January 24, 20188 yr During the RC's I had pointed this out but it seemed to fall on deaf ears. After about 10 days of up-time the kernel logs an error. Jan 24 11:40:20 Tower kernel: traps: emhttpd[10882] trap divide error ip:419f15 sp:1459625d4e00 error:0 in emhttpd[400000+26000] Description: After this the GUI appears to still function for most things but you can no longer check for plugin updates, you can no longer properly shutdown(it won't unmount the disks properly or do a clean shutdown), and functionality gradually degrades. Can someone please look into this? During the RCs I eventually abandoned them and went back to 6.3 and never had this issue again(ran for about 35 days) till I went back up to 6.4 again after it was released. I don't get what 10 days has to do with it but it seems to always happen when uptime crosses it. Another thing is I believe it started happening in the RC's when the GUI was made faster. I've attached the diagnostics. I will say you will see where its switching to battery at times. My UPS seems to be very sensitive to power draw as its a pure sine curve and its just when the printer starts up. Could lime-tech or someone look please? How to Reproduce: Keep server up for around 10 days and check plugins page or check for OS update every day or two. Eventually it will just time out and not work anymore. Expected Results: To have fully functioning GUI and be able to cleanly shutdown Actual Results: Cannot check for updates for plugs, OS, appears mover doesn't function, and none of the buttons for shutting down function either. Edited February 9, 20188 yr by Jerky_san
January 29, 20188 yr Yes I reported this throughout the RC 6.4 as well, its still present in final 6.4.0 for me too. After a bunch of days uptime, some things stop working as you mentioned. Edited January 29, 20188 yr by Inssomniak
January 30, 20188 yr Thanks for the reports. Looking at both syslogs, same crash happening. We have not been able to reproduce.
February 2, 20188 yr Author On 1/30/2018 at 4:53 PM, limetech said: Thanks for the reports. Looking at both syslogs, same crash happening. We have not been able to reproduce. Thank you for responding and looking into it.
February 3, 20188 yr I'm curious about something in your logs. May or may not be related, but this entry should definitely not be there: Jan 23 00:00:01 Tower crond[1641]: exit status 1 from user root /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php >dev/null 2>&1 Can you log in to your server and run this command and post the output /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php
February 3, 20188 yr Author 42 minutes ago, Squid said: I'm curious about something in your logs. May or may not be related, but this entry should definitely not be there: Jan 23 00:00:01 Tower crond[1641]: exit status 1 from user root /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php >dev/null 2>&1 Can you log in to your server and run this command and post the output /usr/local/emhttp/plugins/ca.update.applications/scripts/updateApplications.php Tbh I noticed the same thing and I actually removed that and am testing to see if it crashes again.. I removed it about 8 and a half days ago and still haven't gotten my problem. If it doesn't come back it may be something tied to it but I don't understand why more wouldn't have the problem if it was that.
February 3, 20188 yr It's not the script that would've caused your problem (it ran for days returning the error code before emhttp crashed) But whatever error that it was throwing would maybe point to the cause.Sent from my SM-T560NU using Tapatalk
February 3, 20188 yr Author 7 minutes ago, Squid said: It's not the script that would've caused your problem (it ran for days returning the error code before emhttp crashed) But whatever error that it was throwing would maybe point to the cause. Sent from my SM-T560NU using Tapatalk o-o; guess I wish I wouldn't of uninstalled it then. After I go past 10-11 days without a crash I'll put it back in.
February 5, 20188 yr Author I have made it past the 10 day mark. I am at 10 days 16 hours. So I will see how far I can go . Only thing I did was uninstall the CA update applications plugin.
February 6, 20188 yr Author Day 11 hour 17.. Don't believe I've ever made it this far in 6.4 since the bug occurred. I do wonder how CA update applications plugin would affect it as it only seems to help update the CA system itself. If I can make it to day 20 I am calling it fixed though I question if I can go without the CA update app for long term.
February 10, 20188 yr Author Crashed today about 2 hours ago. 15 days 4 hours.. So made it further at least.. Feb 9 17:43:25 Tower kernel: traps: emhttpd[11311] trap divide error ip:419f15 sp:1469ba8d5e00 error:0 in emhttpd[400000+26000]
February 13, 20188 yr Yes I uninstalled the CA update applications plugin too, in hopes of fixing this, but it crashed within a day.. Ive remove all CA plugins, so far 5 days.
