Re: Disk Volumes added in INIT:INITCMD sometimes cause *CMD to fail
|
That's a good question. I think that the only thing that you need to patch is the 4 character volume name but it's been a long time since I did that so I'm not entirely sure. You might also have to put the public volume number in after the vol name as a halfword
You could test this by using IPLREADER to look at and patch the system during IPL without commiting anything to RamRod. IPLREADER is the program that loads the system into memory after IPL. It has a simple command language which is described on or near page 77 in the operator manual. You could use it to patch the disk table and see if it works before you create a new *IPL.0 file.
Mike
toggle quoted message
Show quoted text
On 21 Apr 2024, at 17:41, John Palmer wrote:
I have additional disk volumes that I add to the system by running SYS:DSK from INIT:INITCMD.?
I am seeing an issue where other processes start ahead of this and if the volume isn't mounted yet, errors occur.
The biggest one is the with *CMD. If the current log file is on one of the volumes that has quite gotten mounted yet, *CMD will not start properly.? In this case, it locks *STATISTICS and everything piles up behind that waiting on the file. In this case, the *CMD job needs to be BLASTed (STOP and GOOSE don't work) and the log file will need to be manually created (and set to SIZE=400P and MAXSIZE=400P)
I see that there appear to be 6 slots for volumes created in TABLES. MTS600 is the only one filled in.??
Do I just need to use RAMROD to fill in the volume ID for the other empty slots so the volumes get mounted right at startup or are there other parameters involved in that data structure that need to be set as well?
|
