LEDs in the front panel stuck after a few days

[Flavio Villanustre]

After a few days of running fine, the LEDs in the front panel got stuck now (probably that bug on server11 that other people referenced). An strace on that process shows it executing a pselect() call in a tight loop and timing out each time, in case it helps troubleshoot it:

 pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

pselect6(1024, [8 9], NULL, NULL, {tv_sec=0, tv_nsec=2000000}, NULL) = 0 (Timeout)

A pidp11 service restart fixed the issue, of course.

Best,

Flavio

[Neal G.]

To be certain this is the same issue; when it happens again check the operation of the rotary switches.

For me only the data and address LEDs freeze. The LEDs associated with the rotary switches continue to operate correctly, responding to switch position changes as expected.

- Neal G.

[Flavio Villanustre]

Thanks for the tip! I'll surely do.

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-11/b4b63b90-5c29-4974-b960-62b4bd0f0867n%40googlegroups.com.

[Johnny Billquist]

I guess the error/bug is on the client side sending data to the server.
Rotary switches as well as the LEDs related to that is completely
handled on the server side.

Johnny

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com

> <mailto:pidp-11+u...@googlegroups.com>.

> To view this discussion visit

> https://groups.google.com/d/msgid/pidp-11/b4b63b90-5c29-4974-b960-62b4bd0f0867n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/b4b63b90-5c29-4974-b960-62b4bd0f0867n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol

[Adam Thornton]

Both my -10 and my -11 get into this state sometimes (every few weeks, seems like).  The simulator's still fine, but the panel's frozen.  Never tried the rotary switch.  I just sigh and take it as an opportunity to update the underlying OS and restart the machine.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-11/4b831e7e-6970-450c-918d-2a542f0a2512%40softjar.se.

[Steve Platt]

Good timing.  The lights on my PiDP-11 are frozen.  The rotary switch lights still work, though.

I'll play a little more with it tomorrow.

Steve

To view this discussion visit https://groups.google.com/d/msgid/pidp-11/CAP2nic2xFEMA1uBegeBxu%3Dpc_11zi7kv3axGgZZr-T5Eo2iPRA%40mail.gmail.com.

[oscarv]

Aha! Bug!

I squashed this one before, but it seems the recent versions of the /opt/pidp11/bin/server11 program reverted to their old bad ways.

The problem, I think (I'm eager for feedback) was a race condition in the buffer that holds the LED status data. (The rotary LEDs are not involved here, so they keep working).

Simh sends a constant stream of LED panel status data to the pidp11 server program. It can choke on the stream, but the odd thing was - only under very specific conditions (I've never seen it myself alas).

The solution:

1. copy the attached historybuffer.c and historybuffer.h to /opt/pidp11/src/07.0_blinkenlight_api/

2. cd /opt/pidp11/src

3. sudo ./makeserver.sh

This will recompile the server11 program (the front panel driver program, client11 is the actual simh simulator) and hopefully, be the solution.

Please let me know if I am right! Then I will update the 'official' install file.

Kind regards,

Oscar.

[George Harris]

That worked Oscar.