5 minutes ago, itimpi said:

 

Thanks - that allowed me to see the error and will issue a fix.    Has a || (logical OR) when I needed | (binary OR).   The issue was specific to the case where backup was running.

 

Awesome!  Glad to help.

4 hours ago, Mat W said:

 

Awesome!  Glad to help.

Fix now available.   Please confirm that it works as expected.

4 minutes ago, itimpi said:

Fix now available.   Please confirm that it works as expected.

 

Update successful.  Hopefully I've got enough parity check remaining to naturally make it passed the backup tomorrow morning.  I'll let you know how it does.

Unfortunately, my current parity check finished 10 minutes before the nightly appdata backup started.  I don't know if it would have resumed properly.

19 minutes ago, Mat W said:

Unfortunately, my current parity check finished 10 minutes before the nightly appdata backup started.  I don't know if it would have resumed properly.

Fair enough.

 

I guess it is just a case of waiting until either someone else confirms it works for them, or at least no new reports of a problem like this.

Running 2023.09.06 (and the others this month) I am getting notifications when my config does not have such options enabled (afaik!) - config:

 

# cat /boot/config/plugins/parity.check.tuning/parity.check.tuning.cfg 
parityTuningScheduled="0"
parityTuningManual="0"
parityTuningAutomatic="0"
parityTuningFrequency="0"
parityTuningResumeCustom="5,15,25,35,45,55 * * * *"
parityTuningResumeDay="0"
parityTuningResumeHour="0"
parityTuningResumeMinute="15"
parityTuningPauseCustom="0,10,20,30,40,50 * * * *"
parityTuningPauseDay="0"
parityTuningPauseHour="0"
parityTuningPauseMinute="10"
parityTuningNotify="0"
parityTuningRecon="0"
parityTuningClear="0"
parityTuningRestart="1"
parityTuningMover="1"
parityTuningCABackup="1"
parityTuningHeat="0"
parityTuningHeatHigh="3"
parityTuningHeatLow="8"
parityTuningHeatNotify="1"
parityTuningHeatShutdown="0"
parityTuningHeatCritical="2"
parityTuningHeatTooLong="30"
parityTuningLogging="0"
parityTuningLogTarget="0"
parityTuningMonitorDefault="17"
parityTuningMonitorHeat="7"
parityTuningMonitorBusy="6"
parityTuningMonitorPartial="1"
parityProblemType="sector"
parityProblemStartSector="0"
parityProblemStartPercent="0"
parityProblemEndSector="100"
parityProblemEndPercent="0"
parityProblemCorrect="0"
parityTuningBackup="1"
parityTuningShutdown="0"

 

GUI:

 

Yet I'm getting 'mover no longer running' and 'mover running' notifications.

 

Any ideas?

 

Thanks!

E

40 minutes ago, estrim said:

Running 2023.09.06 (and the others this month) I am getting notifications when my config does not have such options enabled (afaik!) - config:

 

# cat /boot/config/plugins/parity.check.tuning/parity.check.tuning.cfg 
parityTuningScheduled="0"
parityTuningManual="0"
parityTuningAutomatic="0"
parityTuningFrequency="0"
parityTuningResumeCustom="5,15,25,35,45,55 * * * *"
parityTuningResumeDay="0"
parityTuningResumeHour="0"
parityTuningResumeMinute="15"
parityTuningPauseCustom="0,10,20,30,40,50 * * * *"
parityTuningPauseDay="0"
parityTuningPauseHour="0"
parityTuningPauseMinute="10"
parityTuningNotify="0"
parityTuningRecon="0"
parityTuningClear="0"
parityTuningRestart="1"
parityTuningMover="1"
parityTuningCABackup="1"
parityTuningHeat="0"
parityTuningHeatHigh="3"
parityTuningHeatLow="8"
parityTuningHeatNotify="1"
parityTuningHeatShutdown="0"
parityTuningHeatCritical="2"
parityTuningHeatTooLong="30"
parityTuningLogging="0"
parityTuningLogTarget="0"
parityTuningMonitorDefault="17"
parityTuningMonitorHeat="7"
parityTuningMonitorBusy="6"
parityTuningMonitorPartial="1"
parityProblemType="sector"
parityProblemStartSector="0"
parityProblemStartPercent="0"
parityProblemEndSector="100"
parityProblemEndPercent="0"
parityProblemCorrect="0"
parityTuningBackup="1"
parityTuningShutdown="0"

 

GUI:

 

Yet I'm getting 'mover no longer running' and 'mover running' notifications.

