Re: [BLAST_SHIFTS] Shift summary 09/22/2003 F (20-24)

From: Taylan Akdogan (akdogan@mit.edu)
Date: Tue Sep 23 2003 - 20:37:03 EDT


Thanks Chris, Niki. Yes, that explains it.

Here is what was going on:

The Compton controller was desperately -trying- to send data to
ET when it was inaccessible due to whatever the reason was,
meanwhile, the DAQ histogrammer (on IOC) was waiting a "heart
beat" from the controller, to make sure that everything was
alright, which never came before the watchdog-timeout, so the IOC
decided to reboot itself assuming that it was inaccessible.

It is a safeguard for ANY possible problems that saves us from a
"manual" reset of the IOC which may require a hall access. I am
going to address this inconvenience during the next shutdown
period without any sacrifice to the robustness. And, I am glad to
hear that it is not an "Access Fault" problem which WAS hardware
related.

Meanwhile, the correct approach should be stopping the Compton
run whenever ET is not running; and/or when CODA is stopped.

Thanks,
Taylan

On Tue, 23 Sep 2003, Taylan Akdogan wrote:
>
> Hi Nikolas, Chris,
>
> I noticed the following remark in the last e-log:
>
> "The compton was kind enough to reboot itself every now and then
> but other than that it worked fine."
>
> Can you be more specific? Was it an "Access Fault" or "Watchdog",
> or something else? Was anything unusual going on just before the
> reboot especially about the ET?
>
> Although, I can't do much right now --I'm in Los Alamos--, I'm
> curious...
>
> Thanks,
> Taylan

On Tue, 23 Sep 2003, Chris Crawford wrote:

> hi tylan,
> nik wrote that, but my guess is that the compton et program crashed
> because they left compton on after the run was done. thanks for
> checking up on it.
> --chris

On Tue, 23 Sep 2003, Nikolas C Meitanis wrote:

> Since most of the time I was "playing tetris" with other alarms, I cannot
> be sure 100% for every reboot but the couple of times I caught it it was
> due to the watchdog. It was probably related to the ET being unaccessible
> due to ROC and other problems.
> Hope this helps.
> n

---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---
Taylan Akdogan Massachusetts Institute of Technology
akdogan@mit.edu Department of Physics
---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---



This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST