I can’t show you the name, but here’s an email I received on March 22nd from a customer. For information on CHARON-VAX, check out http://www.stanq.com/charon-vax.html.
Subject: BEANIE MOVE COMPLETE in ERIE PA
Date: Mon, 22 Mar 2010 11:04:46 -0400
BEANIE$ sho mem
System Memory Resources on 22-MAR-2010 11:01:59.03
Physical Memory Usage (pages): Total Free In Use Modified
Main Memory (128.00Mb) 262144 26089 216652 19403
Virtual I/O Cache Usage (pages): Total Free In Use Maximum
Cache Memory 15997 871 15126 167677
Slot Usage (slots): Total Free Resident Swapped
Process Entry Slots 284 178 96 10
Balance Set Slots 280 184 94 2
Dynamic Memory Usage (bytes): Total Free In Use Largest
Nonpaged Dynamic Memory 16000000 10547584 5452416 10421888
Paged Dynamic Memory 8000000 6670048 1329952 6666368
Paging File Usage (pages): Free Reservable Total
PSWAP:[BEANIE]SWAPFILE.SYS;1 994472 994472 999992
DISK$VAXVMSRL5:[SYS0.SYSEXE]PAGEFILE.SYS 4819 -480 5000
PSWAP:[BEANIE]PAGEFILE.SYS;1 907387 610340 999992
Of the physical pages in use, 62343 pages are permanently allocated to OpenVMS.
and a total success…. Here it is with most people logged in and only a bit of swapping and all is well. Remember, the original system had 384MB but this system is fitting fine in 128 using the VMS OS to do what it knows what to do best, handling virtual address space.
But, since I know you would want to know, I did have a problem yesterday that I think I resolved. It was that sometimes you couldn’t log in while it got hung up in a RWMBXW but then cleared itself.
Here is my buddy, Steve Hoffman’s take on this:
However, I resolved it by doing something else that I think was causing the MBX waits… setting PFRATL page trimmer back to ZERO, where I think I will leave it.
3rd paragraph under AWSA, I realized I should NOT have this trimming as the processes on this box never grow to be very large anyway, as we now can see how little is swapped.
It was trimming and then things grew and it a kind of thrashing situation on a bored system (not-loaded at all). Now that PFRATL=0 all is well, I get no RWMBX delays anymore.
I had to leave the system as a VAX-CLUSTER even though it is now only one member as thet CLUSTER alias is used in PROXY (no password) access to PDP-11s. Don’t even ask. I could use Charon-PDPs!!!
Just thought you might want to know that it is virtually complete. 🙂
OH by the way, remember, I had to place the VAX NIC on windows to 10/half and left it there. I now get no COLLISION DETECT CHECK FAILURES in LANCP, which I got alot of when I had it set to 100full.
BEANIE$ mcr lancp show dev eza/count
0 Collision detect check failure
0 Unrecognized frame destination
0 System buffer unavailable
0 User buffer unavailable