OpenVMS clustering, cluster management utilities, protocols, building and managing clusters of any scale.
-
Topic author
sms
- Master
- Posts: 425
- Joined: Fri Aug 21, 2020 5:18 pm
- Reputation: 0
-
Status:
Offline
Post
by sms » Wed Jan 10, 2024 6:41 pm
Code: Select all
I recently tried to use MONI /PROC /TOPCPU /NODE on an IA64 system to
monitor activity on an x86_64 system. After a tachistoscopic message
which might have been something like MONITOR-I-ESTABCON, the terminal
display was cleared, and no output followed. Ctrl/T shows no change in
CPU, and only the usual one IO per Ctrl/T.
Known bug? Already fixed?
ITS $ tcpip show vers
HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.7 - ECO 5
on an HP rx2600 (1.50GHz/6.0MB) running OpenVMS V8.4-2L3
V87 $ tcpip show vers
VSI TCP/IP Services for OpenVMS x86_64 Version V6.0
on a VMware, Inc. VMware7,1 running OpenVMS V9.2-1
V87 has VMS921X_UPDATE V2.0 (and, I believe, all the other latest
stuff for peons).
P.S. Is the HELP for MONITOR just horrible, or am I missing something
obvious (like all the command qualifiers)?
-
sodjan
- Active Contributor
- Posts: 40
- Joined: Mon Apr 24, 2023 3:51 am
- Reputation: 0
-
Status:
Offline
Post
by sodjan » Wed Jan 10, 2024 11:25 pm
As I remember, you'd better use:
$ monitor
Monitor> help
Instead of:
$ help monitor
Btw...
Your posts are always hard to read since they seems to be entered as a whole within code-tags. Have to scroll left/right for each line. Just if you do not already know that...
Last edited by
sodjan on Wed Jan 10, 2024 11:30 pm, edited 1 time in total.
-
Topic author
sms
- Master
- Posts: 425
- Joined: Fri Aug 21, 2020 5:18 pm
- Reputation: 0
-
Status:
Offline
Post
by sms » Thu Jan 11, 2024 12:16 am
Code: Select all
> $ monitor
> Monitor> help
I didn't see the qualifiers there, either. But please feel free to
direct me to the good stuff when you find it. I _did_ find them in the
manual. Look for the (untagged) section, "Command Qualifier
Descriptions", after:
https://docs.vmssoftware.com/
vsi-openvms-system-management-utilities-reference-manual-volume-ii-m-z/#d0e467
Anyone who wants to de-scramble the order of the following items at
https://docs.vmssoftware.com/ has my permission, too:
VSI OpenVMS System Manager's Manual, Volume 1: Essentials
VSI OpenVMS System Management Utilities Reference Manual, Volume II: M-Z
VSI OpenVMS System Management Utilities Reference Manual, Volume I: A-L
VSI OpenVMS System Manager's Manual, Volume 2: Tuning, Monitoring, and
Complex Systems
> Your posts are always hard to read since they seems to be entered as a
> whole within code-tags. Have to scroll left/right for each line.
I like having my white space left unchanged. The lines are normally
limited to 72 characters. If that's too wide for you, perhaps a VT100
would be a better option than whatever you're using. Your complaint, on
the other hand, looked like this here (Firefox on a Mac):
Your posts are always hard to read since they seems to be entered as a whole within code-tags. Have to scroll left/right for
each line. Just if you do not already know that...
which is wider than any of mine were. I normally don't need to scroll
sideways to read my own stuff in my web browser. How wide (narrow?) is
your display?
-
sodjan
- Active Contributor
- Posts: 40
- Joined: Mon Apr 24, 2023 3:51 am
- Reputation: 0
-
Status:
Offline
Post
by sodjan » Thu Jan 11, 2024 2:22 am
My phone shows up to 45 chars when standing up and approx 65 chars when laying down on each line. I expect quite a few forum users are using their mobile devices today. Suggesting a VT100 is of course just silly.
Any non-code-tagged text is automatically formatted to whatever the display properties are of the device used.
All others members posts are displayed and formatted just fine both ways and only code examples (within code-tags) are displayed as-is (as expected).
So again, it is only your posts that takes more work to read. They also stands out with a different look then all other posts. Just thought you wanted to know. Or at least I thought you need to know...
![Wink ;)](./images/smilies/icon_e_wink.gif)
-
Topic author
sms
- Master
- Posts: 425
- Joined: Fri Aug 21, 2020 5:18 pm
- Reputation: 0
-
Status:
Offline
Post
by sms » Thu Jan 11, 2024 8:02 am
Code: Select all
> My phone [...]
Ah. I don't expect anyone to use such a limited device for serious
work.
> All others members posts [...]
I like having my white space left unchanged.
-
dgordon
- VSI Expert
-
Valued Contributor
- Posts: 60
- Joined: Tue May 09, 2023 7:57 am
- Reputation: 1
-
Status:
Offline
Post
by dgordon » Thu Jan 11, 2024 8:50 am
P.S. Is the HELP for MONITOR just horrible, or am I missing something
obvious (like all the command qualifiers)?
HELP MONITOR MONITOR will give you the command qualifiers.
As for your problem, I just tried it on the internal VMS development cluster and it worked. Your two nodes are in the same cluster?