 

Any ideas?

 

Thanks!

E

 

That is currently intended behaviour (I think I discussed it earlier) with the recent releases if it happens during an array operation.  It is only the pause/resume related ones that are currently being controlled by the plugin's settings.   I thought this could well be useful information for most users so they can see if those tasks running coincides with possible performance issues.  Making it unconditional for the time being will mean I get some feedback to see how many want it vs. do not want it.

 

The questions that arise are:

• Is having this notification useful even when not using the pause/resume options for background tasks.
• Should there be a setting to control it?  I could possibly add it as one after the options to pause/resume for background tasks in the current plugin settings with the default being to set it on.

5 hours ago, itimpi said:

The questions that arise are:

• Is having this notification useful even when not using the pause/resume options for background tasks.
• Should there be a setting to control it?  I could possibly add it as one after the options to pause/resume for background tasks in the current plugin settings with the default being to set it on.

 

Apologies, that's my fault for not reading properly!

 

My answer to this seems to be no&yes - whilst there are of course many cases, in my particular case where I know a parity operation already kicked off, having notifications that my mover caused it to stop/start every hour when the mover is chugging along is not useful. 

 

Hope that's at least one perspective!

Thanks again

 

Hi, just to report that on 2023.09.06 version, I still get false "Unclean shutdown detected" notification every time I stop the array. Starting the array again reports that the parity is valid and does not trigger a parity check.

how i can setup check parity in Unraid - '0 0 1 */3 *' every 3 month on 1 day  0:00?

if i set custom i can select months , but no day of month select , only weeks :

i can manually add (0 0 1 */3 *) to parity-check.cron , but not shure it will work it not shown right in webGUI ...

 

 

 

2 hours ago, zellleonhart said:

Hi, just to report that on 2023.09.06 version, I still get false "Unclean shutdown detected" notification every time I stop the array. Starting the array again reports that the parity is valid and does not trigger a parity check.

Have you rebooted since doing the update?   Just asking as this might clear up some ‘debris’ state information by an earlier release.

 

if you DO get after rebooting then to determine why I would need the testing mode active in the plugin settings and then diagnostics taken after the reboot and when this notification has been displayed.

2 minutes ago, Masterwishx said:

how i can setup check parity in Unraid - '0 0 1 */3 *' every 3 month on 1 day  0:00?

if i set custom i can select months , but no day of month select , only weeks :

i can manually add (0 0 1 */3 *) to parity-check.cron , but not shure it will work it not shown right in webGUI ...

 

 

 

No idea I am afraid.   That part of the scheduling is handled by the core Unraid system and not by the plugin.   Maybe you need to raise a feature request/bug about this?

2 minutes ago, itimpi said:

No idea I am afraid.   That part of the scheduling is handled by the core Unraid system and not by the plugin.   Maybe you need to raise a feature request/bug about this?

 

Yes i know , sorry , i will try it ...

On 6.12.4 I keep getting notifications from 'parity check tuning of unclean shutdown detected'... but I'm not sure if these are correct. How can I double check?

Edited September 9, 20232 yr by dopeytree

9 hours ago, dopeytree said:

On 6.12.4 I keep getting notifications from 'parity check tuning of unclean shutdown detected'... but I'm not sure if these are correct. How can I double check?

 If you are on the latest release of the plugin (2023.09.06) then there is a good chance it IS genuine but it would be nice to make sure.   If you enable the 'testing' mode of logging in the plugin's settings and once this has happened post a copy of you system's diagnostics I can check whether it is genuine.   It should also be present in the Unraid syslog but you have to find it there.

15 hours ago, itimpi said:

 If you are on the latest release of the plugin (2023.09.06) then there is a good chance it IS genuine but it would be nice to make sure.   If you enable the 'testing' mode of logging in the plugin's settings and once this has happened post a copy of you system's diagnostics I can check whether it is genuine.   It should also be present in the Unraid syslog but you have to find it there.

 

I did get one also and I know that I have not had an unclean shutdown. And I am on the latest release of this plugin.

