Hi all,
Sorry i am somewhat new to this, even though i have been using it for years (it just worked to date!).
I have 2 x UNRAID installs each running on HP Microservers (old ones with Turion II N40L CPUs). I only use it to serve files to a Plex client running on a separate system on the LAN, so their lack of power hasn't been an issue, and i haven't changed the hardware ever.
Recently one of the servers has been causing me issues. Specifically, one minute you can connect to the GUI over the LAN on a device (phones or windows PC) and all works well, then the next it timesout/no connection. Wait a bit, and it comes back.
It also manifests itself in plex, a film will start working, and then it disconnects and say's media not available. I have also seen my library reduced in number of titles...
I run them headless, though plugging a monitor in i only see CLI, and can't get the GUI running (inpuits at boot dont work).
It was running 6.10.something.
I think one of the HDDs may be dodgy, but even unplugging that doesn't work. Things i've tried (that have made no difference):
- copied across the bz* etc files to update it to 6.12.6
- changed the network cable (and port on the switch)
- installed the drives and Flash drive into a different PC, and setup the server on that
I am looking for a flash drive to try and rule out a corrupt mem stick, but i am skeptical that this is the issue. Any suggestions/ideas - i thought the new hardware would solve it but alas not...
EDIT:
I have changed one thing on my network recently, and i am hesitant to go in to it because i don't think it can be the issue, and its complex (and i am working on it). But basically i have 2 x 5G modems connecting to a Draytek Vigor router, which is bonding the 2 connections to one, and load sharing between the two. I changed one of these modems recently, and it MAY coincide timewise with the issue. BUT they do not do any routing, they exist on a separate part of the network (main is 192.168.0... these are on 192.168.10 and 192.168.20 and so i see no reason why this could be linked to the issue...