Scheduled tasks failing

[darkassassin07 652]

 My scheduled tasks for checking for server updates, and calculating user statistics keep failing. They work fine when triggered manually but i keep seeing these errors with the scheduled tasks (triggered daily).

 

Most recent log attached.

[darkassassin07 652]

Ffs, NOW the most recent log attached.

 

Using this site exclusively from mobile is a huge pain in the ass... Especially if you have to move a remote file or two around to attach. I ran out of memory and had to rewrite this thread three times. (no mem=chrome closing / refreshing pages) 

server-63640598400.txt

Edited September 10, 2017 by darkassassin07

[pju 10]

Had the same problem with Statistics and Server Configuration Backup.

Did a uninstall, restart, install, restart - and that seems to have fixed the problem.

 

Still have a problem with XML Metadata plugin, which runs fine but doesn't do anything.

Seems like this server release has changed something deep in the engine room.

[darkassassin07 652]

Well, I have just re-installed the statistics plugin so I'll report back if I have any more issues with that.

 

Check for updates however is part of the main server and there's no way in hell im reinstalling the whole server if i don't really need to...

[Luke 42080]

@@darkassassin07, it looks like your machine is having trouble connecting to the internet to check for updates. I would suggest giving it a try again later and see if anything changes.

[darkassassin07 652]

Hmm... Thats really odd, the server machine is connected via ethernet and i havent had any power bumps or connectivity issues with any other devices on the same network.

 

Could this have something to do with the server sleeping when not in use and accepting WoL?

Other than that i cant think of anything else that has changed recently

[Luke 42080]

It's possible the network connection went into low power mode during the idle time. In any event, you can also just go to the plugin catalog and update your plugins manually.

[darkassassin07 652]

Well, i have unchecked "allow the computer to turn off this device to save power" in my network adapters properties. We will see if that fixes the issue.

[darkassassin07 652]

Welp, not allowing the computer to shut off the network adapter greys out the wake on lan options and inevitably disables WoL completely (though still allowing the comp to sleep, which shuts off the adapter...). This is simply not going to work. (i require WoL)

 

Any other ideas i could try?

 

Im not sure where to go from here. The server machine is always on, just asleep now and again, with a hardwired ethernet connection that isn't showing any stability issues.

 

How do scheduled tasks react to being triggered while the server machine is asleep? (via auto triggers like daily) will they wake the comp, be skipped completely, or run next time the server is awake?

Is it possible they are being triggered while the comp is in the process of waking up and thats causing an issue?

Edited September 11, 2017 by darkassassin07

[Luke 42080]

You can just ignore the http request errors if you know they are related to the network card powering down.

 

Then see my second sentence in post #7.

 

 

 

How do scheduled tasks react to being triggered while the server machine is asleep? (via auto triggers like daily) will they wake the comp, be skipped completely, or run next time the server is awake? 

 

They will run the next time it is awake.

[darkassassin07 652]

Statistics I'm not too worried about, i tend to trigger that manually befor viewing the stats anyway, however the server checking for and applying updates id really like to be automated.

 

With this problem generating errors/notifications constantly I cant tell if there is a diffrent/actual problem ever unless I comb through the notifications that this would train me to ignore

 

 

 

 

Is it possible they are being triggered while the comp is in the process of waking up and thats causing an issue?

[Luke 42080]

Again see my second sentence in post #7.

[darkassassin07 652]

Statistics I'm not too worried about, i tend to trigger that manually befor viewing the stats anyway, however the server checking for and applying updates id really like to be automated.

 

With this problem generating errors/notifications constantly I cant tell if there is a diffrent/actual problem ever unless I comb through the notifications that this would train me to ignore

 

 

 

 

 

Jesus christ luke did you even read this?

Plugins I am aware I can trigger manually, I said I was fine with that as i tend to manually trigger them anyway.

 

SERVER UPDATING HOWEVER I would like to be automated, ignoring this and manually triggering it is not automated.

 

 

"just ignore it" has got to be the shittiest support I have ever had, short if denying its happening at all.

 

Can I have some actual support for solving this problem please?

Edited September 11, 2017 by darkassassin07

[Luke 42080]

Sorry, the reason I kept saying that repeatedly is because it didn't seem like you were responding to that.

 

Unfortunately I can't tell you why your own machine is unable to communicate with github, which is where we host the update files. It could be something related to your network card sleeping, it could also be a random outage from Github (not common, but does happen).

[Happy2Play 9782]

You could change the scheduled task triggers to when you know your network is available.  Click on the schedule task to bring up trigger options.

[Luke 42080]

Yup that's great advice.

[darkassassin07 652]

Thats the thing, my network is never down as far as i can tell. I have zero problems with any other service/device using the same connection to achieve internet access. I cant find any evidence there is anything wrong with the network it self at all.

 

That leaves possible problems with the server machine, but I dont know where to look or even begin troubleshooting that...

 

The only idea i have is to set up a task scheduler task to run a bat that pings github.com every say 10min(when the server is awake) and write to a log if it was successful or not. Just to see when and if im having connectivity issues.

Problem there is im not sure how to write that script. Im a total noob with command line. I can do some basic java, that's about it.

[pju 10]

You could try a poor mans solution as below.

CMD scripts are neither nice nor readable ;-)

The script pings github.com and counts the lines containg "TTL=" - which only occurs on response.

Unfortunately also succeeds with very long reponsetime where the github api may have given up.

So, I'm not sure how much you'll learn...

 

 

@echo off
setlocal

 

 

set LOGFILE=E:\temp\ping_log.txt
set TESTSITE=github.com

 

 

REM Get a timestamp
for /F "usebackq tokens=1,2 delims==" %%i in (`wmic os get LocalDateTime /VALUE 2^>NUL` ) do if '.%%i.'=='.LocalDateTime.' set TS=%%j
REM DO some substring'ing
set TS=%TS:~2,2%-%TS:~4,2%-%TS:~6,2%_%TS:~8,2%:%TS:~10,2%:%TS:~12,2%

 

REM Count the number of lines containing TTL=
set /A NBR=0
FOR /F %%i in ('ping %TESTSITE% ^| findstr TTL==') DO SET /A NBR=NBR+1

 

REM A default ping send 4 IMCP packets, so 4 is fine
IF %NBR% EQU 4 (
   echo %TS% OK, %NBR% received >> %LOGFILE%
) ELSE (
   echo %TS% BAD, %NBR% received >> %LOGFILE%
)