Project

General

Profile

Bug #2186

Amahi 9 Server Lockup

Added by techsoft over 3 years ago. Updated over 3 years ago.

Status:
Feedback
Priority:
Normal
Assignee:
Target version:
-
Start date:
11/15/2016
Due date:
% Done:

0%


Description

Hi,

I am a retired IT support engineer, but I have only a small amount of knowledge on Linux gleaned from installing a running my Amahi server for the last couple of years.

On the whole my experience with Amahi has been good and I have not had any unexplained problems. However over recent months my server has been locking up with no apparent warning or any error messages being displayed. These lockups happen at various times of the days. Sometimes when i am watching a video from the Plex Media Server streamed via my Apple TV, but mainly when the server could be idle or just performing general maintenance including automatic downloading of video files and syncing with Amazon Cloud Drivce (I have listed below the hardware and software setup and provided links to various logs). Once the lockup occurs all functions appear to stop. I have a keyboard and screen attached, but I am unable to access the console prompt and I am also unable to gain SSH access from my Macbook Pro or iPad. I can not access the Plex server, Sickbeard or SABNzbdPlus. DNS and DHCP stop working. The only action I can take is to power off the server and reboot. Once the server comes back up everything works fine without problem. If I had been watching a video, Plex takes me back in from the exact place the video had stopped. The server will then carry on without issue until the next lockup. That could be within a few hours or could take a few weeks. I have seen no pattern to the lockups, although it does appear to have gotten more frequent over the last few days.

I want to troubleshoot the problems further, but I am not sure where to start. Which logs should i be looking at and are there any special tools I should be using to scans the files for critical errors. Can any one suggest a methodogly for narrowing down the potential problems. With Windows and in the past Novell servers I know exactly where to look. I am now pretty oh fay with Mac's, but with Linux I just don't know where to start

My server is a HP Microserver N40L with 8Gb of memory. I have a Kingston 120Gb SSD setup as the boot drive with 4 x 3Tb drives in the main sata bays and an additional SansDigital TR4M Drive case containing another 4 x 3Tb drives that is connectted to a Syba SD-SA2PEX-2E PCIe 2x e-SATA Ports with SIL3132 Chipset. The server and drive caddie are plugged into a UPS.

On the software side I am running Amahi 9 on Fedora 23. Greyhole is installed managing the 8 x 3Tb disks. The landing zone is on the Kingston SSD. I have only installed Amahi packaged apps as follows: -
Greyhole UI
Disk Stats
Greyhole Monitoring
Plex Media Server
Sick beard
SABNzbdPlus
CouchPotato
IPSec VPN

The Plex Media Server, Sickbeard and SABNzbdPlus are the main programs i use heavily. Apart from daily automatic TV downloads using Sickbeard and SABNzbPlus the Plex Media Server also automatically syncs a number of the current TV programs up to my Amazon Cloud Drive.

Logs: -
Sysinfo - http://paste.fedoraproject.org/482865/14792750
Production - http://paste.fedoraproject.org/482868/47927566
Hda-updates - http://paste.fedoraproject.org/482869/75876147
Watch dog - http://paste.fedoraproject.org/482871/14792761
Samba - http://paste.fedoraproject.org/482872/76421147
Boot - http://paste.fedoraproject.org/482874/47927665

History

#1 Updated by cpg over 3 years ago

  • Status changed from New to Feedback
  • Assignee set to techsoft
  • Priority changed from High to Normal

This, based on our experience, is probably a hardware issue, typically a memory issue. the RAM may be flaky. This typically could lead to unexplained, random crashes that are not reproductible.

The other possibility is that power management was increased and the hardware is not well tuned to power management (e.g. the bios is not great, or the motherboard is not great).

It could be a lot of other things, typically all hardware based.

You can look at /var/log/messages or run jouralctl -x to see the system logs. However, if it's a hard hang, the logs may not even be written, or they may look like random kernel crashes.

Please close once you figure it out.

Also available in: Atom