 

5 hours ago, Ruato said:

did get one also and I know that I have not had an unclean shutdown.

How are you sure?    Lots of people get them without realizing it as it can happen when you press the Shutdown or Reboot button if a timeout kicks in or if for any reason or if Unraid cannot write to the flash drive during these sequences.   About the only way to be certain you have not had one is to successfully Stop the array before attempting to Shutdown or Reboot.

 

At the moment I am waiting for someone on the latest release.to be able to give me some diagnostics (with the plugin ‘testing’ logging mode active)  to confirm that there is still a problem with the Unclean shutdown detection.  I have not managed to spuriously get one in my recent testing when actively trying various scenarios to force iit.

2 hours ago, itimpi said:

How are you sure?    Lots of people get them without realizing it as it can happen when you press the Shutdown or Reboot button if a timeout kicks in or if for any reason or if Unraid cannot write to the flash drive during these sequences.   About the only way to be certain you have not had one is to successfully Stop the array before attempting to Shutdown or Reboot.

 

At the moment I am waiting for someone on the latest release.to be able to give me some diagnostics (with the plugin ‘testing’ logging mode active)  to confirm that there is still a problem with the Unclean shutdown detection.  I have not managed to spuriously get one in my recent testing when actively trying various scenarios to force iit.

 

I have increased the shutdown timeouts a lot and they are 5-10 minutes. Additionally, I followed the latest shutdown from the monitor and it _seemed_ good and went fast. So, at least it wasn't the timeouts.

 

Is there a way to check from the system whether the latest shutdown was clean or not? And is it so that the plugin only checks the latest shutdown? That is, it is not checking if there was an unclean shutdown before that.

 

I am travelling this week so I cannot examine this more now. Hopefully, you get diagnostics before that from someone else.

2 hours ago, Ruato said:

Is there a way to check from the system whether the latest shutdown was clean or not? And is it so that the plugin only checks the latest shutdown? That is, it is not checking if there was an unclean shutdown before that.

 

Not easily from a running system that I know of.    The plugin now detects an unclean shutdown by the presence of the 'config/forcesysc' file on the flash drive during the boot phases as that is created by Unraid when it starts the array after a clean shutdown and deletes it when it successfully stops it.  If you can get at the USB drive while it is not in Unraid you can check for that but by the time you start booting it is probably too late.

Diags attached. 

 

I noticed if it is a 'real' unclean shutdown the system will start a check. VS this message just occurs all the time. I wondered if it was from a historic unclean shutdown or something?

 

I have just completed a parity check after the server crashed a few days ago.

moulin-rouge-diagnostics-20230911-1334.zip

Edited September 11, 20232 yr by dopeytree

3 hours ago, dopeytree said:

Diags attached. 

 

I noticed if it is a 'real' unclean shutdown the system will start a check. VS this message just occurs all the time. I wondered if it was from a historic unclean shutdown or something?

 

I have just completed a parity check after the server crashed a few days ago.

moulin-rouge-diagnostics-20230911-1334.zip 190.52 kB · 0 downloads

I cannot see any sign in the posted diagnostics that the notification about unclean shutdown was ever sent😒.  Are you sure you got it this time around?

Something's triggering it but I'm not worried about it. More curious if anyone else has it.

There should be an entry in the syslog every time such a notification is generated and I did not see any. Which is very strange.

 

Any chance of enabling the Testing mode of logging in the plugin settings; waiting till you get more occurrences of the notification; and then posting your diagnostics.   That would allow me to confirm what is happening under the covers.
 

 

Diags with the debug logging enabled. I stopped the array to cause the notification.

 

moulin-rouge-diagnostics-20230912-0129.zip

Edited September 12, 20232 yr by dopeytree

9 hours ago, dopeytree said:

Diags with the debug logging enabled. I stopped the array to cause the notification.

 

moulin-rouge-diagnostics-20230912-0129.zip 175.87 kB · 0 downloads

 

At the moment I am very confused.   According to the syslog in the diagnostics the plugin has never written a notification (it writes a syslog entry every time before trying to generate a notification).   It is almost as if there is something stuck at the Unraid level that is causing this repeated notification but I cannot think what that might be.  

 

BTW:  You had the 'Debug' level of logging enabled - to do a more thorough analysis I would need the 'Testing' level of logging enabled.