Executive Vice President of InfoServer Engineering at VSI.
-
Topic author
sms
- Master
- Posts: 425
- Joined: Fri Aug 21, 2020 5:18 pm
- Reputation: 0
-
Status:
Offline
Post
by sms » Thu Jan 11, 2024 2:42 pm
Code: Select all
> HELP MONITOR MONITOR will give you the command qualifiers.
Ah. Thanks. I might have been reading "INITIALIZE MONITOR" under
"Additional information available:" as "INITIALIZE_MONITOR" instead of
as two separate topics. (But I also thought that I'd tried MONITOR
MONITOR, and still missed them.) I might have organized the stuff
differently, but there's some chance that I'll find them next time.
> [...] on the internal VMS development cluster and it worked.
Naturally.
> [...] Your two nodes are in the same cluster?
Yes. (I thought that that was a requirement for /NODE.)
ITS $ set term /la36 /wid = 80
ITS $ show cluster
View of Cluster from system ID 1164 node: ITS 11-JAN-2024 13:32:41
+-----------------------+---------+
| SYSTEMS | MEMBERS |
+--------+--------------+---------+
| NODE | SOFTWARE | STATUS |
+--------+--------------+---------+
| ITS | VMS V8.4-2L3 | MEMBER |
| V86 | VMS E9.2 | BRK_NON |
| V87 | VMS V9.2-1 | MEMBER |
| ALP | VMS V8.4-2L1 | BRK_NON |
+--------+--------------+---------+
Interestingly, it works as expected facing the other way: on the
x86_64 system, with /NODE = IA64_system.
Up-time on ITS is about six months, so it's possible that it's
accumulated some confusion over time. If I get bored enough, I might
try firing up another IA64 box for comparison. (Restarting ITS is
almost always inconvenient.)
After getting my x86_64 software up to date, I looked at what was
available on the Service Platform for IA64, but "Only accessible" there
left only "No packages to show", so I don't know if any of the stuff for
real/paying IA64 customers might matter.
-
imiller
- Master
- Posts: 166
- Joined: Fri Jun 28, 2019 8:45 am
- Reputation: 0
- Location: South Tyneside, UK
-
Status:
Offline
-
Contact:
Post
by imiller » Fri Jan 12, 2024 8:39 am
for paying VMS I64 customers patches are on ECOKITS at vsiftp server. Sometimes there are emails about new patches such as the recent update and filesys patches but I periodically check that server for new things.
Ian Miller
[ personal opinion only. usual disclaimers apply. Do not taunt happy fun ball ].
-
Topic author
sms
- Master
- Posts: 425
- Joined: Fri Aug 21, 2020 5:18 pm
- Reputation: 0
-
Status:
Offline
Post
by sms » Sat Jan 13, 2024 3:59 pm
Code: Select all
> [...] If I get bored enough, I might try firing up another IA64 box
> for comparison. [...]
I awakened my spare/experimental rx2600 noisemaker, and, after
updating its License Database and DECnet node registry
(DECNET_REGISTER), I ran the experiment there. It was slower to start
(so that I had time to read the message: "%MONITOR-I-ESTABCON,
establishing connection to remote node(s)..."), but otherwise it worked
as expected.
I tried all permutations, and only one failed:
ITS: MONITOR PROCESSES /TOPCPU /NODE = ITS OK
ITS: MONITOR PROCESSES /TOPCPU /NODE = REX OK
ITS: MONITOR PROCESSES /TOPCPU /NODE = V87 HANG
REX: MONITOR PROCESSES /TOPCPU /NODE = ITS OK
REX: MONITOR PROCESSES /TOPCPU /NODE = REX OK
REX: MONITOR PROCESSES /TOPCPU /NODE = V87 OK
V87: MONITOR PROCESSES /TOPCPU /NODE = ITS OK
V87: MONITOR PROCESSES /TOPCPU /NODE = REX OK
V87: MONITOR PROCESSES /TOPCPU /NODE = V87 OK
The server seems to start on V87 as expected:
ITS $ show syst /proc = monitor* /node = v87
OpenVMS V9.2-1 on node V87 12-JAN-2024 22:28:27.25 Uptime 4 05:17:37
Pid Process Name State Pri I/O CPU Page flts Pages
2340048E MONITOR_SERVER HIB 15 286 0 00:00:00.11 389 349 N
Interestingly (perhaps), I also let the hung command stay hung for a
longer time, and after about ten minutes:
õÃzõÃz$.õÃzO.?POÃIO
%MONITOR-E-NODEXIT, lost connection to node V87. Node removed
-SYSTEM-S-NORMAL, normal successful completion
%MONITOR-E-NODENOTAVAIL, requested nodes are not available
Perhaps the remote MONITOR_SERVER process got tired of waiting?
So, I'm mystified, but the x86_64 system seems to be more blameless
than the one IA64 system (ITS). I'll try to remember to retry this when
next ITS gets whacked in the head.
For the record:
REX $ tcpip show vers
HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.7 - ECO 5
on an HP rx2600 (1.50GHz/6.0MB) running OpenVMS V8.4-2L3
> [...] ECOKITS at vsiftp server. [...]
Thanks for the reminder. When I get bored enough (again), I'll try
to see who here has what installed, and what else might be available.