February 14, 20188 yr If you are still on 6.4.0, please upgrade to 6.4.1. Then upload a fresh diagnostics so we can look for clues. Also read this thread for things to watch out for:
February 20, 20188 yr Author On 2/13/2018 at 7:05 PM, ljm42 said: If you are still on 6.4.0, please upgrade to 6.4.1. Then upload a fresh diagnostics so we can look for clues. Also read this thread for things to watch out for: Sorry I just saw this. Next time I lock up I'll get you the diagnostic. Happened yesterday but didn't see this until after I had already restarted.
February 21, 20188 yr 9 hours ago, Jerky_san said: Sorry I just saw this. Next time I lock up I'll get you the diagnostic. Happened yesterday but didn't see this until after I had already restarted. Once the system locks up I'm not sure that you'll be able to post the diags. Go ahead and upload one now and I'll see if anything stands out. If it is crashing that regularly, you can install the Fix Common Problems plugin and put it in troubleshooting mode.
February 21, 20188 yr Author 28 minutes ago, ljm42 said: Once the system locks up I'm not sure that you'll be able to post the diags. Go ahead and upload one now and I'll see if anything stands out. If it is crashing that regularly, you can install the Fix Common Problems plugin and put it in troubleshooting mode. I won't freeze just all the buttons and other things on the GUI stop functioning. Can't check for updates for plugs, run mover, save changes, and other things. I have at least found a way around not causing my system to be dirty restarted and that is to manually attempt to start the emhttp process. I know your technically not supposed to be able to do it. But by running below command will allow me to restart the server with the buttons/command line. I'll try to post a diagnostic a little later tonight though. I can run into the server via ssh and run the collect diagnostic after emhttpd crashes. /usr/local/sbin/emhttpd & disown
March 1, 20188 yr Author Happened again today. Well by the log last night but still. Here is the diagnostic file. tower-diagnostics-20180301-1614.zip
March 18, 20188 yr Yes I still suffer from this.. Basic install, with Plex Docker. 6.4.1 Edited March 18, 20188 yr by Inssomniak
April 1, 20188 yr Author I updated to 6.5 a 14 days ago and I've not had this occur yet. I guess the jury is still out but it used to happen consistently at 10 days..
April 3, 20188 yr Unfortunately I'm having the same problem as described at the top of this thread. Diagnostics from latest attached. tower-diagnostics-20180403-0919.zip
April 3, 20188 yr Author 6 hours ago, armandsspele said: Unfortunately I'm having the same problem as described at the top of this thread. Diagnostics from latest attached. tower-diagnostics-20180403-0919.zip I would try upgrading to 6.5.. I'm 16 days in almost now and still haven't had my issue re-occur so they might of fixed something in 6.5 or updated something and it fixed the issue without them realizing it.
April 3, 20188 yr 8 hours ago, Jerky_san said: I would try upgrading to 6.5.. I'm 16 days in almost now and still haven't had my issue re-occur so they might of fixed something in 6.5 or updated something and it fixed the issue without them realizing it. Thanks for the suggestion, but I'm already on 6.5. And it didn't help. Now I'm taking a very methodical approach of uninstalling/stopping stuff and see what would help. I've uninstalled the Community Applications (and related -- Backup, Auto-upgrade, etc.) plugins and stopped Docker. And I don't have VM. So, currently just running a "virgin" NAS/File Server. So far, 4h 44m and the emhttp hasn't crashed. Once it stays OK for a day or so, I will start adding things back (starting with turning on docker).
April 4, 20188 yr Unfortunately even after uninstalling all plugins and stopping docker (and not running Virtual Machines either), I've still got that "....emhttpd[...] trap divide error". Perhaps there is some debug mode that emhttpd can be put it to further help troubleshooting? Or some additional logging that can be turned on? Or any other suggestion?
April 4, 20188 yr 6 hours ago, armandsspele said: Unfortunately even after uninstalling all plugins and stopping docker (and not running Virtual Machines either), I've still got that "....emhttpd[...] trap divide error". Perhaps there is some debug mode that emhttpd can be put it to further help troubleshooting? Or some additional logging that can be turned on? Or any other suggestion? There is a change in next release that might solve this. Is your server mostly idle when it happens?
Archived
This topic is now archived and is closed to further replies.