Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
Locked SECTIONS TABLE NEW BUG
|
Hey guys,
I just noticed a random bug, I've had my layout working well for some time and started to do some experimentation with new transits and dispatcher. I went to fix an error i made a while ago that never affected me. The error was i had a sections stopping sensor set as the direction sensor. I never noticed as i hadnt used this section in automation yet. I want to refctify the mistake (set directon sensor to nothing, set forward stopping sensor to LS6) and press update.. the button clicks, but it does nothing. I close it, re-open it and the direction sensor has gone to nothing, the stop sensor is still nothing, and the table of entry points as disapeared. The one block in the sesion (14) is still in the top table. I have to clear blocks and then re-add the block 14 for the table of entry points, and travel directions auto-set themselves. I update this, without touching any sensor info, and it saves it and closes the window and i acan open and close/update this window multiple times with no issues.. but as soon as I set a stopping sensor (to any sensor i found) it clears the table of entry points and all sensor information. I'm going to backdate JMRI and see if this happens on the current stalbe release (I run off the lastest Jenkins updates due to the latest amaizng work on Dispatcher in the recent months) I havfe deffinitley done this task without this issue probably hunfreds of times and the end of last year, hence why I feel it's a new bug. I will post an update when I have tested on stable release. Thanks guys! Mitch -- Thanks Mitch |
|
Mitch,
toggle quoted message
Show quoted text
Are there any messages in the JMRI console when the unexpected sensor behavior occurs? Dave Sand On Sep 15, 2019, at 3:11 AM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
Hi Dave,
as soon as I hit "update" to the section, and have the message log open, i get a long-winded error. It also throws up a window with the same error. I can provide a screenshot, however the first error is: java.lang.Error: Cannot call invokeAndWait from event dispatcher thread followed by about 20 others. most of the errors finish with (unknown source). If you'd like a sccreen shot, please let me know the best place to send/put that. -- Thanks Mitch |
|
The plot thickens!
The Discover Signal mast does not seem to work anymore, and if I have the message log open, it throws an error and crashes. Same error as above. I have a feeling that my file is corrupt. Howver JMRI itself feels buggy, adding signal logic manually instead of automatically even though it was able to in the past. I have to close the windows and re-open them for it to refresh new logic, if it doesnt crash. -- Thanks Mitch |
|
Mitch,
toggle quoted message
Show quoted text
Screen shots are uploaded to the group¡¯s Photo area. If you upload your panel xml file to the Files ProblemsBeingWorkedOn Folder, we can see if there is a data issue. Dave Sand On Sep 21, 2019, at 9:54 PM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
Mitch,
toggle quoted message
Show quoted text
I loaded your panel, deleted all of the signal mast logic and did the automatic generate successfully. The screenshot is curious but does not show the environment. If you have the JMRI console open, click on Copy to clipboard and paste into a reply. This way we get a better view from when JMRI starts and leading up to the error. Dave Sand On Sep 21, 2019, at 10:26 PM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
Mitch,
toggle quoted message
Show quoted text
I see that you have a custom signal system: BR-2003-WE. You will also need to create a zip file and upload that to your problems folder. Dave Sand On Sep 21, 2019, at 10:57 PM, Dave Sand <ds@...> wrote: |
|
Dave,
The auto-generate does work, however S01 should not be able to get to S04, and S01 cant get to S08, but can get to S07, the error being? jmri.JmriException: Path not valid, prtecting block is nill. Protecting block : 08D (true), not connexted to 41. Layout bLock names: I have to manually do this one. I think theres a few others that have this issue, too, but it would mean scouring through what SHOULD be happening vs what IS happening, which is what im in the process of now. In regards to the log error, the whole program freezes so I cant access the log once the actual pop-up comes up with the error. I will try and make the log scren as big as possible and screenshot again. -- Thanks Mitch |
|
Mitch,
toggle quoted message
Show quoted text
You can upload the session.log file. It is at the settings location/logs. See Help >> Locations. Dave Sand On Sep 21, 2019, at 11:28 PM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
Mitch,
toggle quoted message
Show quoted text
Don¡¯t worry about BR-2003-WE. The internal name for BR-2003 turns out to be BR-2003-WE. Dave Sand On Sep 21, 2019, at 11:48 PM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
Mitch,
toggle quoted message
Show quoted text
I was able to get the SML from S01 to S08 to work. If you have been making a lot of block changes, the block routing can get confused. The unique double crossover at LT2-3-4-5 and its blocks appear to have contributed. In the Block table menu, select Paths >> Delete Paths. Acknowledge the prompt. Save the panel xml file, stop and restart JMRI. This rebuilds the block paths. You may want to delete and rediscover the SML if you have been trying different things. Dave Sand On Sep 21, 2019, at 11:28 PM, Mitchell via Groups.Io <mitchell.scott93@...> wrote: |
|
okay so..
When I do the delete paths routine, save, reload, and generate signal pairs.. the S01 to S08 path works, SML shows the blocks used etc. I save the file, close JMRI, re-load and then check, and that is the ONLY SML that loses its blocks, and creates the null error as described before. Panel S01 also shows red when should be green for S08, proving the point. So delete paths works Create new paths works when retarted because the SML works when first created.. and as soon as the XML is saved, it stops working after a restart. -- Thanks Mitch |
Roger Merritt
to navigate to use esc to dismiss