[Info-vax] strange cluster behaviour: 2

[Phillip Helbig---undress to reply]

In article <4d57e0e3 at ns.langstoeger.at>, peter at langstoeger.at (Peter
'EPLAN' LANGSTOEGER) writes: 

> In article <ij8e46$5sv$2 at online.de>, helbig at astro.multiCLOTHESvax.de (Phillip Helbig---undress to reply) writes:
> >When setting up new satellites, I get the standard message that global 
> >sections is too low for DECwindows to start, and get asked whether 
> >AUTOGEN should be run to correct it.  If I say yes, it doesn't get 
> >corrected, i.e. after the automatic reboot the same message comes again 
> >and would do so forever if I didn't correct it by hand. 
> 
> This is a VMS/MOTIF bug for decades now. Value added by MOTIF installation
> to CLU$PARAMS.DAT is way too low (600)...

OK.

> Add MIN_GBLSECTIONS=1024 to SYS$SYSTEM:MODPARAMS.DAT

I have 950; AUTOGEN has never set it to anything higher than that.

Would a power of 2 be better?

> Few utils manipulate the MODPARAM.DAT themselves (Pathworks comes to mind)
> and AUTOGEN.COM itself is not one of them (reads it, but not change it).

I was thinking it was the cluster-configuration procedure which might 
modify it, since it creates it in the first place.