June 7, 20179 yr I decided to give this its own post, although it has been mentioned in another post. Link at the bottom of this post. When accessing the UI at times I get a 502 Bad Gateway response from nginx. This relates to a reoccurring issue in syslog. Today I opened the UI and accessed the tabs randomly 16 times before this issue occurred, then after typing part of this post went to get diagnostics and it happened again Today I opened the UI and accessed the tabs randomly 16 times before this issue occurred, Jun 7 10:33:06 Tower php-fpm[11851]: [WARNING] [pool www] child 9740 exited on signal 7 (SIGBUS) after 2842.695768 seconds from start Jun 7 10:33:06 Tower php-fpm[11851]: [NOTICE] [pool www] child 32388 started Then after typing part of this post went to get diagnostics and it happened again Jun 7 10:35:53 Tower php-fpm[11851]: [WARNING] [pool www] child 16479 exited on signal 7 (SIGBUS) after 18786.486124 seconds from start Jun 7 10:35:53 Tower php-fpm[11851]: [NOTICE] [pool www] child 1173 started tower-diagnostics-20170607-1035.zip This has already been mentioned in the following post Edited June 7, 20179 yr by Tuftuf
June 7, 20179 yr Have you tried to clear browser's cache and remove cookies? See if that makes a difference. And - if possible - try another browser Edited June 7, 20179 yr by bonienl
June 7, 20179 yr Author I'm seeing the issue accessing the system from a MacBook in Safari and Chrome, Windows 10 VM Edge and Chrome and a separate corporate laptop running Edge and Chrome. Each occurrence relates to the error shown, I don't think it can be browser related. I will clear all cache etc on one of the machines just to be sure though. EDIT - After clearing cache it took over 30-40 clicks but it occurred. It's random when it happens, and always accessing the same URL again works. EDIT - Closed all other browsers, cleared cache again. Issue occurred within about 20 clicks. These tests performed with Chrome, avoiding Safari for main tests due to other issue seen with CPU stats. Edited June 7, 20179 yr by Tuftuf
June 7, 20179 yr Author I don't mind making the server accessible if you wanted, would need an IP address to allow access. But as far as reproducing goes I can't suggest any steps as it just happens. Plugins installed. -- Could to remove one or more if needed. Apache Http Server support - Port 8118 CA Backup / Restore Appdata CA Cleanup Appdata CA Docker Autostart Manager ca.update.applications Community Applications Diskspeed Test Dynamix SSD TRIM Dynamix System Statistics Dynamix System Temperature Dynamix webGui Fix Common Problems Nerd Tools Preclear Disks Ransomware Protection Recycle Bin Unassigned Devices unRAID DVB Edition Deployed ports from Docker binhex-plexpass (Static IP) unifi3478 6789 8080 8081 8443 8843 8880 minecraftpe 19132 tvheadend 9981 9982 tftp-server 69 netdata (Host, using 19999) Edited June 7, 20179 yr by Tuftuf
June 7, 20179 yr Author Rebooted into Safemode with GUI. The UI loaded in FireFox and I pressed Shares link, 502 Bad Gateway. I had not even started the Array yet. Edited June 7, 20179 yr by Tuftuf
June 7, 20179 yr I would also see this randomly under RC0. Haven't seen it yet under RC1/2Sent from my LG-D852 using Tapatalk
June 7, 20179 yr 7 hours ago, Tuftuf said: The UI loaded in FireFox and I pressed Shares link, 502 Bad Gateway. I had not even started the Array yet. Does anything show up in syslog? How about /var/log/nginx/error.log?
June 7, 20179 yr Author I have been checking error.log since RC1 and so far I've not seen anything written to it. The file is empty. Today I restated my system at 12:34 and did not access the box at all. Later I started my windows VM around 14:40, accessing the ui caused the error and appear to syslog again. Did not access the UI again and remained within windows. It's now 20:10 and the only error reported was 14:40. It's not much but based on that I need to be accessing the UI to cause the pool www restart error shown in syslog. Just now, restarted into safe mode with GUI, this time it took 9 minutes of clicking buttons. Accessing every tab on the settings, user page, tools and feedback etc. Almost thought it wasn't going to appear. The only two lines from syslog are the same as before, cannot find any other messages around the same time that are related. Jun 7 20:15:45 Tower php-fpm[2170]: [WARNING] [pool www] child 4840 exited on signal 7 (SIGBUS) after 129.155766 seconds from start Jun 7 20:15:45 Tower php-fpm[2170]: [NOTICE] [pool www] child 9085 started EDIT - 9 minutes was based on system uptime, clicking will of been a little less. Edited June 7, 20179 yr by Tuftuf
June 10, 20179 yr Upgrade to RC2 2 or 3 days, today got Nginx - bad gateway, the Web console lost but SSH no problem. What I am doing is changing some cache disk config. ( array was stop ). My system was simple, not much plugin, no VM / docker. I only have one time this error, I also got "[pool www] child" error, but system still running normal. Use Ryzen 1700 too, I notice you have 2 ethernet ( Intel + killer ? ), does you apply bonding / bridging / IPV6, would you try disable them ? Good luck. Edited June 10, 20179 yr by Benson
June 11, 20179 yr Author Based on my experience I don't think the Bad Gateway error actually causes any issues with the system itself. I've been changing my cpu assignments for VM's and even if I receive the Bad Gateway error message while submitting settings, they are still saved. Correct about the LAN ports, Initially I only had 1 connected and later connected the second. It works as expected. Likely an issue on my side but one of the ports did come up as 100Mbit the first time I connected it, simply replugging this to switch corrected it to 1000Mbit. Standing bonding settings. (no changes, my last setup also used vlans but that will be later)
Archived
This topic is now archived and is closed to further replies.