tonifarres13 Posted September 23, 2015 Share Posted September 23, 2015 Hello, I have frequent crashes of the web interface of my Unraid box (version 6.1.2). In syslog I have a general protection error but I don't know what is causing this. It happens randomly when I try to update some page (Main page, VMs, Dashboard, etc). I've been unable to run web interface without crashes for more than one day. I ran memtest all night and it's working fine so I'm pretty sure it's not a memory issue. Sep 23 01:09:46 MEDIASERVER kernel: traps: emhttp[4474] general protection ip:403c8d sp:7ffef1392408 error:0 in emhttp[400000+1f000] Any ideas of waht is causing this? Link to comment
itimpi Posted September 23, 2015 Share Posted September 23, 2015 The most likely cause of General Protection Fault messages is dodgy RAM. You might want to run an extended MemTest (e.g. overnight) to see if it passes. Link to comment
tonifarres13 Posted September 23, 2015 Author Share Posted September 23, 2015 I ran MemTest overnight two days ago without a single error. Everything works fine (VMs, Dockers) but emhttp keeps crashing. Link to comment
itimpi Posted September 23, 2015 Share Posted September 23, 2015 How much RAM do you have? It may be that there is not enough for the workload being run and emhttp is being purged fro memory? It might be useful if you could provide the diagnostics file covering a period where the crash has happened. If you have no web GUI access then this can be created using the 'diagnostics' command from a console/telnet session. The resulting ZIP file is put into the logs folder on the flash drive. Link to comment
tonifarres13 Posted September 24, 2015 Author Share Posted September 24, 2015 Thank you for help. I have 32 Gb RAM, my board is a Supermicro X9SCM-II with a Xeon E3-1230 V2. I'm running 3 dockers (Crashplan, Plex Server and No-IP) and 2 VM: Ubuntu Server (4 Gb) and Windows 10 (4Gb). I have another VM with 2G (Ubuntu Desktop) running occasionally. I think the workload is not so high for 32 Gb RAM. It seems that the crash only happens when running Windows 10 with memory balloon. If I start the VM with fixed memory everything is ok. As soon as I configured the VM with 2 Gb of RAM with a max memory of 4Gb and started working with it forcing the VM to pull more RAM, crash happenned again: Sep 23 22:55:28 MEDIASERVER kernel: traps: emhttp[4437] general protection ip:403c8d sp:7ffd2c319488 error:0 in emhttp[400000+1f000] Maybe a problem with the Virtio drivers for Windows 10? I used the latest ones from Fedora page (0.1.109) I'll do some tests with the stable version (0.1.102) and memory configuration and see what happens. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.