Unexplained Console Lockup

[tim lloyd_1]

Hi, I maintain a legacy system running VMS V8.2-1. The system features 2 Itanium computers - one master and one standby.

Each computer uses a separate thin client (running windows XP) as its console. OPA0: is connected to the serial port of the thin client PC which runs a PC emulator called Powerterm (Ericom systems).

We experienced an unusual situation last week where the console locked up on the master computer. I could access the computer remotely and the onsite operators could access the computer via telnet.

I remotely initiated a reboot of the computer but this did not appear on the console. The only way the operators could retrieve the situation was to physically power down the computer.

I am confident that noone had pressed CTRL-S (lock the window) on the console. Although I didn't think of checking that at the time.

I know this is a vague outline but has anyone experienced similar? Or any suggestions on how to troubleshoot?

Thanks

[John Gillings]

Tim,

If this happens again, remove the PC and try accessing the console port directly (dust off a real VT220 or similar).

Rebooting the OpenVMS system is extremely unlikely to have any effect, if it's the PC that's blocking the flow. Maybe rebooting the PCs would help, but I'd physically disconnect them from the console port first.

The console is a VERY dumb subsystem, just a serial line with low speed characters flowing through it. Don't try to run it at high speed - 9600 baud should be adequate for the kind of tasks the console NEEDS to be used for. If you don't NEED to use the console to do something - DON'T.

Although you can attach more or less anything that speaks serial to the console, it doesn't necessarily work all the time/circumstances.

Consider purchasing console management software for your PCs. There are several to choose from.

[Hein van den Heuvel]

Control-Q is always my first try...

Last time this _almost_ happened to a customer I worked with there was a SET HOST /DTE to la-la-land running on the console back out to the other serial port.
We tried just about every control-sequence except ^\ to release it.
It was only seeing DTEPAD with SHOW PROC/CONT that we realized what was happening.
Still, reboot would have cleared that, and did not for you.

Do those Itaniums not have an MP processor?
I find telnet access to the console more reliable and much more handy than serial line.

I hope the operators tried recycling the PC thin client before power cycling the Itanium.

Smells like a hardware / Uart failure in this case. Ouch.

fwiw,
Hein

[Jan van den Ende]

Tim,

from your Forum Profile:


I have assigned points to 26 of 57 responses to my questions.


Maybe you can find some time to do some assigning?

http://forums1.itrc.hp.com/service/forums/helptips.do?#33

Mind, I do NOT say you necessarily need to give lots of points. It is fully up to _YOU_ to decide how many. If you consider an answer is not deserving any points, you can also assign 0 ( = zero ) points, and then that answer will no longer be counted as unassigned.
Consider, that every poster took at least the trouble of posting for you!

To easily find your streams with unassigned points, click your own name somewhere.
This will bring up your profile.
Near the bottom of that page, under the caption "My Question(s)" you will find "questions or topics with unassigned points " Clicking that will give all, and only, your questions that still have unassigned postings.
If you have closed some of those streams, you must "Reopen" them to "Submit points". (After which you can "Close" again)

Do not forget to explicitly activate "Submit points", or your effort gets lost again!!

Thanks on behalf of your Forum colleagues.

PS. - nothing personal in this. I try to post it to everyone with this kind of assignment ratio in this forum. If you have received a posting like this before - please do not take offence - none is intended!

PPS. - Zero points for this.

Proost.

Have one on me.

jpe

[tim lloyd_1]

Thanks for the help so far. Firstly Jan: Noted about assigning points. When I first started using the forum I didn't realise the significance. I will begin to address this but I will have to take a bit of time to familiarise myself with the individual threads first. Anyway, thanks for pointing this out.

John/Hein, I note that the consoles were moved earlier in the year and are a significant distance from the servers. I am wondering whether cabling is an issue. So some investigation is required here. A simple ping test will get things going.

Based on your responses I should be able to knock up a meaningful checklist for next time this occurs (typically around 2am - ouch).

I'll leave this thread open for now. And advise if/when I get any updates.

[abrsvc]

Please keep in mond that if there is a broken wire within the cable conencting to the serial console, it will make a wonderful antenna that can cause any number of interesting characters to generated.

It seems that this kind of problem exists just like many years ago although with less frequence. I still recall a VAX getting completely hosed everytime an elevator moved...

Dan

[cnb]

Tim,

It depends upon the system hardware model platform.

If the system is configured to directing OPA0: to the MP, they have a physical reset button which *usually* will clear these types of issues without affecting the OS.

See this document:

http://h71000.www7.hp.com/doc/83final/ba322_90076/6676pro_005.html

HTH

Rgds,

[Steven Schweda]

> [...] a significant distance [...]

What would that be in some standard unit or
other?

My dim recollection is that RS232 is supposed
to work as far as 50 feet. Longer can work,
of course, depending on the actual hardware,
but N*spec tends to cause problems for large
N.

An oscilloscope can help to resolve doubts
about signal integrity in situations where
it's in doubt.

[tim lloyd_1]

Hi All, I am closing this thread now as we have identified a boring old software problem.

FYI I have narrowed it down to either an operation to delete virtual address space or to rename the file connected to the VA. One of these operations is locking up a process which controls the console.

There is still plenty of digging to do as there are some imponderables (like how come a disk/memory problem occurred simultaneously on 2 servers?).

So I may be solicting information on a different topic in the near future :)

Thanks All

[tim lloyd_1]

I have identified a problem not related to this thread.