is 4030 on CU 4000 on CHPID 80. So first I > try to define the CHPID for our LPAR (TEST): > > CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 > othersys3 TEST INIT TEST > > But I get: > > HCPCCO864E Partition TEST does not have access to CSS 0 > > So I try to authorize our partition: > > cp modify chpid 80 add access_list TEST > HCPCCO864E Partition TEST does not have access to CSS 0 > > H, well the chpid exists (even though it shows as undefined in QUERY > CHPIDS), so I try to modify the chpid to add the CSS: > > cp modify chpid 80 css 0 add_css rcss 0 init CAPS > HCPZPM6704E Missing token at end of line -- Gregg Reed "No Plan, survives execution"
Re: Adding a second CSS
Mike, is "init CAPS" a typo/some email malfunction? The CSS value is the target CSS and the RCSS value "is the reference channel-subsystem image from which the channel-path characteristics are to be copied" INITial_access is optional, but ACCess_list is required and as you know the INIT and ACC lpname(s list) do not need to be the same in the INIT and ACC lists I believe the modify chpid should be CP MODify CHPID 80 CSS 0 ADD_CSS_access RCSS 1 INITial_access TEST ACCess_list TEST > The spindle I’m trying to reach is 4030 on CU 4000 on CHPID 80. So first I > try to define the CHPID for our LPAR (TEST): > > CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 > othersys3 TEST INIT TEST > > But I get: > > HCPCCO864E Partition TEST does not have access to CSS 0 > > So I try to authorize our partition: > > cp modify chpid 80 add access_list TEST > HCPCCO864E Partition TEST does not have access to CSS 0 > > H, well the chpid exists (even though it shows as undefined in QUERY > CHPIDS), so I try to modify the chpid to add the CSS: > > cp modify chpid 80 css 0 add_css rcss 0 init CAPS > HCPZPM6704E Missing token at end of line -- Gregg Reed "No Plan, survives execution"
Adding a second CSS
Hi Folks, We have an LPAR with an I/O subsystem on CSS 1. There is some other DASD we need to access in CSS 0. We don't have HCD set up, and can't do a POR at the moment, so I'm trying to do DEFINE/MODIFY commands (we do have dynamic IOCP active). Here is what our CHPIDS look like: q chpids 0 1 2 3 4 5 6 7 8 9 A B C D E F 0x + + + + + + + + . . . . . . . . 1x . . . . + + + + . . . . . . . . 2x + + + + + + + + + + + + + + + + 3x . . . . . . . . . . . . . . . . 4x . . . . . . . . . . . . . . . . 5x . . . . . . . . . . . . . . . . 6x . . . . . . . . . . . . . . . . 7x . . . . . . . . . . . . . . . . 8x . . . . . . . . . . . . . . . . 9x + + + + + + + + + + + + + + + + Ax . . . . . . . . . . . . . . . . Bx . . . . . . . . . . . . . . . . Cx . . . . . . . . . . . . . . . . Dx . . . . . . . . . . . . . . . . Ex . . . . . . . . . . . . . . . . Fx . . . . . . . . . . . . . . . . The spindle I'm trying to reach is 4030 on CU 4000 on CHPID 80. So first I try to define the CHPID for our LPAR (TEST): CP DEFINE CHPID 80 PCHID 290 CSS 0 TYPE FC SHARED ACC othersys1 othersys2 othersys3 TEST INIT TEST But I get: HCPCCO864E Partition TEST does not have access to CSS 0 So I try to authorize our partition: cp modify chpid 80 add access_list TEST HCPCCO864E Partition TEST does not have access to CSS 0 H, well the chpid exists (even though it shows as undefined in QUERY CHPIDS), so I try to modify the chpid to add the CSS: cp modify chpid 80 css 0 add_css rcss 0 init CAPS HCPZPM6704E Missing token at end of line What am I doing wrong here? -Mike