From: Narayanan, Suresh <[email protected]>
Sent: Wednesday, August 25, 2021 10:04 AM
To: Zhang, Qingteng <[email protected]>; Yasukazu Nakaye <[email protected]>; Hiroki Yoshida <[email protected]>
Cc: Takuto Sakumura <[email protected]>; Yasutaka Sakuma <[email protected]>;
Kazuyuki Matsushita <[email protected]>; Dufresne, Eric <[email protected]>; Lang, Keenan C. <[email protected]>
Subject: Re: Rigaku/epics
Dear Nakaye-san, Hiroki-san,
Out of curiosity, I tried changing the low energy threshold and I did not see
the number of non-zero pixels increase.
I lowered threshold to as low as 0.5 and saw not a single pixel increase.
Question is, is the field exposed to the library?
Keenan,
Can you check what happens in epics when the field is updated?
Thanks.
Suresh
RE: Rigaku/epics
Yasukazu Nakaye <[email protected]>
Mon 8/30/2021 7:00 PM
To: Narayanan, Suresh <[email protected]>; Zhang, Qingteng <[email protected]>; Lang, Keenan C. <[email protected]>; Hiroki Yoshida <[email protected]>
Cc: Takuto Sakumura <[email protected]>; Yasutaka Sakuma <[email protected]>; Kazuyuki Matsushita <[email protected]>; Dufresne, Eric <[email protected]>
Dear Suresh,
Sorry about my late response, but Hashimoto-san and Yoshida-san already informed QZ with instruction how to recover from the current situation.
Hashimoto-san found the cause of the problem and now we are working to fix the problem to update the server.
With current version, it looks like, if you set the threshold below 4 keV, server cannot find the files for
calculation and fails subsequently.
Yasukazu
Error logs
Re: Rigaku/epics
Narayanan, Suresh <[email protected]>
Fri 8/27/2021 12:40 PM
To: Zhang, Qingteng <[email protected]>; Yasukazu Nakaye <[email protected]>; Lang, Keenan C. <[email protected]>; Hiroki Yoshida <[email protected]>
Cc: Takuto Sakumura <[email protected]>; Yasutaka Sakuma <[email protected]>; Kazuyuki Matsushita <[email protected]>; Dufresne, Eric <[email protected]>
Dear Nakaye-san,
It seems like the entire controls is very fragile. I was using yesterday in the slow mode and it worked well.
Now the IOC wont take any images and see the “error: 33619976”
If these things keep happening, then we cannot run user operations. There are experiments that will fully rely
on this and we cannot get away with running Lambda for example. If slow mode does not work, we can get away
with Lambda. But if the whole thing fails, then we have to cancel the user experiment.
Why does this error keep happening? Is there a troubleshooting process that can be documented that we can
go through in steps? Usually, power cycle fixes most things but it is not clear here what to do.
Keenan, do you know what this error is?
Suresh
Re: Rigaku/epics
Narayanan, Suresh <[email protected]>
Mon 8/30/2021 9:11 PM
To: Yasukazu Nakaye <[email protected]>; Zhang, Qingteng <[email protected]>; Lang, Keenan C. <[email protected]>; Hiroki Yoshida <[email protected]>
Cc: Takuto Sakumura <[email protected]>; Yasutaka Sakuma <[email protected]>;
Kazuyuki Matsushita <[email protected]>; Dufresne, Eric <[email protected]>
Dear Nakaye,
Regarding log files, it would be nice if the most recent file always has the same name (maybe use a
sym link to always point to the current file), will be a good practice to manage things better.
Also, a database of error code and what it means could save a lot of trouble.
Suresh |